Flame eater questions

[lazylathe]

Hi All,

I recently picked this engine up on fleabay for a steal!


It seems to be in pretty good condition, jut needs a good cleaning and polish.

I am having a hell of a time trying to get it to run!
The original spring steel valve(?) did not cover the whole inlet port so i made a new one that fits a lot better and actually gives the engine
a bit of compression.


It has a graphite piston.
The only ball bearing is the one that runs on the cam.




I am wondering if the return springs tension is too high.
It takes a lot to spin it over.

I have tried lots of different flames, sizes and positions but no luck.
I have also read through all the builds here and tried a few of the mentioned fixes to no avail...

Any help would be greatly appreciated!!
My fingers are aching from flicking over the flywheel for the past 2 hours....

Andrew

[Stilldrillin]

Andrew.
I guess you know what it is.......    http://www.pmresearchinc.com/store/product.php?productid=3095&cat=0&page=1

I don't own one of this model, but they are said to operate very easily. Used (almost) as a toy.

http://www.toysteambible.org/PM_Research/Flame_Licker.htm


All my flamelickers need the cylinder warming with the burner flame, then will start easily.

David D

[lazylathe]

Hi David,

Thanks for that link!!!
I was still a bit unsure of what i had bought...
A little research from the link you provided gave me some insight!

It appears to be from around 1986!
Could be one of the first prototypes as mentioned here about half way down:
http://www.craftsmanshipmuseum.com/other.htm

There are some differences though...
Mine does not have the oil holes drilled into the crankshaft supports.
The cam follower is a different design.
And the valve it had on when it arrived looked home made. I changed it for a better fitting one.
The base is exactly the same minus the little plaque.

Thanks for the help!!

Andrew

[Stilldrillin]

Yer welcome Andrew!

No doubt, others on here who have actually built 'likkers, will give you more technical advice..... 

Good luck! 

David D

[lazylathe]

I have been in contact with PM Research and sent them a picture of my engine.
They have never seen one that old and of that design.
The new owners took over in 1990 so it is prior to that!

Mine was missing a few essential bits that i have ordered from PM Research.
Hopefully once they arrive and are installed it will roar into life!

For now i think i will polish it up!
20 years of grime does not look good with my other nice and clean engines!

Andrew

[Stilldrillin]

Some crisp pics here, of an original engine.....     http://www.ebay.co.uk/itm/VACUUM-ENGINE-SOLAR-3-FLAME-EATER-/270825644643?pt=UK_Trains_Railway_Models&hash=item3f0e776263

David D

[NickG]

Hi Andrew,

Just seen this, I've built three of these, all runners (last one on Monday night!) but they can be very tricky to get to run. First thing to check is the friction ..take the cam off, leave the valve open and see if it spins over nice and freely. I would say it should do 15 to 20 revolutions with a firm flick over of the flywheel. The next thing is, you are quite right, that spring does look a bit strong to me.

The cam should shut the valve fully shut about 45 degrees before bottom dead centre. To clarify, the piston is on it's outward stroke and the valve shuts before it reaches it's extremity, roughly 45 degrees but it shouldn't be too critical, both cam operated engines I made I set by eye and they ran 1st time. The overlap of the valve with the port shouldn't be too much ... about 1mm or so will do but noticed yours has a turned up end on the valve so take this into account with the overlap. Although I can see why you've done it, this feature is not the best idea as it will not aid sealing. If it's thin enough however, it should still work. What gauge shim is it made from?

When the pressure increases inside the cylinder to above atmospheric, the valve (again if thin enough) will push off its face so it's sort of automatic opening. As long as the cam is right, the spring will open it. It must be fully open again by top dead centre to start the cycle again.


The other thing to ensure is that you have a big flame covering the whole port if you can. This isn't critical on some engines but that is quite a big port you have there.

Hope this helps.There must be many of these engines under the bench destined never to run though as it's easy to get frustrated with them!

[geoff_p]

Gents, I've just joined Modders and have found your topic, with Nick's insights.  I am at the 'frustrating' stage with my first Flame-eater, trying to get it to go.

My valve is graphite/carbon - an electric-motor brush - moved up-and-down on the cylinder head, and 'farts' quite nicely on compression.  For the moment the valve spring is an elastic band (mounted away from the flame, I hasten to add) so I can easily vary its tension.  The valve is operated through a sort of bell-crank from a cam.

The port is about 3/4 of the piston-diameter by perhaps 3/8 wide, and the head has been thinned-and-tapered in the port area.

But the cam is giving me heartache!  OK, Nick, you suggest closing about 45 degree before BDC, but when should it open?

I have written some CNC-code so I can make different cams fairly easily, and so far have tried 90 degree dwell, 120, and 180 dwell all with quite snappy opening and closing actions.  However, none of them work!

Any advice will be hugely appreciated,

Geoff
Thailand

[NickG]

Hi Geoff,

Is your flame eater to your own design? The port sounds a bit large to me if I'm honest, would it be simple to make it smaller? There are so many parameters that affect when the valve should open! Myself and a member of our club were discussing it a few weeks back at an exhibition. We were surprised to find that mine opened well before TDC on poppin - at least 45 degrees.

