MadModder

Gallery, Projects and General => Project Logs => Topic started by: sorveltaja on November 06, 2009, 07:14:49 PM

Title: Another version of how not to build a model engine
Post by: sorveltaja on November 06, 2009, 07:14:49 PM
This has been my main project for a while:
(http://www.kolumbus.fi/kz1706/022.jpg)

It's a small four-stroke engine, that is based on my own(borrowed some ideas from E.Westbury) drawings. Bore is ~10 mm and stroke ~12 mm. Compression ratio is something between 3:1 and 5:1 (wild guess).

So far I've encountered countless pitfalls on my design. Anyway, the engine shows signs of life, so rather than desperate, I'm curious about it.

In this topic, I'll do my best to reveal both good and bad things, that I've faced during the project, as most of those are common for everyone, that builds his/her first IC-engine.

Progress might be a bit slow, but more to follow...


 
Title: Re: Another version of how not to build a model engine
Post by: Bernd on November 06, 2009, 09:00:10 PM
Sounds good. Looking forward to your posts.

Bernd
Title: Re: Another version of how not to build a model engine
Post by: sbwhart on November 07, 2009, 02:35:37 AM
Looks a good project look  :thumbup: forward to you posts

Stew
Title: Re: Another version of how not to build a model engine
Post by: Gerhard Olivier on November 07, 2009, 03:13:57 AM
Im hooked already

Send more soon

Gerhard
Title: Re: Another version of how not to build a model engine
Post by: bogstandard on November 07, 2009, 05:25:07 AM
Sorveltaja,

I'm just trying to figure out which bit goes with what.

Any chance of a few more views, especially from the side.

Bogs
Title: Re: Another version of how not to build a model engine
Post by: NickG on November 07, 2009, 05:38:49 AM
Sounds great. Is a compression ratio of 3:1 high enough? Can't wait to see more.

Nick
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 07, 2009, 12:03:46 PM
Thanks for replies.

An overview of the engine's construction: Base is 5mm aluminum; crankcase and cylinder head are both machined from solid alum.
Cylinder liner's material is some sort of stainless steel. Easily machinable though. I used plenty of time, when lapping it. Glad I did.
---------------------------------------------------------
Pearlitic cast iron is used for the piston. It has no oil grooves or piston rings, as it provides nice tight compression without them. Even without any oil.
It's the material that I highly recommend for pistons/cylinder liners for low-power, slow running engines. No wonder, that Jan Ridders prefers to use it on his engines...
-----------------------------------------------------------
Carburetor on the left side is third or fourth version. Needle is from an ancient .10 glow engine. It still needs tweaking, maybe a new one(fifth version). Tried to make the needle system myself, but lack of lathe's top slide prevents turning any decent taper.
Currently, carb's throat size is 1.2 mm. Engine runs with that, but only at (too)fast speed. Positive thing is that no parts were flying out, when it cranked :clap: .
Testing different carbs with different throat sizes has gave me a hint of what the engine likes. Next version of the carb will have 0.6 mm throat, as it's the smallest drill size, that I have. If it's no go, then drill it to 0.7 mm, and so on. 
----------------------------------------------------------------
Sharp eyes might notice, that left side (inlet)valve doesn't have a tappet. Currently it's spring-loaded, as I ruined inlet cam by over-filing it.
-------------------------------------------------------------
To be continued..
(http://www.kolumbus.fi/kz1706/023.jpg)
(http://www.kolumbus.fi/kz1706/024.jpg)
(http://www.kolumbus.fi/kz1706/025.jpg)
Title: Re: Another version of how not to build a model engine
Post by: chuck foster on November 07, 2009, 12:42:31 PM
very interesting build  :thumbup:

you can spend allot of time trying to find the right combination of parts.............but when you do  :ddb: :ddb: :ddb:

please keep us posted and let us know how it all turns out  :clap: :clap:

chuck  :wave:
Title: Re: Another version of how not to build a model engine
Post by: NickG on November 07, 2009, 01:02:18 PM
Brilliant effort. Well done for getting it to run.  :bow: An IC engine is not far down my list of projects as it is something I have yet to do.

Nick
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 07, 2009, 01:26:30 PM
Thanks for kind comments.

The engine ran today, and I was going to shot a video of it. But the breaker points failed again(http://emoticons4u.com/crazy/1087.gif).
I'm going to replace them anyway with proper ones, that's used on cars. I've ordered few sets of those, and expect them to arrive next week. 

Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 10, 2009, 04:41:08 PM
In the meantime, I made another temporary breaker points to continue the testing.

Coil(an ex-transformer), that I used for ignition, stopped working, and made itself useless by sparking inside the coil:
(http://www.kolumbus.fi/kz1706/027.jpg)
After abusing(and destroying) several ones during the tests, I must admit, that transformer coils are not suitable for an IC ignition purpose. Not only because they can't produce reliable sparks, but they just can't handle repetitive strong current pulses.

Luckily I had an appropriate replacement part at hand:
(http://www.kolumbus.fi/kz1706/028.jpg)

What comes to breaker points, at first I tried several micro-switches:
(http://www.kolumbus.fi/kz1706/026.jpg)
They seem to fade quite fast.

Besides the temporary breaker points, that I made, today arrived the replacement for those also:
(http://www.kolumbus.fi/kz1706/029.jpg)
Next thing is to figure out, how to attach that to the engine, as the spring is too stiff for the humble power of small engine.
------------------------------------------------
Short video of the engine:

http://s323.photobucket.com/albums/nn471/sorveltaja/?action=view&current=01.flv

Carburetor needle is extremely sensitive, making the engine behave like a donkey.

Today it ran about half minute, at fast speed, then stopped due to heat expansions, I assume, as the cylinder head was too hot to touch.
Title: Re: Another version of how not to build a model engine
Post by: bogstandard on November 10, 2009, 05:20:53 PM
I found that a hall effect sensor and a rare earth magnet was much easier and more compact to use. I used the Tim 6 configuration.

Here is a link to other ignition types, the hall sensor takes the place of the contact breakers.

http://www.5bears.com/tim4.htm

Or go with Jan ridders version by using a piezo, which does away with everything, and fires the plug direct.

Here are a couple of pics of the last ic engine I made, after having a long running in session. It is still sitting in my shop in the same condition. I never wanted to, or ever will finish it off after getting it to run OK. It just knocked ic engines completely out of my system, after struggling with such terrible castings. It would have been ten times easier to have made it out of bar stock, as you have done, and how I made previous engines.

The second picture shows the mounted hall sensor and the magnet was mounted in edge of the small ali disc. Those two parts take the place of the points and condenser. No wear, and you can cover it in oil, and it still runs perfectly


Bogs


Title: Re: Another version of how not to build a model engine
Post by: NickG on November 11, 2009, 05:38:03 AM
sorveltaja,

The car points spring might feel stiff but remember they don't have to move much and it will only be a very small cam throw so it may not cause too much power loss when used in the correct manner - give it a go, I know the webster uses that method.

Jan used the piezo with some success but they sap quite a lot of power. He has also used the coil set up but with a microswitch as you show with success. http://heetgasmodelbouw.ridders.nu/Webpaginas/pagina_drukgestuurde_tweetakt/druktweetakt_frameset.htm

John,

I've never seen that one before, nice engine  :thumbup: what is the coil on it? Why is there an inline switch shown on the Tim 6?

Cheers,

Nick
Title: Re: Another version of how not to build a model engine
Post by: bogstandard on November 11, 2009, 06:54:01 AM
The engine is actually the Westbury Whippet, and the castings for it are available from Hemingway. But they are so bad they are almost unusable, especially the head area ones. I had to get a friend to build up areas of the casting with ali weld, just so that I could carry on with it. They were up to 1/8" out in places.

Those coils are now not available, and were made for fitting into model aircraft with 4 stroke petrol engines. They are only about 3/4" diameter.

The inline switch is in fact where you would mount the contact breaker or in my case, the hall sensor.


John
Title: Re: Another version of how not to build a model engine
Post by: NickG on November 11, 2009, 07:54:33 AM
John,

I thought I recognised it. A friend of mine is building the Chenerey V Twin Aero Engine with castings from Hemingway but he has also had all sorts of problems with the head castings especially. Valve bushes in the wrong place, wrong angle and blow holes everywhere. I think he tried 3 that I know of but then resorted to machining from solid. It looks a much better job and he said it was far easier.

Incidentally, if anybody is interested Hemingway say they supply a similar model electric coil to John's and they supply the TIM 6 in kit form, but they'll cost you an arm and a leg! http://www.hemingwaykits.com/acatalog/Ignition.html

Oh right, I see now, they just haven't shown it in the same way as the TIM4. I don't get TIM 7 ...  :scratch:

What sort of battery did you power yours from? I know Jan uses cordless drill batteries and the like.

Nick

Title: Re: Another version of how not to build a model engine
Post by: crankshafter on November 11, 2009, 11:24:59 AM
NickG

"Incidentally, if anybody is interested Hemingway say they supply a similar model electric coil to John's and they supply the TIM 6 in kit form, but they'll cost you an arm and a leg"

The coil they supply is the Exiter.

I recently lost one of my legs/hands  :bugeye::lol:
cs
Title: Re: Another version of how not to build a model engine
Post by: NickG on November 11, 2009, 11:27:49 AM
 :lol:

But as long as your model engine works it was worth it right?!  :thumbup:

Nick
Title: Re: Another version of how not to build a model engine
Post by: crankshafter on November 11, 2009, 11:53:56 AM
NickG
Time will show. The coil and TIM6 will be used on a Hoglet Twin that I have under construction. You can see some of the parts of it at my Avatar.
Hope to have it running before newyear.
cs
Title: Re: Another version of how not to build a model engine
Post by: bogstandard on November 11, 2009, 12:09:04 PM
Nick,

I think they are supposed to run off 3.6 volts, but I used a small 6volt 4ah sealed lead acid, and they give out around 7volts, and I had no trouble with them at all. The engine would run for hours using that battery.

I actually bought a Tim 6 circuit board kit from the US, half the price, and made up some of the cards myself. Made about a dozen of the units for friends for about 2 squid each. I gave all the rest of the bits to make them to a friend not too long ago. I think the coil was about 20 squid, again from the US.

If anyone is interested in making their own coils and magnetos, the late Bob Shores book (of which I have a copy), Ignition Coils and Magnetos in Miniature is a must. It tells you how to make them from scratch, at very small cost.


John
Title: Re: Another version of how not to build a model engine
Post by: crankshafter on November 11, 2009, 12:55:07 PM
John.
You are rigth. The Tim4 is for up to 4 volt and the Tim6 up to 6volt batt.
Actually Hemmingway kits get them from US of A. ( made by the well known Jerry Howell)
CS
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 11, 2009, 03:13:49 PM
Thanks for replies!

Today I made a new carburetor, and tested it. First was 0.7mm throat, then bored it to 1.0, and finally to 1.5mm. None of them brought the engine alive. Most likely reason for that is the fuel that I used - it was an unopened can of well known brand lighter fluid.

I have used that succesfully before, but this time even the odor was different. It smells more like teen spirit mouth-wash, than gasoline.
Did the burning test on a small steel cup with it, and the amount of smoke and smell told already, that it was something else/heavier, than pure gasoline.

