Another Halo

[kvom]

I was inspired by Bogstandard's starting to build this engine, and it looks as if it would be a good (or ambitious) step up from my first two engine builds.  For those not familiar with this engine, it's a 5-cylinder radial:  http://lineymachine.googlepages.com/l5

Since I still have a bit to finish on the paddleduck engine plus blinging it and Brian's beam engine, I will work on this in fits and starts as time allows.  In the meantime, I still need to accumulate tooling that I don't already have.

The first bit is a very thin lathe grooving tool to cut the fins on all the cylinders.  The plans are showing .025" grooves, which seems almost paper thin to me.  I started by grinding the end of a 3/8" HSS blank on grinding wheel, till I had a width of around .07.  The remainder of the forming was done on a manual surface grinder at school.  The sides need relief, and I guessed at a 5-degree angle on each side (10 degrees included).  To achieve this a sine vise was set at 5 degrees using gauge blocks, and the tool clamped in the vise with the botom side of the tool on the high side of the vise jaws.  The left side of the tool must be ground first in order to support the tool with a parallel when grinding the other.

I actually ground it down to .03", but decided to stop as the bottom of the blade was so thin.  I think that 5 degrees is too much; 2.5 should be better.

Here's the result:



The grooves in the cylinders aren't critical dimensionally, and aren't deep.  We'll see how the tool holds up.

Another necessity for this build is a rotary table, and some of the parts will require that it be mounted vertically.  I'd been shopping for a rotab for some time, and found the PhaseII 8" H/V on sale at Enco.  Unfortunately they don't honor the free shipping on these tables, and shipping would have raised the price by $75.  I called MSC, which has a local presence, and was told that they'd match the Enco price.  I picked it up yesterday.

Here's the packing crate, along with a piece of 2" thick 8" diameter aluminum that will be a mounting plate for a 3-jaw chuck eventually.  Alongside is a D1-3 backplate that I will use to attach the aluminum to the lathe to cleam up the faces and edges.



Inside the crate:



Once out on the workbench, the only pre-install work to do is to fill it up with spindle oil and clean the table surfaces with mineral spirits to remove the residue from the packing paper.  A thin coat of oil and it's ready to be mounted.



Eventually I will try to separate the vise and rotab a bit, but I didn't want to take the time to retram the vise.  There is no vertical slot for the right side.  Instead it must be clamped with this supplied bolt/nut/clamp combo:



There are two other similar clamps for mounting the table vertically.

Here's the first method I decided to try to center the table under the mill spindle.  I chucked a conic piece of metal that I use as a bull center on the lathe, and positioned the table so that it fit snugly all around when lowered into the center hole.  Then I measured with a DTI, which showed that I was less than .005 out of dead center; good enough for me!

[Brass_Machine]

Way to go Kirk. You are wacking these projects out! 

Eric

[CrewCab]

Nice choice Kirk, (Engine and Rotab    ) .............. you'll do fine, looking forward to the build.

CC

[bogstandard]

Kirk, I am half way thru the finning on the cylinders, but I am unable to finish them at this time.

I ground up half a dozen 0.025" wide by 5/16" long slotting tools, I ground them up 3 at a time. But I do ensure that my machine is set up exactly square on the topslide and toolpost, so I can get away with no side relief, as you are plunging in only a matter of 0.125" at the deepest point. Doing them with no side relief, they are not so fragile.


John

[kvom]

The rotab doesn't come with t-slot nuts, and the slots are smaller than the Bridgeport table.  I measured the slots and did some internet searching.  One eBay seller has the size that fits at a price of $22 for 10.  I can't see needing 10, so I decided to make some.  The width of the slots at table height indicate the need for 7/16" studs.  I happen to have a foot or so of 7/16-14 threaded rod, so I went to the local hardware store to purchase the locking nuts.  Once back home, I discovered that my tap&die set has a 7/16-20 NF tap but not the NC.  Back to the store to purchase a tap.  I then spent a good 90 minutes milling the necessary t-nut form in a 6" length of mystery steel, then cut off 4 1" pieces.

After dinner I need to drill and tap them.  Hoping not to break the tap. 

[Bernd]

Kirk,

I've got almost the same table, except mine doesn't stand horizontal. I discovered the same thing about the bolts and T-nuts. I improvised for now but need to make a good pair like your doing.

This is going to be a very interesting thread. Why? Because I think you did an excellent job on the Paddleduck engine and I'd also like to build a radial engine but steam power it and put it in a boat. Sort of like an air boat but with a steam driven engine.

Bernd

[Brass_Machine]

...I'd also like to build a radial engine but steam power it and put it in a boat. Sort of like an air boat but with a steam driven engine.

Bernd

Holy cr@p Bernd... we need to talk. I want to do this as well!