However, when the engine is actually running, because of the valve design on poppin, you can get a kind of latching effect, where the 'vacuum' or sub-atmospheric pressure in the cylinder will keep the thin valve against the port face until the pressures equalise and it is ready to open.

If your cam is such that the valve is forcibly opened at a set point, if you open it too late, the pressure in the cylinder will start to rise above atmospheric and you will be trying to compress the gas, sapping power. It's a trade off with the how long the power stroke is, I would try around 45 deg before top dead centre as a starting point.

Another guy on here called madjackghengis did quite a few successful flame eaters and a lot of experimentation. We worked out that one of the good design points of 'poppin' was the whole valve mechanism ... low friction pivots, low spring tension required and a very thin valve that lent itself to sealing well. Also, the cam forcibly shut the valve at a pre-determined point, however, it didn't forcibly open it again, that was essentially left to its own devices providing the spring pressure wasn't too great. This meant it wasn't that critical and relatively easy to get a running engine ... both of my poppin engines ran with the first flick of the flywheel which was a revelation considering the Jan Ridders one I made!

The cam dwell is 110 degrees on poppin but as I said it's very insensitive to cam timing ... I set both by eye and both ran that well I didn't bother messing around with it!

Hope this is of some use and you can get  it running ... they are fantastic to watch, don't give up, I nearly did on my first one but glad I persevered. Incidentally, that still doesn't run too well, takes a long time to warm up and is very tempremental. I am planning to one day make a graphite piston and valve for it.

Best regards,

Nick

[geoff_p]

Many thanks Nick, for that detailed reply. 

Yes, it's my own 'design' (though that is a rather posh term for it) and is made from my own castings.  Originally it was going to be a hit-n-miss petrol engine but I have to save-up for some gears, so I thought perhaps I can use the parts for something-else in the meantime.

Since most flame-lickers I've seen have cooling-fins I thought the water-hopper cooling should be good.

The main bearings are ball-races and the Big-end runs on graphite/carbon - two motor-brushes held together and bored out.  The crankshaft is cast aluminium, and the graphite & aluminium combination seems to offer very little friction - it spins very easily. 

The 25 dia x 30 stroke cylinder and piston are both aluminium and match very well: with a hint of oil they have only a little blow-by.

The valve is another carbon-brush, 17 x 11mm, for a port which is 13 x 6, and is held against the port-face by spring-tension (emulated here by an elastic-band until I can find my box of springs!) and that same tension acts downwards on the lever-arm to keep the cam-follower in contact with the cam.

This morning I made another cam, very similar to Nick's and calculated to be 90 degrees opening-dwell but although the engine pops-and-farts with it, I didn't get a run.  I'll try again later, varying the timing.
Geoff

[NickG]

No problem Geoff,

Always try to offer advice on these as I know how frustrating it can be getting them to run! The only other obersvations I can think of at the moment are:

Try not bothering with any water in the hopper .... sometimes these engines (which you may think is contradictary to how they work!) need to get quite hot before they'll run! It's possible to get a quenching effect where the gases are cooled too quickly before the valve has closed!

Bearings and friction reduction measures seem good.

Aluminium on aluminium for piston / cylinder wouldn't be my first choice, because they tend to scratch easily giving a high coefficient of friction. However, if they are a very good fit and the friction is minimal there is no reason it shouldn't work, at least initially. I have to use oil on my Jan Ridders flame eater or it doesn't run. The oil helps both seal and prevent combustion products from adhering to the parts increasing friction massively. I use engine oil as it can cope with the high temperatures and is very thin when hot.

The port, now I have seen it looks about the right size to me.

The only other thing is, as I mentioned before, your valve is being forcibly opened rather than closed, it is good that your valve slides upwards away from the flame, but it may be better if the belcrank operated from the top of the cam, closing the valve for the required time and position rather than opening it. Also, make sure the elastic band isn't too strong, I suspect the farting noise is the valve lifting off its face which is desirable anyway, like a sort of automatic relief valve when the pressures equalise ... but you want this to happen very easily.

I don't think it's far off, other than above, what fuel are you using? I use denatured alcohol or industrial meths - a cleaner burning methylated spirits. Make sure your flame covers the entire port, if you draw in cold air it can scupper the whole thing!

Nick

[geoff_p]

Thanks, Nick.
In no particular order:
I haven't been putting water in the hopper except when the whole thing gets too hot to handle.

Having tried 'nightlight' candles, gas cigarette-lighter, blowtorch and all else in between, I now have a source of Gel Fire in tin-can packages similar to nightlights.  The carton lists the fuel as 'Alcohol Fuel' and 22gms each.  At 2 Baht (4 pence) each, they are cheap enough for experimenting, though the flame is quite massive - it certainly covers the port.

Last night, in the dark, I was able to observe that the flame is blown away from the port by the exhaust air and because of the lump-of-graphite (AKA 'valve') above the port deflecting it, that often also blows the flame out.  Watching a few YouTube videos I noted that many open the valve downwards so the exhaust air goes up-and-away from the flame.