To keep the engine's fuel thing straight, avoid lighter fluids, unless you are sure, what you get. 

Anyway, next I'm going to obtain real gasoline. And start the carburetor testing -phase from the start.

Learning curve might be quite steep, but I'm getting used to it. Having the engine, that really wants to run, is the source of inspiration :thumbup:.
Title: Re: Another version of how not to build a model engine
Post by: bogstandard on November 11, 2009, 05:25:40 PM
How are you starting it S?

If you look at the pic of the back of my engine (opposite to flywheel),  you will see a lump of bar with a hole in it.

That holds a one way clutch bearing. Stick a bit of 6mm rod in my drill, put it into the bearing and turn on. When the engine fires and runs, the rod just pulls out. It saved hours of finger aching starting during setting up trials.


Bogs
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 12, 2009, 04:13:05 PM
Bogs, at the earlier stage I used a Dremel with rubber wheel against flywheel for starting.

Recently it hasn't been necessary, since the engine starts easier. If it doesn't even kick, I know it isn't going to run. During this project, I have learned, what the general faults are, that prevent running. Mostly it's flooded spark plug, or failing breaker points.
Earlier the biggest problem was the cylinder head cover not being leak proof, due to my somewhat faulty design. But more of that later.

One by one, I've managed to stomp the bugs as they appear.

On the left is previous needle system, that snapped right where the needle tip comes out. In the middle is newest carburetor body with 1.5mm throat.
On the right side is a new needle, that I made off of compass needle. It however might need some tweaking. 
(http://www.kolumbus.fi/kz1706/030.jpg)

Title: Re: Another version of how not to build a model engine
Post by: bogstandard on November 12, 2009, 04:20:15 PM
S,

A lot of people use sewing or darning needles in their carbs.

Cheap and cheerful, but very accurate.


Bogs
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 13, 2009, 01:26:29 PM
I tested the new carb, and it floods the engine too easily. Most likely reason for that is the M3 thread, that holds the needle,
being too coarse(0.5 mm pitch), preventing any tuning.

The glow engine needle, that I used earlier, has 0.35mm pitch on its thread. It seems that even finer thread (~0.1-0.2mm) pitch would be better in this case.

I can make that fine thread on the lathe, at least outer thread. Small diameter inner threading is more of a problem. It would require threading tap.
Is there any other way to accomplish that?

Easiest way could be to buy a whole new needle system. But not yet.

In the meantime, I'm going to make a vapor carburetor, as it surely can't flood the engine :dremel:.




Title: Re: Another version of how not to build a model engine
Post by: Darren on November 13, 2009, 01:32:37 PM
How about making a tap on the lathe?  :dremel:

You can have whatever pitch you need then.
Title: Re: Another version of how not to build a model engine
Post by: sbwhart on November 13, 2009, 01:40:20 PM
You can get 60 TPI  (0.423mm pitch) from her http://www.tracytools.com/tapsanddiesfinethread.htm

Hope this helps

Stew
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 13, 2009, 01:45:41 PM
How about making a tap on the lathe?  :dremel:

You can have whatever pitch you need then.

It's indeed possible, but I'm completely newbie, what comes to required hardening process. On the other hand, if the tap is used to thread ptfe, it wouldn't need any hardening.... The needle body doesn't have to be made of metal?

Anyway, thanks for the tip, I'm going to test how I can manage :thumbup:.
Title: Re: Another version of how not to build a model engine
Post by: Darren on November 13, 2009, 02:00:40 PM
If you want to tap a softer material you could thread a Silver Steel/Drill rod to the sizes required.

Then grind a groove down one side, or maybe two sides. Touch it up with a stone to sharpen the cutting edge.

I'm guessing, cos I have never done it, then you will now have something between a regular cutting tap and a thread forming tap.

If you want to harden it, just heat to a bright red and quench in water. Then re-heat noting the colour change required and allow to self cool in air.

You say a newbie as if that should limit you?  Not so ...ask they guys here where they started ... just go for it  :thumbup:
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 13, 2009, 03:32:58 PM
Thanks, I will :dremel:.

I tested making a 0.2mm pitch thread on the lathe, and found out, that the cutting tools tip has to be very sharp, without any radius.

I wonder, how watchmakers sharpen their thread cutting tools. With the aid of microscope perhaps?

Back to sharpening I go ::).
Title: Re: Another version of how not to build a model engine
Post by: Darren on November 13, 2009, 04:33:15 PM
Forget the radius, far too fine a pitch to manage that I would have thought.

In essence you don't need to worry about exact angles and stuff because you will make both the male and female parts from the same cutting tool  :thumbup:
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 14, 2009, 07:24:43 AM
Another thing popped to my mind. There is noticeable difference between the glow engine needle system, and the one I made.

The former doesn't have its needle going across the carb throat. Thus allowing smaller throat size.
When using the latter with same throat size, needle itself restricts the air flow.

So I drilled the throat to 1.9 mm size, and the engine starts easier with that. Engine flooding is also reduced greatly.
Title: Re: Another version of how not to build a model engine
Post by: NickG on November 14, 2009, 09:52:25 AM
Was just thinking, have you considered making it glow ignition? It might be easier to trouble shoot carb problems if you have reliable ignition initially? Just a thought.

Nick
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 14, 2009, 10:24:35 AM
The ignition seems to work, as the engine did just run at a horribly fast speed. I thought it was going to explode :borg:.

To tame it, I'll add an air shutter to the throat.

Glow ignition would be simpler, but it wouldn't allow low rpm.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 14, 2009, 06:09:21 PM
Is a compression ratio of 3:1 high enough?

Nick

The approximation, that I presented about the compression ratio, is based to my own calculations, and has proven to be somewhat misleading(http://emoticons4u.com/crazy/1261.gif).
The math really isn't my strongest virtue :smart:.

The engine behaves like it had something like 8:1-10:1 comp ratio on it, regarding the speed it has run.

Next time I'm going to warm up the tachometer to get some readings.


Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 15, 2009, 08:57:23 AM
The shutter at the carb's throat doesn't seem to solve the adjusting problem :scratch:.

Anyway, I finally got a tachometer reading, about 4100 rpm, but the engine goes even faster than that, if allowed. Mostly I have to kill the spark to stop it.

Any advices on that carburetor thing would be greatly appreciated.
Title: Re: Another version of how not to build a model engine
Post by: John Hill on November 15, 2009, 01:33:49 PM
How about a restrictor on the exhaust?  That should tame the beast! :coffee:
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 15, 2009, 04:02:39 PM
How about a restrictor on the exhaust?  That should tame the beast! :coffee:

Thanks for the tip :thumbup:. I was just getting to testing it, when I heard 'crack' and the engine stopped suddenly, resulting this:
(http://www.kolumbus.fi/kz1706/031.jpg) (http://www.kolumbus.fi/kz1706/032.jpg)

Reason was the M2 nut, that keeps the connecting rod in place. It was loosened, and finally it fell off.
So far, I haven't found any other damages.

Lesson learned: The nut only isn't good way to secure the connecting rod end.

As an alternative, I'm going to attach a bearing disk to crankcase backplate.

What comes to crankshaft, only the shaft is bent, so I'll drill it off, and replace it with a hardened 3 mm rod.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 16, 2009, 07:01:31 AM
Here is the crank disk and new rod, that are to be silver soldered together:
(http://www.kolumbus.fi/kz1706/033.jpg)

Bits and pieces:
(http://www.kolumbus.fi/kz1706/034.jpg)

Another view of cylinder head and liner:                                    From the top:
(http://www.kolumbus.fi/kz1706/035.jpg)                          (http://www.kolumbus.fi/kz1706/036.jpg)

Crankcase is fastened to base with two M3 screws:
(http://www.kolumbus.fi/kz1706/037.jpg)

Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 16, 2009, 11:22:20 AM
Repaired crankshaft with ball bearings:                                 
(http://www.kolumbus.fi/kz1706/038.jpg)(http://www.kolumbus.fi/kz1706/039.jpg)
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 16, 2009, 03:52:31 PM
About that bearing disk, that I mentioned earlier; it really isn't 'bearing' disk, as it's a static part.
Since I don't have suitable size steel stock, I'll make that disk by stacking an M8 washers, and silver solder them together:

before...                                                                       ...and after.
(http://www.kolumbus.fi/kz1706/040.jpg)                (http://www.kolumbus.fi/kz1706/041.jpg)

The disk in place:                                                           
(http://www.kolumbus.fi/kz1706/042.jpg)               

It needs to be shortened, so that it can be attached to crankcase backplate.
(http://www.kolumbus.fi/kz1706/043.jpg)

If my memory serves, this kind of system has been used on glow engines.
Title: Re: Another version of how not to build a model engine
Post by: NickG on November 17, 2009, 04:02:01 AM
sorveltaja,

I think the nut would have been a fine way of stopping it coming off. It shouldn't have come undone, the only reason I can think it did was if the conrod was slightly thicker than the length of the crank pin, it should have been the other way around.

Still, your new method should work as long as there is no binding between rod and that disc.

Nick
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 17, 2009, 04:23:50 PM
Nick, the recent nut-system should have worked, as my purpose was to make a slow runner. But the engine runs at too fast speed, causing more vibrations, than I ever expected.

But it's just one of those bugs, that needs to be stomped(http://emoticons4u.com/crazy/061.gif).

Here is the bearing plate attached to backplate. Fastening is done by two M2 screws:
(http://www.kolumbus.fi/kz1706/044.jpg)

And here are the plate + crankcase backplate in place:
(http://www.kolumbus.fi/kz1706/045.jpg)

So it's about time to re-assemble rest of the engine :ddb:.

Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 18, 2009, 01:06:18 PM
Before assembly, slots for primary gear and flywheel's grub screws needs to be done:
(http://www.kolumbus.fi/kz1706/046.jpg)

Testing that they fit ok:
(http://www.kolumbus.fi/kz1706/047.jpg)
Title: Re: Another version of how not to build a model engine
Post by: bogstandard on November 18, 2009, 01:19:36 PM
In all honesty S, your over revving just might be because you have a too large carburettor throat. I would try about half the diameter you have now and see if it makes a difference, you can always open up in stages of 0.1mm until you reach a good compromise. On small engines like yours, 3mm or 4mm would be classed as a racing bore, designed for getting the engine into a higher rev range by over fuelling it.

The reason I suggested sewing needles is because they have a very shallow taper, so allowing you to use a coarser thread for adjustment, plus they will be much smoother than a needle you make yourself.


Bogs
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 18, 2009, 02:02:45 PM
Bogs, you are right, but as I have tested throat sizes from 0.6 mm to 1.9mm, there doesn't seem to be any sweet spot.

Too small throat size causes flooding, no matter what; with ~1.2-1.9mm throat, the engine starts easier, but runs too fast.

Maybe the compression ratio is too high; at the beginning of this project, I was going to make a cylinder head, that allows the comp ratio to be adjusted.

Back then, it did seem too complex thing to achieve. Not impossible anyway.

Lowering the comp ratio is just one of the options to test.