Eric

[Bernd]

...I'd also like to build a radial engine but steam power it and put it in a boat. Sort of like an air boat but with a steam driven engine.

Bernd

Holy cr@p Bernd... we need to talk. I want to do this as well!

Eric

Ok buddy lets talk. Start a thread. I've got some ideas.

Bernd

[kvom]

Theoretically the Halo can run on steam as well as air.  However, it doesn't have a reversing mechanism AFAIK.

[kvom]

After working on the Jeep all afternoon, I still had a little energy for some machining.  I haven't decided which chuck to buy for the rotab, so I started to clea up the aluminum mounting plate.  The first order of business was to center the lathe backing plate and drill four holes for the 3/8" mounting bolts:



Note that since the disc is 8.25" in diameter, I couldn't clamp it normally in the vise.  I reversed the hard jaws on the Kurt to allow clamping it for drilling.  Once the holes were through, I pulled out this countersink that I'd gotten along with the mill in order for the bolts heads to be below the surface of the plate.  Useful gadget.   



Now I could mount it at the lathe to clean up the face and sides.



Then I flycut the reverse side, hopefully yielding a plate that will keep the enventual chuck square to the rotary table.



The next step will be to cut 4 mounting notches into the side to attach it to the rotab.



In the meantime, I will try to start on one part via the CNC lathe at school: the cam housing.  The first screenshot show the simulation of the g-code for drilling and boringL



Assuming that works properly, the piece is turned around in the vise to turn the front profile:



I'll try them out on Monday afternoon.

[kvom]

Still tooling up.  Today I ordered two lathe chucks.  One, from a guy on PM, is a 8" 3-jaw chuck with D1-3 mount and 2-piece jaws.  This will go on the lathe when I need soft jaws.

The other is a 6" plain back 3-jaw with 2-piece jaws that will go on the rotab using that Al disc as a backplate.  Found it on eBay.  Assuming TIR isn't too bad I should be able to center work on the rotab.  Otherwise I can make some softjaws and center them that way.

Otherwise no shop time today as I was helping a friend swap a transmission into his Jeep all afternoon.

[kvom]

I ran the first half of the cam housing programs on Monday night.  Since we have to share the lathe setups, I was limited to using the existing chuck jaws and could add only the boring bar to the tool holder.  The existing 3-jaw chuck was set for 2.5" stock, so I needed to adjust the program to turn down from there as opposed to the 2" I had previewed.  Since I couldn't drill the stock on the CNC lathe, I needed to cut off 1.5" thick discs of aluminum bar and then drill them 11/32 on the manual lathe.

Using the Haas CNC lathe there is a longish setup process where you touch off the X and Z coordinates for each of the tools your program will use.  In this case I was using a turning/facing tool, the boring bar, and the parting tool.

Eventually I was ready to cut, but the first pass is always done by single stepping the program to make sure there are no gross errors that would break a tool or the machine.  This took about 15 minutes to make the first piece.

Next I measured the critical dimensions.  The boring depth was .008" too shallow as measured by a depth mic; since this is the critical dimension of the part, I needed to adjust an offset for the boring bar to correct for it.  The second part was run a bit faster, and then remeasured to confirm the dimension is fixed.

The last two copies were run at full speed, taking 7 minutes machine time for each.  One of the pieces received a slight ding, likely by hitting a chuck jaw when parted, so I have two "scrappies" and two good parts.



The second program to form the nose will likely need to wait until next week, as a new set of jaws needs to be bored to chuck the 1.675" diameter.  The jawqs on these lathes are hydraulic and have a fairly short range of motion, so they need to be bored to fit a given depth and diameter, similar to soft jaws.

So producing the two good parts has taken approximately 6 hours, including programming, setup, and run time.  It's interesting to work on, but obviously inefficient for small quantities.

[kvom]

My cheap Chinese 6" plain back chuck arrived today, so I spent the afternoon working to get it mounted on the rotary table.  The first order of business was to verify the bolt circle of the 3 mounting holes in the back.  I used the DRO to verify that the diameter is 142mm.  The holes are threaded M8-1.25, so it was off to the hardweare store to purchase bolts of the correct length.  After drilling the holes, I countersunk them with a .5" endmill, but the hole was still a few thou too small.  A few minutes with the grinder reduced the heads so that I could attach the chuck to the plate.

With the chuck mounted, there is a 3/4" flange remaining for mounting slots.  I milled these with the side flutes of a 1/2" endmill.  Here's the result mounted on the rotab:



I removed one of the top jaws for measurement, as I intend to make a set of soft jaws.  The next step will be to center the chuck body on the table, and then meaure the runout with the supplied jaws.

[kvom]

I got the rotab chuck dialed in today, although it took longer than it should have.  I was holding the DTI in the spindle, but a better idea would have been to use the mag holder on the table.  Indicating on the body of the chuck, I got it to within .001".  Next I need to clamp a round bar and indicate to see how concentric the jaws might be.