So this morning I'm lowering the valve - it will now open downwards and close upwards.  Which will, by coincidence cover your point about the cam forcing it closed.  Checking the cam-lift, it looks as though the timing can be adjusted to suit.  I hope!

Mind you, I may have to wait till after dark to see the blooming flame.

Geoff

[NickG]

Geoff,

Good point there ... but, the valve might get in the way of the flame being sucked in this way around ... worth a try though and if it solves the other issue with the cam closing the valve rather than opening it, it may work,

Your flame sounds good.

With regards to blowing the flame around, they can be very sensitive to that ... on poppin, if you put the flame directly in front of the port it runs very fast, but then gets so fast it blows the flame out! On the Jan Ridders engine, Jan did lots of work to find the optimum flame position ... which is actually not in front of the port but to one side ... bit different as it's a side port on that though. It will not run at all with the flame in front of the port though. Try experimenting moving the flame around to many different positions if you can easily do so.

All of that said, if the engine in sucking the flame / hot gases in properly and the valve is closing at the right time, the engine should actually exhale (if that's the right word!) much air, only right near the top of its stroke with internal and external pressures equalise.

Nick

[geoff_p]

I wish I had your analytical mind, Nick. 

You are dead right about the 'lump' getting in the way of the flame. It looks as if there is only a tiny amount 'bending in' to the port, and that only sometimes.  Mind you, when it does, there is a definite vacuum lock - as soon as the valve closes (turning the engine with my finger in the flywheel-spokes) it becomes very hard to continue through to Bottom Dead Centre.

Aha!  Retard the valve-closing.  Nada! Zilch! Nothing at all. not even a 'fart'.

Try something radical:  I turned the engine 90 degrees and held the wood base in the vice so the hopper faced me and the port was vertical.  With a candle's yellow flame I could see the flame sucking into the port quite strongly, so I swapped for a blowtorch (turned down to a very low flame) and suddenly I had power.  A lot! But not for long as it threw/blew the valve off the engine after a few revolutions.  But it had definitely accelerated away from my fingers.

So now I will rotate the cylinder-head 90 degrees and am rebuilding the valve gear to move sideways.  Pictures to follow when I have made a bit more progress.

Geoff

[NickG]

Geoff,

as soon as I read the bit about the vacuum lock I thought ... retard the valve closing ... then read your next sentence!

It sounds like it's almost there then, it's surprising how sensitive to things like flame position a lot of them are and I always thought a vertical cylinder would be better too. Poppin seems to run just sucking a little flame in, it runs best with the flame at a slight angle so as not to get disturbed by the residual air exhaling, but the other won't even try if there's the slights mis placement of flame or a flicker.

Nick

[lazylathe]

WOW!!!!
Just seen that this post has some activity!!!

Great additions guys!!
Love the look of the flame likker you are building Geoff!!!

Just an update on my find!
I did get it all shined up and running like a champ!
Takes a bit of heat to get it warmed up before it will run but then it runs till there is no more fuel.

Here is a link to a short clip of it running:

Keep up with the updates Geoff!!!
I would love to see a video of it running in the future!

Andrew

PS: I have recently acquired another basket case flame eater...
Also a non runner, just not had enough time to look it over carefully.
Here is a clip of it:

[geoff_p]

Sorry Chaps, there is little chance of a video - I don't have any means of taking one: my phone makes phone-calls only, and my camera takes still photographs.  However, if I can get this beasty to run, I will try to borrow a moving-pictures machine.

Meantime, back in the jungle ....

The following is a lash-up!  I'm not inclined to use the brass-bar I have on suck-it-and-see parts, and cardboard just won't stand being near the flames.  But I do have a sizable sheet of 1/2mm brass and a pair of tin-snips.

Hopefully the photo-titles will make sense so here goes..

And, by the way, the flywheels are now filled with lead and together weigh about a kilo.  With the port held open, the thing will spin on for perhaps 30 to 40 revolutions, showing there is little friction.
The raggedy sheet of brass at the back is a windshield because my workbench is outdoors.

All suggestions on a new Royal Mail first-class stamp, please.

Geoff

[geoff_p]

Oops!  I forgot to mention, the spring (in top-view of bell-crank) presses the valve onto its face and, by Newtons Law of reaction, also 'encourages' the cam-follower to "Follow that cam!"

I originally had a separate spring pulling the cam-follower but it fell off and I realised it wasn't doing much good.

Geoff

[NickG]

Well done Andrew, yours looks excellent and runs superbly now!

Geoff, it sounds like everything is nearly there - I don't know what else to suggest now except going back to that blow torch flame when you felt it kick - play around with the timing with that setup and you might get some joy. If you've got low friction & good seals that's half the battle but there are many other factors too. It's hard to say past that as my 3 engines were proven designs so I already knew the basic parameters worked i.e. bore, stroke, port size, flywheel size etc.

Have you tried lighter flywheels? They may be too heavy for it to turn over under its own power. The only other thing to note is your piston and cylinder materials aren't ideal, although at least they are the same so differential expansion shouldn't be an issue, but it might be worth trying to measure them hot just to rule something else out. I tried an alloy piston in my first one and the whole thing locked up solid when hot but that was with a cast iron cylinder.