But in the mean time, I'll assemble the engine.:
(http://www.kolumbus.fi/kz1706/048.jpg)   


Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 18, 2009, 02:25:03 PM
Cylinder head/liner combo in place:
(http://www.kolumbus.fi/kz1706/049.jpg)
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 18, 2009, 03:11:28 PM
And finally, the assembled engine:
(http://www.kolumbus.fi/kz1706/050.jpg)


Title: Re: Another version of how not to build a model engine
Post by: Darren on November 18, 2009, 05:22:37 PM
Sure is an interesting looking engine ... nice to see you got it fixed up after the little disaster :thumbup:

Now don't you be going thrashing it again ....  :wack:
Title: Re: Another version of how not to build a model engine
Post by: Russel on November 18, 2009, 10:35:28 PM
Wow, until you added the matchbox in the photo I thought your engine was about 6" tall. I'm impressed, a very nice little engine!

Russ
Title: Re: Another version of how not to build a model engine
Post by: Stilldrillin on November 19, 2009, 03:17:50 AM
It`s much smaller than I thought too......  :bugeye:

David D
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 19, 2009, 01:35:18 PM
What comes to that carb needle -dilemma, it's time to make a needle body, that has grub screw to hold the needle, as it allows testing different needles.

The idea came from this:
(http://www.kolumbus.fi/kz1706/052.jpg)
Above, the grub screw doesn't actually hold the needle, but instead possible extension piece.

And here is more than enough meat for practicing:
(http://www.kolumbus.fi/kz1706/051.jpg)

The safety pins are of 1mm wire, so it's possible to use a mini-drill for making different tapers.
Title: Re: Another version of how not to build a model engine
Post by: bogstandard on November 19, 2009, 02:26:18 PM
S,
Quote
so it's possible to use a mini-drill for making different taper

Are you using a tapered bore in your inlet?

If so, you will not get very good control as you are using two tapers, when they start to open up, they grow apart too quickly. You must only use one taper (the needle) to get good results.

Please see C-o-C.

Bogs
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 19, 2009, 02:49:33 PM
Bogs, thanks for demonstration. Anyway, I intend to use the taper on needle only, to keep things simple.

I'm not even quite sure, how to make a decent tapered bore, so I have kept myself away from it.
Title: Re: Another version of how not to build a model engine
Post by: bogstandard on November 19, 2009, 03:11:32 PM
S,

I found that sewing needles had a much longer taper, so aiding much finer adjustment, and they were stuck into the threaded holder with Loctite. If you need to go up a size, just warm the holder up and the loctite will give way, allowing another needle to be fitted.


Bogs
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 19, 2009, 03:56:43 PM
Bogs, the Loctite thing is a fast thing for sure, but it would require cleaning, when changing the needle.

So I'll stick with that grub screw -schema stubbornly(http://emoticons4u.com/crazy/1087.gif) .

Ready for needle shaping:
(http://www.kolumbus.fi/kz1706/053.jpg)
(http://www.kolumbus.fi/kz1706/054.jpg)

And the finished needle:
(http://www.kolumbus.fi/kz1706/055.jpg)



Title: Re: Another version of how not to build a model engine
Post by: NickG on November 19, 2009, 05:50:21 PM
S,

Don't know whether it's the camera playing tricks but that needle looks parallel except for a very short tapered section on the end. If you used a sewing needle as bogs said you'd be sure to get a finer adjustment. Worth a try.

Nick

ps on compression ratios, from looking at Jan ridders designs he seems to aim for about 6:1. I know 6:1 is low for a 4 stroke auto engine these days so it might pay to go through some recognised plans and see what they come out at?

Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 20, 2009, 12:47:42 PM
Nick, it wasn't the camera, needle in the pic had a short taper.

Here is the needle with longer taper, after some polishing, and the needle body with grub screw:
(http://www.kolumbus.fi/kz1706/056.jpg)
(http://www.kolumbus.fi/kz1706/057.jpg)
(http://www.kolumbus.fi/kz1706/058.jpg)
Title: Re: Another version of how not to build a model engine
Post by: NickG on November 20, 2009, 06:40:55 PM
That looks much better, fingers crossed.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 21, 2009, 06:08:05 PM
So I got the engine assembled, but as always, at least in my case, it was(is) quite challenging(despairing) to get the spark timing even close to position, where the engine starts to show signs of life.

I'm quite sure, that I need to make a spark tester, that shows exactly when the spark occurs.

There is a fine example on Jan Ridder's page: http://heetgasmodelbouw.ridders.nu/Webpaginas/pagina_vonktester/vonkentester_frameset.htm



 
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 23, 2009, 02:12:48 PM
My previous post's problem was caused indeed by the temporary breaker points, that I made :wack:.

As I looked back the replies, there was good suggestions, such as using TIM4, or piezo igniter.

5bears's TIM4 has the Hall IC, HAL506UA, that has became obsolete. Nowadays it seems to be a 'hard to find' -part. Couldn't find the replacement either :scratch:.

The piezo ignition is ruled out, because it requires far too much force to make a spark. With bigger engines, that have the grunt, it wouldn't be any problem though.

What's left, is to fit an automotive points to the engine:
(http://www.kolumbus.fi/kz1706/059.jpg)
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 24, 2009, 10:59:59 AM
Dark piece in the middle is roughly cut piece of 1.5mm mild steel.  
(http://www.kolumbus.fi/kz1706/060.jpg)(http://www.kolumbus.fi/kz1706/061.jpg)

Here with holes to attach the contact breaker. Aluminum part is for tightening the assembly to crankcase backplate:
(http://www.kolumbus.fi/kz1706/062.jpg)

...and yes it fits!! ::):
(http://www.kolumbus.fi/kz1706/063.jpg)

The steel plate has some excess material in it, so after before cropping, it looks like this:
(http://www.kolumbus.fi/kz1706/064.jpg)
Title: Re: Another version of how not to build a model engine
Post by: NickG on November 25, 2009, 05:27:59 AM
S,

Looking good, what sort of cam are you going to use? I know the webster engine uses car points and he just has a cylinder of metal with a flat milled on it - seems to work well.

Nick
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 25, 2009, 01:01:37 PM
Nick, I was going to make a roller for ignition cam, but it would have been too complex. Simpler the better :beer:.

Brass was used for the cam, and polished it a bit:
(http://www.kolumbus.fi/kz1706/066.jpg)(http://www.kolumbus.fi/kz1706/065.jpg)
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 25, 2009, 03:53:35 PM
Addition: After previous testing, that was made using ethanol, the engine had couple days of rest.

When it was time to fit the breaker points to it, I noticed that the crankshaft was really stiff to rotate.

So I sprayed some wd-40 from spark plug hole, and gave it spinning. Stiffness was gone. I should have done that right after testing.

That problem never occurred after using pure gasoline...

Correct me if I'm wrong, but ethanol, if I dare to say, is water based liquid :scratch:. If even small amount of it is left inside the engine, rust and corrosion is to be expected :wack:.
Title: Re: Another version of how not to build a model engine
Post by: Russel on November 25, 2009, 11:02:18 PM
Even if it is pure ethanol, it will tend to draw moisture out of the air if there is any humidity. That can be a big problem with chainsaws and gasoline that has a small amount of alcohol in it.

Russ
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 27, 2009, 02:49:37 PM
Russel, thanks for the information. That's what I was after. Not water-based, but that ethanol absorbs water from the environment  :beer:.

After installing new breaker points/cam -combo, they seem to work just fine. At first I had a doubt, if the breaker's spring was too stiff, but it doesn't affect the engine's behavior.

When testing, once the engine fired, it ran at fast speed for a short time. I noticed by the ear, that something was preventing it to run longer.
At earlier stage, I had to kill the spark to stop the revving.

But anyway, I checked the ignition system(numerous times), and it was ok. Although the compression felt quite right, there was something wrong.

I have avoided opening the cylinder head, because I've finally managed to get it airtight; But no other way to check the valves.

Most excellent chance to reveal the interior of the head also.

I couldn't get the gasket off willingly, so I'll let it be:
(http://www.kolumbus.fi/kz1706/067.jpg)
(http://www.kolumbus.fi/kz1706/068.jpg)

Above pics show the major flaw in my design; middle holes are far too close to the ignition chamber; or the cavity is too wide.
Making a gasket for that is an adventurous thing(http://emoticons4u.com/crazy/134.gif).

Cylinder head:
(http://www.kolumbus.fi/kz1706/069.jpg)

As this picture shows, there is a brass button to increase the compression ratio. It has something to do with earlier cylinder head sealing problems.
(http://www.kolumbus.fi/kz1706/070.jpg)

Valves as they are; far from being similar to each other:
(http://www.kolumbus.fi/kz1706/071.jpg)

Cylinder liner, made of stainless steel, having two brass parts silver soldered to it:
(http://www.kolumbus.fi/kz1706/072.jpg)





    
Title: Re: Another version of how not to build a model engine
Post by: bogstandard on November 27, 2009, 05:26:15 PM
S,

Very interesting design for a side valve.

If you don't mind me giving you a suggestion which you might benefit from.

Even a basic bit of gas flow design might help with the problems you are experiencing.

It seems that the fuel might be being held around the valve area instead of the charge being sucked down into the cylinder.

If you look at the C-o-C that I have attached, I think just cutting the corner off the brass slug, like in the two top pics, would improve the gas flow a lot to give a more efficient burn.

If you ever needed to make a new head, the bottom picture shows how it could be improved even further, by shaping the head a little more, and moving the plug a little more into the corner of the cylinder.


John
Title: Re: Another version of how not to build a model engine
Post by: 75Plus on November 27, 2009, 06:12:53 PM
S, The head gasket can be made of copper foil which should contain the heat and pressure. I don't know about your location but here in the US copper foil can be found in stores that carry craft supplies.

Joe
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 28, 2009, 06:55:15 AM
Bogs, thanks again for the demonstration :thumbup:. The reason for the spark plug placement is, that I originally was going to add 'dieselish' compression screw(and contra piston) to cylinder head cover.

I'm not sure yet, what to do with that brass button. Maybe I'll bevel it.

In the meantime, to replace the M2 screws, I drilled cylinder head's 2mm holes to 2.5mm, and made an M3 threads to them:
(http://www.kolumbus.fi/kz1706/073.jpg)

My M3 threading tap has snapped, thus being so short. About 2mm too short for the head. I took an M3 screw, and ground there a bevel, so with that I got the threads finished:
(http://www.kolumbus.fi/kz1706/074.jpg)

Joe, I actually have 0.1mm(~0.04") copper sheet. Don't know if that is thin enough, but I'll give it a try:
(http://www.kolumbus.fi/kz1706/075.jpg)
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 29, 2009, 06:26:09 AM
I made a bevel to brass button, and copper gasket also, and re-lapped the valve seats. With M3 screws it's possible to apply more fastening force to the head cover, than with earlier M2 screws:
(http://www.kolumbus.fi/kz1706/076.jpg)

So it's about time to assemble rest of the engine.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 29, 2009, 10:38:21 AM
At this stage, head cover is fastened, and spark plug in place. I flooded the engine a bit with wd-40, and gave it a spin, to find possible leaks.

The head cover seems to be ok, and valves seem to seal ok.
(http://www.kolumbus.fi/kz1706/077.jpg)

But there is a leak, where the cylinder and head meet:
(http://www.kolumbus.fi/kz1706/078.jpg)

I was tempted to put some loctite there, but instead I used 'high temperature instant gasket' paste, that vulcanizes at room temperature.
Head screws were opened just enough to get the stuff between the parts, and then tightened:
(http://www.kolumbus.fi/kz1706/079.jpg)

When the paste is cured, it's like a soft rubber, making it easy to remove the excess material with a sharp knife.

But anyways, if the gasket paste fails under pressure, I'll get back to that loctite-schema :ddb:.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on November 29, 2009, 02:25:37 PM
During the acid running tests, there hasn't been leak problems so far.

Current setup:
(http://www.kolumbus.fi/kz1706/080.jpg)

To save my fingers, I used trusty Dremel with rubber wheel to spin the engine:
(http://www.kolumbus.fi/kz1706/081.jpg)

After some fiddling, the engine fired. Short high speed runs, no response for the carb's needle, just as it did before. Fuel was ethanol.

Next thing was to test with gasoline. At first, the engine had same running habit, that it had with ethanol. I kept on trying, but letting the engine to cool down between the tests.

At first time in its history, the engine started to show even faint response to carb needle :bugeye:.

Carburetor has 1.9mm throat, but with 1.5mm ptfe plug, it worked better.

So there is a new carby to be made...



Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 01, 2009, 06:35:37 AM
I made a new carburetor, that differs from previous ones, so that it doesn't have the needle going across the throat. Parts are ready to be soft soldered together:
(http://www.kolumbus.fi/kz1706/082.jpg)(http://www.kolumbus.fi/kz1706/083.jpg)
Title: Re: Another version of how not to build a model engine
Post by: NickG on December 01, 2009, 07:31:56 AM
S,

Nice work on that carb  :scratch: I don't really understand it but it looks good! Just wondered - have you considered Jan Ridders petrol vapour carb? That could solve your mixture issues drawing in a pre-mixed vapour, just a thought. Even if you didn't want to pursue that route, or felt it was cheating or something at least it would give an indication as to where the problem lies.

Nick
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 01, 2009, 09:16:19 AM
Nick, the vapour carburetor is most certainly next thing to test, if the current carb doesn't give good results.

I soldered the parts together, and noticed that the carb throat was filled with solder also(http://emoticons4u.com/crazy/1261.gif).

Took it apart, drilled the holes open, and removed rest of the solder. After cleaning the parts, plan B was to try super glue. It seems to work better in this case :thumbup:.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 01, 2009, 03:01:40 PM
I had to add an extension piece for carb, so that it's possible to put in place and remove it without removing breaker/cam combo:
(http://www.kolumbus.fi/kz1706/084.jpg)

Now the carb throat has 0.7mm(~0.027") diameter, and the engine fires with that, but doesn't run.

Next thing is to drill the throat to 0.8mm(~0.031"). Cleaning the carb after drilling might be more difficult, without having to disassembe it, as it's little more complex, than the previous ones.
Title: Re: Another version of how not to build a model engine
Post by: sbwhart on December 01, 2009, 03:03:41 PM
Thats developing into a great little engine:-  :clap: :clap: :clap:

Stew
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 01, 2009, 04:06:52 PM
Thanks for kind comments :wave:.

Here is a cross section sketch of that carburetor:
(http://www.kolumbus.fi/kz1706/085.jpg)

Basically the principle is same, that has been used at least on small commercial glow engine's carbs. That concept makes it possible to make a carb, that's needle doesn't intrude across the throat.
Title: Re: Another version of how not to build a model engine
Post by: Bernd on December 01, 2009, 09:34:28 PM
S,

I don't see any "venturi", just a straight hole to the engine in your drawing. I recall that the glow plug engines all had some kind of venturi on the carburator before the fuel got mixed in with the air.  :scratch:

Bernd
Title: Re: Another version of how not to build a model engine
Post by: NickG on December 02, 2009, 04:47:32 AM
S,

THat extension to the inlet manifold might affect things. You have now brought another variable into the system and you don't know the effects.

Good work though, hope you find a solution.

We want to see a video too!!

nick

Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 02, 2009, 01:50:57 PM
Bernd, you're right, there isn't venturi. I have wondered, how does one affect the engine's behaviour :scratch:. Many of the old engines have just straight tube as venturi.

Here is an example of an engine, that is roughly about the same size as the current project, although it's two-stroke: http://www.modelenginenews.org/schroeder09/index.html

I'm not sure, if 2 and 4 stroker carbs are comparable, but on that Schroeder's 09, the carb throat is quite big (4.37mm(0.172")).

I'm bit confused about those :smart:.

Nick, the extension piece doesn't seem to affect, as the engine behaves like it did before it.

Anyway, I'm also looking forward to shoot a video of it running. If lucky, at this week.

 



Title: Re: Another version of how not to build a model engine
Post by: John Hill on December 02, 2009, 04:12:06 PM
At the risk of showing myself to be a complete Wally I will venture my thoughts on the venturi...

A venturi causes a change in pressure in the airflow and I assume a carb with a venture will more readily atomise or evan vaporise the fuel.  So I assume that where a venturi is not used the fuel must be more volatile.
Title: Re: Another version of how not to build a model engine
Post by: Bernd on December 02, 2009, 04:18:11 PM
Ok, I looked at the Schroeder 09. I can see that it dosen't have a vebturi.

Well it was guess as to what might be wrong. Looks like you don't need one to run the engine.

Keep up the good work. I'm sure you'll get it straightened out eventually.

Bernd
Title: Re: Another version of how not to build a model engine
Post by: bogstandard on December 02, 2009, 05:26:34 PM
You will find that you will get very little fuel suction as you have it.

The way the one you showed works is the spraybar acts as a backwards venturi. Instead of having the curved in walls of a standard venturi, speeding up the air flow and thus the drop in pressure, the airflow around the spraybar speeds up the airflow and it is that which causes the pressure drop, and so the fuel is sucked into the inlet. It is not as efficient as a venturi, but it is good enough for these small engines.

C-o-C explains the principle.


Bogs
Title: Re: Another version of how not to build a model engine
Post by: Darren on December 02, 2009, 05:34:35 PM
I think the problem was too much fuel and not enough control ?
Title: Re: Another version of how not to build a model engine
Post by: AdeV on December 02, 2009, 05:39:20 PM
A venturi causes a drop in air pressure (swapping pressure for velocity). This serves two purposes: It draws fuel into the airstream by the difference in atmospheric pressure; and, as a happy coincidence, the faster-moving air helps to atomise the fuel.

I think that the "venturi" effect in the Schroeder engine is created by the tapered cut (creates a large surface area at the end of the tube), the air would be accelerated as soon as it's in the main body of the tube. The effect will be quite small & as a result inefficient, but I suppose on an engine that size, you can trade efficiency for simplicity.


With regards to this engine, I suspect if you cut a slight taper in both sides of the main carb body and open out the fuel/air mix passage into the engine, as per the diagram below, it'll work better. Go easy on the size increases & test often, that way you shouldn't overshoot the sweet spot.

(http://lister-engine.com/pics/carb.jpg)

There's a pretty good explanation here: http://www.explainthatstuff.com/how-carburetors-work.html

This page may also help you out a bit: http://www.hooked-on-rc-airplanes.com/nitro-engine-tuning-tips.html

I see Bogs has come up with a better explanation of the Schroeder carb than the one I'd stumbled around at.
Title: Re: Another version of how not to build a model engine
Post by: bogstandard on December 02, 2009, 06:11:43 PM
You have to be very careful with angles in a venturi, if they are the wrong combination, you can actually get the air stalling and causing all sorts of problems, with dead air in places you don't want

I have just made a venturi for my spraymist system using a couple of basic angles that I know work. That is 22 degrees for going in and 6 degrees coming out, both inclusive, but I did away with the 6 degree one because of the way I will be using it. That should provide a steady pressure drop and hence suction once the engine is turning over.


Bogs
Title: Re: Another version of how not to build a model engine
Post by: AdeV on December 03, 2009, 12:24:38 PM
You have to be very careful with angles in a venturi, if they are the wrong combination, you can actually get the air stalling and causing all sorts of problems, with dead air in places you don't want

I should point out that IANAFD (I Am Not A Fluid Dynamicist), so yes - if the OP can find the correct angles for a basic carb, that would be better. Or use the spraybar, which is probably easier to machine anyway.
Title: Re: Another version of how not to build a model engine
Post by: bogstandard on December 03, 2009, 01:16:25 PM
I was picking up on what you had said Ade, not criticising, but trying to be of assistance

I have stated before, I am no super expert on it, but I do tend to remember bits of two years of aerodynamics that I was force fed many many years ago.

As you suggested, in this instance, a spraybar would be the easiest way to go.


Bogs
Title: Re: Another version of how not to build a model engine
Post by: AdeV on December 03, 2009, 03:48:04 PM
I was picking up on what you had said Ade, not criticising, but trying to be of assistance

Don't worry, I took it to be an assist rather than a crit. And, after I read your post, it occurred to me that every model carb I've ever seen has quite a steep taper on the inlet, not a nice shallow one like wot I drew. IstillNAFD, though (http://helmies.org.uk/Agree.gif)


Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 03, 2009, 04:27:57 PM
I must admit, that I enjoy reading the discussion about that carb thing :thumbup:.

There sure isn't plain right or wrong aspects, but rather options, of how things can be achieved.

As much as I'd like to test variations of tapered venturi's, lack of required tooling keeps me on that basic straight tube -venturi schema.

In the mean time, testing goes slowly, but evenly. With 0.8mm carb throat, the engine fires, and almost runs.

In between, I tested also with the previous carb(1.9mm throat), and it has turned out to be a kind of reference. With that, the engine runs, making it possible to check once in a while, that valves are seating ok, spark timing is ok, etc.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 04, 2009, 01:28:24 PM
After some more testing with 1.9mm and 0.8mm carbs, the engine wants to run only at high speed with both.

Part of the problem might well be the spring loaded intake valve, having tension, that allows only limited rpm range:
(http://www.kolumbus.fi/kz1706/086.jpg)

It's kind of difficult to adjust its spring tension :scratch:.

I think I'll start a new camshaft. With it, the intake timing should be more constant, than that spring loaded, regardless of rpm :dremel:.


Title: Re: Another version of how not to build a model engine
Post by: NickG on December 04, 2009, 01:35:11 PM
That is a good point S. But many slow running engines have an atmospheric intake valve. The spring must be really very weak in such a system so that the suction of the piston will open the valve. From what you are saying it's too strong and the engine can only open it at high rpm?

Nick
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 04, 2009, 01:55:57 PM
Nick, it's something I wonder also. When turning the engine slowly by hand, the inlet valve makes snorty sound, indicating that it lets the air in, and also, that the spring tension isn't too strong.

I may be wrong though. But for example, on hit'n'miss engines, I assume, the spring tension isn't as critical, because there is a governor, that effectively prevents the engine from running too fast. 

Hmm... should I mod the current engine to hit'n'miss(http://emoticons4u.com/crazy/134.gif). Tempting option, but we'll see.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 05, 2009, 05:34:31 AM
Before making whole new camshaft, I'll try to revive the old one:            Rest of the over-filed inlet cam machined off, and new cam blank to be                                                                                                       silver soldered:
(http://www.kolumbus.fi/kz1706/087.jpg)         (http://www.kolumbus.fi/kz1706/088.jpg)

Before..                                                          ...and after cleaning. Cam is now ready for forming:
(http://www.kolumbus.fi/kz1706/089.jpg)(http://www.kolumbus.fi/kz1706/090.jpg)
Title: Re: Another version of how not to build a model engine
Post by: NickG on December 05, 2009, 06:46:41 AM
S,

this might work better by as you say relating the timing directly to engine speed - as long as you get the timing right. Great stuff, fingers crossed it works. 

Nick
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 05, 2009, 02:43:01 PM
Forming the inlet cam entirely by filing isn't too attracting option, at least for me, so i made a quick 'n' dirty jig, so most of the excess material can be machined off in lathe:
(http://www.kolumbus.fi/kz1706/091.jpg)

Grub screws keep the camshaft in place, allowing the shaft to be rotated between machining:
(http://www.kolumbus.fi/kz1706/092.jpg)

This simplified jig doesn't necessarily need the tailstock support, as only light cuts are taken.

There are no dials to measure the angles, instead I'm going to perform the machining process by following the basic instinct hunch.

If it fails, with that jig, it's easier to make new cams anyway :ddb:.
Title: Re: Another version of how not to build a model engine
Post by: John Hill on December 05, 2009, 09:37:49 PM
If you think there is a problem with the inlet valve maybe it is bouncing instead of closing cleanly?  If it is bouncing I think you will see a small cloud of fuel being blown out of the carb? :scratch:

I greatly admire your progress on this engine! :bow:
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 06, 2009, 02:29:55 AM
John, can't say for sure, would bouncing be the problem, as I haven't noticed fuel blown out of the carb, except on cases, if there is impurities in the fuel, that gets to inlet valve, causing it to fail.

Usually compressed air is enough to clean the valves.

Before next running tests, I'm going to make a fuel filter.
Title: Re: Another version of how not to build a model engine
Post by: NickG on December 06, 2009, 03:54:14 AM
S, now it is controlled by cam you can put a stronger spring to avoid any bouncing. I don't think it would though, as soon as it gets to its compression stroke it would shut and stay shut until there is an underpressure in cylinder.

Nick
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 06, 2009, 09:47:13 AM
Repaired camshaft:
(http://www.kolumbus.fi/kz1706/093.jpg)

Stronger spring and tappet added to inlet valve. Engine is now ready for testing:
(http://www.kolumbus.fi/kz1706/094.jpg)
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 06, 2009, 12:29:44 PM
Testings with new inlet cam didn't make the engine run. At first, the cam was rather egg-shaped, having too long opening/closing time, so I filed the cam a bit, to make the valve opening and closing more rapid:
(http://www.kolumbus.fi/kz1706/095.jpg)

After that, engine fires, and also runs at fast speed again, although I have tested it only with 1.9mm carb this far.

Addition: I tested the engine with 0.8mm carb also, and got similar results, as with spring-loaded intake cam. Engine fires, and almost runs.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 09, 2009, 02:38:17 PM
Current status of the engine. Hopefully it's noticeable on the video, how the engine responds to carb needle:


Previous sparking plug started to push out its ptfe insulation, causing severe running problems, so I had to make new, bit modded version, that should keep the insulator in place much better.

Test setup on the video:
Fuel: regular 95 gasoline with small amount of motor oil
Carburetor is the older one 1.9mm throat with 1.5mm hole ptfe plug in its butt its throat.
Ignition timing was about tdc(top dead center). I have tested both advancing, and retarding the timing though.

Generally, the engine has began to show more responsibility to the carb, and also it starts easier.

Maybe it just needs plenty of short running-in sessions.
Title: Re: Another version of how not to build a model engine
Post by: bogstandard on December 09, 2009, 03:14:29 PM
S,

It has only just clicked with me :doh:, you are not using a throttle on the carb, so all you will ever get is high speed running, which is correct. When you adjust the needle, all you are doing is making the mixture more or less efficient, either weak, correct or rich. Just like a normal model free flight or control line engine.

What you really need is a throttling carburettor to give you speed control, which is a little more difficult than what you have now, but achievable.


John


Title: Re: Another version of how not to build a model engine
Post by: NickG on December 09, 2009, 06:02:31 PM
Seeing that video it's obvious now! I have been studying Jan's engines for a couple of years now and I am so used to that petrol vapour carb that I didn't notice lack of throttle body either. Jan effectively changes the mixture to alter the speed but he is also choking the air at the same time so it's a bit like a throttle? (please correct me if I'm wrong). S , as John said you are purely changing the mixture, the throttle is full bore all the time.

Looks to be running really nicely by the way.

Nick
Title: Re: Another version of how not to build a model engine
Post by: chuck foster on December 09, 2009, 08:13:54 PM
sorveltaja : that's one sweat running little engine but as others have said you need a carb that has a throttle.
i might have some plan's to build a little carb that might just do the trick for you.
i will dig them out and email them to you if you like.

chuck  :wave:
Title: Re: Another version of how not to build a model engine
Post by: NickG on December 10, 2009, 04:40:25 AM
Not sure what model ones are like but I know from working with car SU carburettors they effectively have a cam on the throttle butterfly rod that pushes the needle down as you need to richen the mixture as you open the throttle. Not sure how fixed jet carbs work, number of jets that become uncovered by a rotating disc?

Nick
Title: Re: Another version of how not to build a model engine
Post by: Darren on December 10, 2009, 09:04:01 AM
You could try a washer, or plate with a small hole to restrict the airflow. Expect to have to adjust the needle as well though.

Get it going and place the washer over the intake to see if it slows down. Different sized holes should give you different speeds. Worth having a go before making something.
Title: Re: Another version of how not to build a model engine
Post by: NickG on December 10, 2009, 10:46:40 AM
Good point Darren.  :thumbup:
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 10, 2009, 02:54:14 PM
Thanks for replies and suggestions :wave:.

Today I was testing the engine with different carb settings, and suddenly it just refused to start or run. So I double-checked every possible things.

Fuel system was ok, ignition system was ok, but the new spark plug started causing shortcuts, no matter what. Yesterday it worked just fine :scratch:.

As gasoline itself doesn't conduct electricity much, in this case it was enough to cause almost electrochemical pair between spark plug and its core. Resulting black goo between them. With previous plug, that never happened.

It must have something to do with new spark plugs body material. It is made of an ordinary M6 screw, as was the previous one. Apparently they aren't same material.

Old and new plug:
(http://www.kolumbus.fi/kz1706/096.jpg)
Two cores, left was made of an M2 screw, right is stainless steel. Both was tested with new plug, but no difference.

The battery-led combination is a poor man's continuity tester. The 9V battery in picture is old, and it has low enough voltage, so it doesn't burn the led.   

Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 10, 2009, 03:34:31 PM
Old plug is on the left, new on the right:
(http://www.kolumbus.fi/kz1706/097.jpg)

Right one might have too little sparking surface, when compared to the left one. That also could be one thing, that causes the shortcut phenomena.

edit: I remember seeing great spark plug build tutorial on HMEM, and there it was: http://www.homemodelenginemachinist.com/index.php?topic=5653.0

When making the next spark plug, I think I'll follow gbritnell's advices :dremel:.

Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 11, 2009, 12:41:11 PM
New spark plug body, made of stainless steel, except the nut, that was silver soldered:
(http://www.kolumbus.fi/kz1706/098.jpg)

Finished work:
(http://www.kolumbus.fi/kz1706/099.jpg)

Anyways, now it seems that the engine testing can continue  :wave:.
Title: Re: Another version of how not to build a model engine
Post by: Andy on December 11, 2009, 01:59:13 PM
Sorveltaja, I've been following this with interest, and I admire your persistence.  :headbang:

I'm afraid I can't offer any advice, but hope you get it going to your liking very soon!
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 11, 2009, 04:25:11 PM
Andy, thanks for your reply :thumbup:.

I did running tests with new plug, and it seems to work ok. But the engine still has quite a reckless behaviour.

I have read the previous replies, and there is good suggestions about that carburetor thing.

Hopefully I finally can concentrate to test different carb variations, besides stomping bugs. One by one, if needed  :coffee:.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 12, 2009, 12:50:50 PM
I made ptfe washer for the spark plug, and it seems to 'live', so that it's required to tighten the plug occasionally.

That's what I did. Until the exhaust valve hit the sparking plug, and got stuck :bang:.

Anyway, I got the exhaust valve to work, after some persuasion. Not sure how well it seals, but before starting to make a new valve, I'll test if it's possible to get the engine running with current valve.
Title: Re: Another version of how not to build a model engine
Post by: DICKEYBIRD on December 14, 2009, 02:21:44 PM
I just finished catching up on this thread.  Great job and a wonderful example to all of us to persevere with our projects!

One thing to try is an exhaust throttle.  I have had great success with my small (Cox .010 to .049 c.i.) model airplane engines using exhaust throttles and fixed venturis.  Granted, these are 2-stroke engines and yours is a 4-stroke but it's certainly worth a try.  You might be surprised.

Very small throttleable carburetors are difficult to make work properly due to scaling issues.  Reynolds numbers and all that rot.:)
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 14, 2009, 02:50:08 PM
Luckily I opened the cylinder head, as whole combustion chamber and valves were covered with black goo. No wonder, that the engine refused to run.

Maybe I used too much oil in the fuel. It's a run-in season for the engine anyway, so I thought, that more is better, than too little (http://emoticons4u.com/crazy/1261.gif). Apparently this isn't glow engine after all, but fire breathing plain gasoline one.

That led me to discover another problem with recently made spark plugs. 

Distance between center electrode and plug body was too short, making it prone to conduct between them, or worst, cause a shortcut.

So I made a new spark plug, after seeing some commercial plug cutaways:
(http://www.kolumbus.fi/kz1706/100.jpg)

Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 14, 2009, 04:34:22 PM
Dickeybird, good point. I have heard about that exhaust throttle thing... but as far as I can remember, it's only good for two-strokers, as their performance might even advance, when suitable, counter-pressure creating tuned exhaust pipe is used.

Back to the project. I've made some running tests, after the recent exhaust valve hit spark plug -thing. The engine barely runs. Exhaust valve doesn't seat well enough. I'm surprised, that it isn't bent at all :thumbup:.

The engine is now disassembled, and I'm going to re-lap both valves (may be the 30th time, but who counts, as it's test engine :wave: ) . 
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 15, 2009, 12:46:34 PM
The current crankcase cover is made of 1mm brass, and it's too thin, as it holds the breaker points assy also. When in use, it bends just a bit, letting the crankcase oil to leak out.

To fix that, I'll make a new one using 1.5mm mild steel:
(http://www.kolumbus.fi/kz1706/101.jpg)
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 16, 2009, 01:00:59 PM
New crankcase cover in place, after some cleaning:
(http://www.kolumbus.fi/kz1706/102.jpg)

While inspecting other parts, I noticed, that both piston top and brass button had clear mark on them, caused by knocking on each other. Not good.
That should not have been possible :scratch:. Anyways, I lowered the brass button about 1mm or so, to make sure, that it and piston doesn't get a chance to knock anymore: 
(http://www.kolumbus.fi/kz1706/103.jpg)
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 16, 2009, 03:16:50 PM
So far, but not so good; when assembling the engine, piston didn't fit to the cylinder anymore, as it did before.

Cylinder liner is bent on its lower part, being more oval shaped, than round. I guess, that something went wrong, when I did the recent sealing/tightening with it and the cylinder head. Cylinder liner has only 0.5mm 0.2mm(~0.0078")wall thickness, so no wonder, that it finally gave up:
(http://www.kolumbus.fi/kz1706/104.jpg)

Just another unexpected weak spot in my design :wack:.

Not sure, how long it takes, but I'll have to make a new cylinder liner and piston, as current ones are now useless.

During this project, I've somehow began to get used to make new parts, as needed.

Cylinder/piston fitting is one of the most challenging thing to do on any engine, but then again, I consider it as a good chance to practice my skills.

I'll post the process pics as soon as I get a chance.

Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 17, 2009, 04:05:19 PM
At this point, cylinder lapping is at the stage, where the surface begins to be satin smooth, and almost even:
(http://www.kolumbus.fi/kz1706/105.jpg)

Bore diameter will be 0.5mm smaller, than on previous cylinder, thus allowing the old piston to be re-machined, and fitted.

Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 18, 2009, 11:08:53 AM
Todays effort:
(http://www.kolumbus.fi/kz1706/106.jpg)

New brackets are made of 1.5mm(~0.059") mild steel, to add more rigidity, while previous ones were made of 1mm(~0.039") brass.

Next thing to do is to drill screw holes to both bracket, and file the excess material off.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 18, 2009, 01:28:22 PM
As can be seen, once a perfectly working cylinder has now became two sacrificial drilling templates:
(http://www.kolumbus.fi/kz1706/107.jpg)

New brackets were super-glued in place, before drilling. Screws on the above pic are there securing, just in case, if that gluing gives up.

So it's time to start the filing process.
Title: Re: Another version of how not to build a model engine
Post by: NickG on December 19, 2009, 09:32:35 AM
S, the cylinder wall looks extremely thin! Is there any particular reason for that?

Nick
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 19, 2009, 11:45:54 AM
Nick, when looking it afterwards, I have no idea why I made it so thin. Must have been some brainstorm (http://emoticons4u.com/crazy/134.gif).

But in any event, new cylinder wall is about five times thicker (1.2mm(~0.047")), than on the previous one (0.2mm(~0.0078")).

Brackets are now filed, and ready for silver soldering:
(http://www.kolumbus.fi/kz1706/108.jpg)

If all goes well, after soldering, I'll continue the lapping session.
Title: Re: Another version of how not to build a model engine
Post by: NickG on December 19, 2009, 11:53:45 AM
The new one looks much more substantial, good luck.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 19, 2009, 01:58:03 PM
Soldering session went ok. Parts soldered together, after some lathe treatment, and cleaning:
(http://www.kolumbus.fi/kz1706/109.jpg)
Title: Re: Another version of how not to build a model engine
Post by: chuck foster on December 19, 2009, 10:24:07 PM
looking good sorveltaja  :thumbup: ............... i know the feeling about making part's over and over again.  :dremel:

the new cylinder looks allot better, more solid. good luck with the lapping.
looking forward to the next write up.  :ddb:

chuck  :wave:
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 21, 2009, 05:34:46 PM
Thanks for the replies :thumbup:.

Piston/cylinder fitting is in progress. Although 3/4 is done, it's quite tedious job to do, as I have to do it manually. I tried using lathe for that, but piston gets too easily stuck in the cylinder. So tight, that it has to be hammered out.

I really don't prefer to do that. That's the reason, why I do the job manually.

One may ask: "why don't you use o-rings?". During the process, I have indeed looked for an easier way to make tight cylinder/piston fitting.

I have read about successful use of Viton as piston ring(s). Viton is available here, but it's rather expensive.

Anyway, I have enough ptfe(teflon) to make o-rings, as an alternative, as it stands heat, and doesn't require lubrication. 
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 22, 2009, 09:01:26 AM
Piston fitting is now done. After short run-in in lathe with plenty of oil, piston moves smoothly in cylinder:
(http://www.kolumbus.fi/kz1706/110.jpg)   

Compression feels good, and it's again about the time to restore the engine to its glory running condition :ddb:.
Title: Re: Another version of how not to build a model engine
Post by: Bernd on December 22, 2009, 09:45:41 AM
S,

Can't believe the dedication you have put into this engine. Your perseverance has been unbelievable. I hope this engine turns out to be a good runner after putting that much effort into it. I know I wouln't have had the patients to carry on as far as you have on one project.  :clap:  :thumbup:

This thread will hopefuly inspire others to carry through to the end on their project.

Bernd
Title: Re: Another version of how not to build a model engine
Post by: NickG on December 22, 2009, 11:20:38 AM
Very true Bernd. Today I've been looking at the poppin plans (another flame licker) incase I can't get mine going  - even down to what materials to buy! But that is pretty negative so am going to follow in S's footsteps and concentrate on trying to get mine running for a bit first!

Nick
Title: Re: Another version of how not to build a model engine
Post by: Stilldrillin on December 22, 2009, 12:15:51 PM
S,

Can't believe the dedication you have put into this engine. Your perseverance has been unbelievable. I hope this engine turns out to be a good runner after putting that much effort into it. I know I wouln't have had the patients to carry on as far as you have on one project.  :clap:  :thumbup:

This thread will hopefuly inspire others to carry through to the end on their project.

Bernd

S`s avatar prompts me to say , he`s certainly showing "dogged determination".......  :lol:


Sorry.....  ::)

David D
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 22, 2009, 05:02:54 PM
Thanks again, guys :wave:.

Hopefully this thread shows the pitfalls, that are to be avoided, when designing/building a test engine. And also, how to make the construction more durable.

David D, that saying isn't familiar to me, but I'm quite sure, that it fits for me, as I'm too stubborn to give up on this project.

Anyways, the engine is almost assembled, and the valves and camshaft are adjusted.

Tomorrow it's time to test, if how the engine runs.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 23, 2009, 03:47:52 PM
Engine was bit stiff at first, but now it runs again :beer:.

But the recently made spark plug (on the left) failed, and started to leak (http://emoticons4u.com/crazy/1261.gif):
(http://www.kolumbus.fi/kz1706/111.jpg)

It is now replaced with the older plug (on the right side).
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 23, 2009, 06:06:20 PM
At this time, the engine takes more persuasion to get it to run, than before.

But when it finally gets going, it runs more stable than ever. And even longer time than before.   

Crappy video:
&feature=channel
Title: Re: Another version of how not to build a model engine
Post by: Darren on December 23, 2009, 06:08:52 PM
Sounds like you have it under control now  :clap: :clap: :clap:
Title: Re: Another version of how not to build a model engine
Post by: sbwhart on December 24, 2009, 02:25:43 AM
Great work sounds real well  :clap: :clap: :clap: :clap: :clap: :clap:


Cheers

Stew
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 24, 2009, 08:40:32 AM
When testing, I've noticed that the spark plug has to be cleaned throughly between runs.

Reason might be that the gasoline has still too much oil in it. Even when running the engine on ethanol without oil, the spark plug had to be cleaned before next run, or engine fails to start.

After all, there is still some run-in to do, to find out how well the new piston/cylinder -combo keeps its tightness. 

Title: Re: Another version of how not to build a model engine
Post by: Bernd on December 24, 2009, 11:07:54 AM
Sounds like a fine ticking watch running.

Bernd
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 24, 2009, 04:27:01 PM
Today was the longest running session so far (~30 minutes), at a wonderfully low speed (1200-1400 rpm) :nrocks:.

I'm not sure if it's possible for an engine that small to run any slower, but I'll try it anyway :wave:.

Title: Re: Another version of how not to build a model engine
Post by: jim on December 24, 2009, 05:43:08 PM
 :clap: :clap: :clap:

well done!

its reading posts like this that keeps us all going :thumbup:
Title: Re: Another version of how not to build a model engine
Post by: NickG on December 25, 2009, 05:51:35 AM
Well done, looks like it's running superbly. Might cooling be an issue running it any longer than that?

Nick
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 25, 2009, 10:43:32 AM
Nick, the engine ran out of gas, after that 30 minutes ::).

It seems that the engine speed is low enough, so that it doesn't cause an overheating.

Adding some cooling element(s) would be cool, but not quite yet..(http://emoticons4u.com/crazy/134.gif).
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 26, 2009, 12:13:01 PM
To lower the idle speed, I increased the weight of the flywheel by adding smaller brass flywheel to it.

Also a bit of fiddling was done with crankshaft camshaft timing.

Result:


Anyways, I consider that as a giant leap for mankind in this project, closer to goal :ddb:.

Title: Re: Another version of how not to build a model engine
Post by: chuck foster on December 26, 2009, 01:47:51 PM
now that's a big improvement  :thumbup: :thumbup: 


chuck  :wave:
Title: Re: Another version of how not to build a model engine
Post by: NickG on December 26, 2009, 04:39:24 PM
It's spot on now! Were you timing it with a strobe light or something there?

What are your overall plans for this then S? Is it to dress this one up a bit or now you have the baseline are you going to start a new version?

Well done  :bow:  :thumbup:

Nick
Title: Re: Another version of how not to build a model engine
Post by: John Hill on December 26, 2009, 05:42:19 PM
Really nice result and it runs so well too! :thumbup:
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 27, 2009, 01:07:42 PM
No strobe light or any other measuring devices were used. First I adjusted the spark timing just by listening the engine(just as I did before).

Then was camshaft adjustment. It was made by the same procedure.

Next I probably have to make a wooden base, as the engine has kept my machining vise occupied ::).

New version of the engine? It's possible, but can't say yet about that.



Title: Re: Another version of how not to build a model engine
Post by: NickG on December 27, 2009, 01:30:29 PM
Just wondered what the red flashing light in the vid was, or was that getting the speed?

Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 27, 2009, 02:54:05 PM
Flashing red light came from the camera. I just wonder what its purpose is :scratch:.
Title: Re: Another version of how not to build a model engine
Post by: Bernd on December 27, 2009, 03:47:06 PM
Flashing red light came from the camera. I just wonder what its purpose is :scratch:.

The light looked blue to me, but that doesn't matter. I think it's to remove "red eye" from humans in the picture, but on a video? Maybe for focusing?

By the way S, nice running engine. Starting to sound like a nice Swiss watch ticking away.  :thumbup:

Bernd
Title: Re: Another version of how not to build a model engine
Post by: dsquire on December 27, 2009, 09:54:54 PM
sorveltaja

We are definitely going to have to give you an A++ for dedication and determination. A lot would have given up or put it aside by now but every day you kept trying something new. Watching it run now you have a very sweet running engine. If you decide to build another one I can only imagine how much sweeter it would be after having learned so much from this one. Thanks for letting us look over your shoulder as you built this one.  :ddb: :ddb:

Cheers  :beer:

Don
Title: Re: Another version of how not to build a model engine
Post by: Darren on December 28, 2009, 07:10:11 AM
It deff looks red to me, but no matter  :lol:

It's so you can't secretly film people without their knowledge  :thumbup:
Title: Re: Another version of how not to build a model engine
Post by: Darren on December 28, 2009, 07:15:36 AM
You have certainly had perseverance with this little engine ... top marks on the development  :thumbup:
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on December 28, 2009, 02:44:10 PM
Thanks again  :wave:.

This project has teached me a whole lot about how to make a four-stroker run :med:.

Especially the sidevalvers have always fascinated me, so the next project is most probably a small flatty twin four-banger.

Current engine has became kind of reference, and I'll continue the testing with it once in a while.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 06, 2010, 02:57:14 PM
The sparking plug, and especially its ptfe insulator, has became a bottleneck, when testing the engine, as it erodes away too fast.

I have made few rings out of ceramics, that are to be used on the business end of the plugs. Hopefully that eliminates the insulation wearing -problem.
Title: Re: Another version of how not to build a model engine
Post by: madjackghengis on January 08, 2010, 10:56:22 AM
As a matter of course, I would suggest the nut would work just fine, but only needs a drop of locktite to keep it retained.  I've been building Harley engines for some thirty odd years and the only thing that keeps all their parts on the engine is locktite or self-locking nuts, cotter keys, bunged up threads and the like.
    As a rule of thumb for an engine, a carb bore of a quarter the cylinder bore is usually a good size to start with.  This works well with multi-cylinder engines as well, as only one cylinder is fed at a time.  A smooth venturi with a good taper leading up to it helps keep a consistent vacuum to pull fuel in as well.  For what its worth,
mad jack
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 08, 2010, 04:12:30 PM
I've tested sparking plugs with those ceramic rings, and they definitely don't wear out as easily as ptfe.

On the other hand, as this engine is on the experimental state, even the spark plugs need to be easily disassembled and cleaned, if needed.

Using ceramics doesn't make that easier. Not to mention, that it's quite delicious delicate job to make parts from it, that fit.
Plaster anyone(http://emoticons4u.com/crazy/134.gif)?

Big white is the resistor body, that was originally attached to an automotive ignition coil, that I use. Next is one of the spark plugs.
And the marble looking thing is just one of the rings, that I managed to hack out of that big white. Bluish colour comes from felt tip, that I used to mark, where the cut was made.
(http://www.kolumbus.fi/kz1706/112.jpg)

Conclusion at this point is, that the engine still runs at too rich fuel mixture, causing contamination on the sparking plug rather easily.

--------------------------------
Mad jack, somehow that cyl.bore/4 = carb bore makes sense, as it says, that I'd need to enlarge current carbs throat from 1.9mm(~0.075") to 2.6mm(~0.102").

Anyways, the engine already runs too fast with that 1.9mm carb bore, if allowed(http://emoticons4u.com/crazy/1087.gif).
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 09, 2010, 12:13:12 PM
Today I disassembled the engine, and whole combustion chamber was again coated with bituminous black dirt :wack:.

It is now cleaned, and before going back to ordinary carburetor, I'm going to make the vapor carb, to test if it gives cleaner combustion :coffee:.
Title: Re: Another version of how not to build a model engine
Post by: NickG on January 09, 2010, 04:21:03 PM
S,

Somebody over on HMEM just finished an i.c. engine and used the vapour carb with really good results. He also tried the same tank / carb on his hit & miss engine and that ran really good too, so much so that he is making another jan ridders vapour carb for that too!

http://www.homemodelenginemachinist.com/index.php?topic=6352.105

Nick

Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 10, 2010, 03:49:03 AM
Nick, thanks for the link :thumbup:.

I occasionally visit HMEM, but somehow I've missed that excellent thread.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 10, 2010, 07:42:08 PM
I have started testing with vapor carb, and also discovered the blowback -phenomena.

Not a new thing, as the very first running test were made with same kind of carb.

Back then I tried the ball valve to prevent blowbacks, but couldn't get it to work.

But this time I'll be using adjustable spring loaded valve:
(http://www.kolumbus.fi/kz1706/113.jpg)
(http://www.kolumbus.fi/kz1706/114.jpg)

Valve needs to be lapped, and spring added. Also the bigger brass part needs some brass tube, as the whole thing is to be installed between the engine and fuel tank.
Title: Re: Another version of how not to build a model engine
Post by: NickG on January 11, 2010, 05:28:14 AM
I think you need to find where the blowback is coming from and get to the route cause - there must be something leaking on the compression stroke for that to happen. I would try submerging the engine in water, turning it over past the compression cycle and look for the bubbles.

Nick
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 11, 2010, 03:20:22 PM
Nick, most probable source of leak is the inlet valve, but as the engine has tight springy compression, leak must be quite minimal. But even that is enough to cause flowback.

I guess that the air in the fuel hose "lives" between intake strokes. Don't know how to explain it, must be one of those hunch things :smart:.

In the meantime, I have made primitive vapor carb:
(http://www.kolumbus.fi/kz1706/115.jpg)

Upper brassy thing is simple air/fuel mix adjustment, and it is to be modified, as it's bit too touchy to adjust.

The inlet tube goes below the fuel level, causing bubbles, as Jan Ridder's carb does.

Then there is the spring loaded valve combo, that was mentioned earlier.

With it, flowbacks are gone :ddb:. Even the short running tests have proved, that it works very well.

Therefore, I strongly encourage to test that, if flowbacks occur.
Title: Re: Another version of how not to build a model engine
Post by: NickG on January 11, 2010, 03:27:04 PM
S,

On Jans later version he stopped the inlet tube short of the petrol. Can't remember why  :doh: but he said it improved matters!

Good work, it only needs to be primitive to test the concept out. Does it solve your over rich mixture and sooting up problems?
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 11, 2010, 03:42:34 PM
Nick, the spark plug stays dry, and that hopefully makes it last longer, and keep the engine cleaner, than with recent carbs. I could even pass that ceramic rings -session ::).
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 12, 2010, 01:25:51 PM
I attached that spring loaded valve thing straight to engines inlet. But the engine ran poorly after that.

So I went back to setup, that is shown in the previous picture.

There is about 5cm(~2") hose between the valve and engine. Strange, but with that the engine runs much better :scratch:.

But for example, on backpressure valves, that are used on gas welding equipment, the valves aren't attached to handle, are they?

There is most certainly a piece of hose between handle and backpressure valves.

Conclusion is, that backpressure valves require some matter, be it oxygen, acetylene or air, between them and the source to work.
Title: Re: Another version of how not to build a model engine
Post by: NickG on January 13, 2010, 03:39:52 AM
Another strange phenomonen. I know I keep banging on about Jan Ridders engines, but this is because he has build a lot and done a build log of his findings with all of them. I noticed on a couple of his engines he has some sort of additional chamber with check valves in line. Not quite sure why they are needed without reading through it again but they are!

Nick
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 13, 2010, 12:07:11 PM
Well, the unlucky sparking plug saga continues.. Any spark plug, that I've made, doesn't seem to last in use(http://emoticons4u.com/crazy/1087.gif).

I have considered purchasing commercial ones, as they have ceramic insulation. Smallest size available here is NGK CM-6.

Anyways, I suspect, that 12V/2.5A, that my current power supply gives, isn't enough, causing sooty contamination.
Title: Re: Another version of how not to build a model engine
Post by: jim on January 13, 2010, 12:25:59 PM
CM-6 are 10mm thread, NGK do spark plugs down to M8 thread, don't know how long they are though :scratch:
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 14, 2010, 12:22:14 PM
In the meantime, I took the ignition system under inspection. Especially the ignition cam caught my eye (finally).

The flat part seems to be too large, as it causes two sparks. First being weaker one, when the ignition points close, and second being the desired one, when the points open.

So I turned the ignition cam's diameter ~0.5mm(~0.019") smaller, thus making the flat part smaller also, just to see, does it have any effect on that double sparking -problem.
(http://www.kolumbus.fi/kz1706/116.jpg)

Shame on me, as it might be too hasty to tell any results yet :wack:.

But it seems, that faster close/open sequence works better. Earlier, same amount of test running got the spark plug(s) sooted badly.

Now the plug stays cleaner anyway, for some reason :scratch:.
Title: Re: Another version of how not to build a model engine
Post by: Bernd on January 15, 2010, 10:08:52 AM
S,

Wouldn't a lobe like on a car engine cam work better that a flat spot? Just wondering out loud. :scratch:

Bernd
Title: Re: Another version of how not to build a model engine
Post by: NickG on January 15, 2010, 11:23:48 AM
I think this is the same sort of cam as you find in a car distributor for opening / closing the points - been so long since I've seen one though i'm not 100% sure!

Nick
Title: Re: Another version of how not to build a model engine
Post by: Stilldrillin on January 15, 2010, 12:09:18 PM
A car points cam has a peak (or 4) not a flat (or 4)......

The points should be together for the maximum time..... Not held open!

I think!  :scratch:

David D
Title: Re: Another version of how not to build a model engine
Post by: Bluechip on January 15, 2010, 02:10:11 PM
Dave SD

Quite right. I doubt if there is sufficient 'dwell angle' to allow the coil core to reach a high enough magnetic field to give any sort of reasonable spark when the points open, and the field collapses to give said spark. You may need about 120 deg., depends on the coil and the internal resistance of the source of DC. Has yours got a currrent limiter? The instantaneous current may be causing that to kick in, and that may drop your apparent DC input voltage. I had a cheapo Chinese PSU, limited at 5A, if I took pulses out of it at near 5A, the output voltage bounced all over the place. CARP !!
Briefly demoted to battery charger, on it's way to the Wheelie Bin.

Don't know whether you have one, but you should have a capacitor across the points too. About 47nF or 100nF usually work. 100VDC rating will usually survive long enough.

Dave BC

Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 15, 2010, 03:47:00 PM
Thanks for the replies and suggestions  :wave:.

Ignition circuit, that's in use, consist:

12V/2,5A PSU from an ancient (DEC/Digital)computer, that has a shortcircuit protection. That bugger seems to last, no matter how hard it is abused used.

standard 12V automotive coil

Velleman's transistor ignition kit K2543

Besides all of those bells and whistles, reducing the ignition cam's flat part/ignition points close/open sequence, as mentioned earlier, has given positive results.

Spark plug stays a lot cleaner than before. Meaning, that I don't have to make a new ptfe insulator every other day anymore :thumbup:.

The vapor carb is still under testing, engine runs with it, but especially the idle adjustment is bit tricky.

Anyway, I'm working on it :dremel:.
Title: Re: Another version of how not to build a model engine
Post by: Bluechip on January 15, 2010, 05:04:50 PM
Sorveltaja

Not sure what the kit is, but if it uses a power transistor to switch the coil primary, then NO capacitor is needed.
In fact, it usually stops it working if present.

Dave BC
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 16, 2010, 06:24:38 AM
Bluechip, there is no additional capacitor, as the kit's instruction manual emphasizes not to use any. Also the resistor, that was originally attached to the coil, was removed.

Here is the current fuel tank, that is now modified to vapor carb:
(http://www.kolumbus.fi/kz1706/117.jpg)

To simplify the flowback valve, I removed the previous one, that was outside of fuel tank, and made it to the lower end of the inlet tube:
(http://www.kolumbus.fi/kz1706/118.jpg)

That is a 4mm(3mm id) brass tube, and it has 3mm steel ball, that acts as valve. Bit of tube shrinking was done, to form the actual closing surface for that ball.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 21, 2010, 03:27:42 PM
So far, I haven't been able to get the vapor carb to work good enough. Engine runs with it yes, but only at high rpm.

Numerous setups and combinations were tested. None of the different size fuel hoses makes it any better :scratch:. Be it inlet or outlet hose/tube.

I have seen the vapor carb used successfully on bigger engines. Who knows, maybe it just doesn't work on engines this small.

Also the double sparking -problem still exists, even though I made proper ignition cam with lobe.

If this project had a name, it would be an ML. Murphy's Law(http://emoticons4u.com/crazy/134.gif).
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 22, 2010, 07:44:47 AM
Addition to that vapor carb thing, I suspect that the engine ran too lean mixture with it, no matter what :scratch:.

In the meantime, I've changed back to ordinary needle carb, that gives more richer mixture, that the engine seems to prefer.
Title: Re: Another version of how not to build a model engine
Post by: madjackghengis on January 22, 2010, 09:38:41 AM
Sorveltaja
   A couple of things regarding your fuel mixtures:  I suspect Jan stopped the fuel inlet just before going beneath the surface because of the substantial increase in vacuum necessary to get air flow through the liquid, when there is a layer of fuel vapor hovering above the fuel.  Secondly, there is a need for space between the check valve and the intake, as the intake charge is of low mass and rather low velocity, and it is its inertia which opens the check valve, and it needs room to build up velocity, given the low vacuum presented by the intake stroke.  Check valves on torches are right at the valves because a "backfire" provides plenty of power to the flame that is trying to climb backwards up your hoses and blow you up, by the energy of the ignition and its "pop".  In low velocity systems, inertia is the single most important energy factor, which is why long intake manifolds work well with small engines.  As size goes down by the square, mass and volume go down by the cube, and it is mass which is the smaller of the factors in inertia, and velocity which is the major factor in the energy of the intake charge.  For what its worth, that's my input.  I hope you get it to a happy state soon.
jack
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 22, 2010, 03:04:21 PM
Jack, thanks for the reply :wave:.

I tested the vapor carb also with vertically adjustable inlet tube, so it was possible to take its lower end as close to fuel surface, as possible. Still no low rpm running though.


As size goes down by the square, mass and volume go down by the cube.

Hmm.. that's interesting. I'm thinking of some simple formulas to calculate needed values, but there seems to be so many factors (and variables?) on the engine and fuel tank, and between them, that I can't figure out how to interpret all that mathematically :smart:.
Title: Re: Another version of how not to build a model engine
Post by: madjackghengis on January 23, 2010, 09:40:56 AM
Bluechip,
   Just for background information, dwell angles of about thirty degrees are standard on points operated engines, so the flat on the round cam should be sufficient, dwell angle being adjustable by point gap with a degree wheel, a light bulb of the right voltage, and a pointer to set when the bulb goes out, and to show the degrees when it comes on with the points open.  Typical primary voltage in a twelve volt system is two amps for a car engine, anything close to an amp is sufficient for small engines.  Thirty degrees will allow full coil "saturation" at eight grand on a racing engine, thus a full "fat" spark at the plug.  Plenty of dwell for an old Harley.
mad jack
Title: Re: Another version of how not to build a model engine
Post by: madjackghengis on January 23, 2010, 09:53:12 AM
By the way Sorveltaja, with the vellman transistor kit, you should be able to use a micro switch to trigger the ignition without any problems of burning it up, as there is very little current in the trigger circuit.  I've used one of those conversion kits in a truck to good results.  It should give you a fixed dwell angle established by the circuitry.  Are there any magneto experts out there who have managed a miniature magneto?  I'm looking at the possibility of one for my radial, never made one before so any insight would be appreciated.
   On the fuel tube, you might try putting it right below the surface, with a tiny hole right above surface level, that's what some regular carburetors appear to use to control low speed mixture.  A slash cut end, with just a bare bit of opening showing above the fuel level might work as an alternate.  Quite a lot of dedication, getting this engine running up to snuff, looking good :bang:
Title: Re: Another version of how not to build a model engine
Post by: NickG on January 25, 2010, 08:08:32 AM
Madjack, I have also wondered about miniature magnetos. I have a book which must have belonged to my grandad, model petrol engines by edgar t westbury. It has a section in there about them. I will see if there's anything useful in there and let you know.

Nick
Title: Re: Another version of how not to build a model engine
Post by: madjackghengis on January 25, 2010, 09:49:59 AM
Madjack, I have also wondered about miniature magnetos. I have a book which must have belonged to my grandad, model petrol engines by edgar t westbury. It has a section in there about them. I will see if there's anything useful in there and let you know.

Nick
Hi Nick, I've got an old book of ETW's on miniature ignitions and magnetos I'm studying, it has a lot of information and gives dimensions and numbers for winding miniature coils and the like, as well as some ideas on miniature magnetos.  It was written in the thirties of forties, with the new "rare earth magnets" available, I think I can get a working magneto in scale size certainly for a single cylinder hit or miss, and with one of the chapters on the Scintilla magneto, and how it works, maybe one that would run nine cylinders in at least close to scale size.  ETW was the mother lode for model engines for forty or fifty years, and designed some stirling engine generator/radio sets which were dropped behind occupied lines in WWII and used by the French resistance, to comm with the Brit forces.
It's getting hard to find any of the books he's written, at least here in the states.
mad jack
Title: Re: Another version of how not to build a model engine
Post by: Bluechip on January 25, 2010, 11:11:42 AM
Folks

Got this myself,   absolutely excellent.

Was sold by Camden in UK, no longer stocked by the looks of it tho.

scroll down the site a bit ..

http://www.bobshores.com/

Dave BC
Title: Re: Another version of how not to build a model engine
Post by: NickG on January 25, 2010, 03:36:33 PM
Mad Jack,

Wow, that's impressive, didn't know he was that well known!

The bob shores one sounds interesting too BC.

Nick
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 28, 2010, 08:01:03 AM
That magneto thing is interesting... how small can be made?

I have ETW's 'Atomag Minor' plans for miniature magneto and the coil. Drawing has the magnets diameter ~30mm(~1.18"), and its material is alnico.

With modern neodymium magnets, maybe even smaller size (and more efficient?)would be possible?
Title: Re: Another version of how not to build a model engine
Post by: Bluechip on January 28, 2010, 10:59:52 AM
Folks

The principal problem with mags. miniature or not, is the HV coil.

This consists of some 15,000 turns of very fine magnet wire. about 44 AWG, or it's metric/SWG equivalent.

The breaking strain on this stuff is near zero. If it lets go, you start again. I have wound coils on a commercial winder with wire a good bit larger. I broke that! You won't do it on a lathe unless you are very lucky indeed.

(BTW, those coils were not for ignition stuff, just rewinding solenoids).

Assuming you have succeeded in making the coil, you then have to vacuum impregnate it. This needs to be done to give a path through the coil to get rid of the heat. Otherwise, you get a hot spot, and the coil will fail.

Then, there is the problem of sourcing the armature stampings ...

There was. at one time, a UK maker of Mini-Mags .. no longer in business I think. (Jim Shelley).

Some reference here.. by some one we once knew, I think.

http://www.floridaame.org/discus/messages/14/321.html?1182878736

There is a wealth of practical information in Bob Shores book. I strongly suggest anyone even contemplating home building coils/ magnetos gets a copy and reads it before committing any significant time or money.

Dave BC
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 28, 2010, 11:37:42 AM
Dave, it sure is tricky to wind hair-thin wire. I've been trying that, and there is way too big chance to snap it.

Unless you have a delicate winding setup.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 28, 2010, 01:40:46 PM
In the meantime, I've done some more testings with vapor carb.

Earlier I just couldn't get it to work with ethanol. But today I almost accidentally found out, that it's all about temperature.

With gasoline it hasn't been a problem, since it has lower flash point, than ethanol(just a guess).

I remember reading somewhere, that it could be tricky to start the engine, that uses ethanol with vapor carb on the cold circumstances.

I was in doubt, when I tested following:
(http://www.kolumbus.fi/kz1706/119.jpg)(http://www.kolumbus.fi/kz1706/120.jpg) 

But it seems, that even few degrees increase of temperature makes ethanol to vaporize :ddb:. Once the hand is removed, lukewarm is gone, and soon the engine stops.
Title: Re: Another version of how not to build a model engine
Post by: Bluechip on January 28, 2010, 01:45:49 PM
Dave, it sure is tricky to wind hair-thin wire. I've been trying that, and there is way too big chance to snap it.

Unless you have a delicate winding setup.

Bob Shore mentions 16,000 turns of  44 AWG  ( 0.05mm  or 0.002")   I can't see it, let alone wind it.   :scratch:

This might be handy ...

http://wires.co.uk/acatalog/conversion.html

Not for me ...

Dave BC
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 28, 2010, 02:25:41 PM
Once I was interested in making my own pickups for the electric guitar. Wire is somewhat similar size, that is used on ignition coils.

There was the site, where the winding how-to tricks were shown. I think, that same techniques applies to both pickup and the ignition coil winding.

Only the latter being easier, since it's cylindrical..
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 30, 2010, 04:32:10 PM
I've been testing vapor carb with gasoline and ethanol. Neither provides low rpm, no matter how many setups are tried.

So I moved back to ordinary carb. With it, engine runs at very low speed, but only randomly. Otherwise it likes to run horribly fast.

After all, I have a strong feeling, that it should indeed be possible to get even that small engine to run nice and slow (huff'n'puff'n).
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on January 31, 2010, 03:57:06 PM
Last, but hopefully not least, I'm going to make a more standard needle and barrel throttle carb.

I can make the engine run already, and that should be enough, but maybe just one more test(http://emoticons4u.com/crazy/1087.gif), before this project is finalized.
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on February 01, 2010, 02:21:44 PM
New carb so far:
(http://www.kolumbus.fi/kz1706/121.jpg)

On the left side is needle, next is part, that needs an inlet for the fuel to be made. It has 1mm(0.039") hole going through, and piece of smaller tube (od 1mm, id 0.5mm), that goes to carb's throat.

Then is carb body, that has different taper on inlet, and outlet. It has 7mm(0.275") hole for barrel. There is also pair of d-bit reamers, that were made of otherwise scrappy drill bits. Fastening part of the cheap drill bits seems to be softer, allowing them to be filed, if desired. No hardening was done, and still they cut the brass better than I expected :thumbup:.

At the right side is the barrel, that has 2mm(0.078") throat. 
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on February 02, 2010, 01:17:43 PM
So I tested the new carb, and it's just as super sensitive, as earlier ones.

But anyway, that's it folks about this project ::).
Title: Re: Another version of how not to build a model engine
Post by: Stilldrillin on February 02, 2010, 01:21:59 PM
So I tested the new carb, and it's just as super sensitive, as earlier ones.

But anyway, that's it folks about this project ::).

And what a project it has been!  :clap:

Tremendous patience and perseverance.......  :bow:

Thank you!  :thumbup:

David D
Title: Re: Another version of how not to build a model engine
Post by: NickG on February 03, 2010, 05:20:55 AM
Yes well done S, it has been a fascinating project to watch unfold and you have a really good runner in my book. What's the next project?  :proj:

Nick
Title: Re: Another version of how not to build a model engine
Post by: sorveltaja on February 03, 2010, 06:00:24 AM
Next is probably a side project, to make a simple top slide for my lathe :coffee:.