[kvom]

I chucked one of the "bad" cam housings in the rotab chuck and measured the runout.  It was .0015, so I "think" I will be within tolerances for building the parts that need the rotab.

Now for another issue, I needed to move the rotab closer to the center of the table.  Now to re-center under the spindle, I no longer have access to the precise center hole of the rotab table.  I think I will probably just use an edge finder againt the chuck body, or else on whatever workpiece is chucked.

I could also try using a DTI against the table's outer diameter, although that's a bit more time consuming.  Since the chuck runout is small, I don't think I gain anything that way.

In the meantime, I ordered and received a parts list from Liney that has all the screws, bearings, balls, washers, etc. that is needed to complete the engine.

[kvom]

I turned the other side of the cam housings on the Hass lathe last night.  It turns out one of my "good" ones was bad, so I have 2 good ones left.  Hopefully I can drill all of the necessary hole (20 of them) without screwing up on at least one of the two.



Since I have machine time this week, I'm going to CNC the driveshaft as well.  I cut off 4 2.25" pieces of 3/4" 12L14 before leaving.  I'll face off at home, and turn them on Thursday if all goes well.

[kvom]

I turned the driveshaft on the Haas lathe at school last night.  I started with 4 blanks and managed to get only one good part.  Since I won't have access to the machine anytime in the near future, this is the last part for this engine I'll be able to make via CNC.  However, at least I was able to run the two parts that have the critical dimensions.



I still need to drill and tap 3 holes on the part.  I bought a package of 2-56 taps on eBay yesterday, so hopefully I won't fubar this.

[kvom]

I received the package of HSS 2-56 taps I bought on eBay a couple of days ago, so today I thought I'd test one before using it in anger.

The test victim was a piece of 1/4" aluminum plate.  I drilled a through hole with the requisite #50 drill.  One impression is that the Bridgeport spindle is not very sensitive with drills this small.  I pecked it mainly looking at the swarf coming out of the hole.

Then off to the tapping stand.  While the tap cut the threads very cleanly, again it was difficult to sense any feedback through the fingers.  Nonetheless a SHCS screwed into the hole, so I can claim a success.

This engine uses over 100 of these screws, so I'd better get used to drilling small holes and tapping them. 

[kvom]

Having made the soft jaws for the rotab, I wanted to use them for the cam housing.  Here the portion that needs to be clamped is the 1" diameter nose.  So I needed a pocket 1" in diameter wth a maximum depth of .400".  

First, I positioned the rotab as close as possible so that its center is lined up with the mill's spindle.  The jaws are clamped onto a steel round to keep them fixed.  With a 5/8" end mill, I drilled down into the jaws .400" and turned the rotab 360 degrees.  Then with the mill turning I locked the y-axis and moved the x-axis .05", then again rotated the table one full turn.  I repeated this until I moves the X-axis 3/16", resulting in a pocket 1" in diameter.  With the pocket formed, I then used the endmill to level the top of the jaws so that the piece would sit level when clamped.  Here's the result:



Note that it is not sufficient to bore or drill the pocket using the mill, as the spindle may well be several thousandths off from the center of the rotab.  Doing it this was ensures that the center of the pocket coincides with the center of rotation of the table.

Next I chucked the piece in the soft jaw pocket, and using a DTI in the spindle indicated its inner diameter.  Thisa allowed me to line up the center of the cam housing with the spindle, and thus also with the center of the rotab.  Now with the rotab in the horizontal position I can drill the 10 mounting holes, with each pair 72 degrees from the next.  Note that I don't have to worry about drilling through holes into the jaws.



Once these holes are drilled, I will need to place the rotab in the vertical position without removing the cam housing, so that the pushrod holes can be drilled 72 degrees apart but 36 degrees from the mounting holes.

Now with the holes drilled:



I put the rotab into the vertical position, square it to the table, and then center the workpiece on the Y-axis using an edge finder.  I quickly discover a small "problem":  the jaws are too close to the work and too long to reach with a drill in the spindle. 



So my original plan won't work.  I didn't want to cut the soft jaws down far enough to allow the drill to reach, as I want to re-use the jaws for other parts.

The solution I came up with is to rotate the table so that the holes will be drilled between the jaws.  I'll reclamp the work for each set of holes, and use the height gauge to ensure the proper alignment.  I put pair of drill bits in the mounting holes, and then rotate the work so that the scribe of the gauge touches both;  once in position I tighten the chuck.

[NickG]

Good stuff, you're making good progress on this.

 

Nick

[kvom]

I have been working on the crankcase over the past few nights as time permitted.  I'm sure John will make this a lot differently as he has 6 to do, but here's my effort.

Since I had a precision 1" bore in the rotab chuck's soft jaws, I chucked a piece of 2" dia. aluminum rod in the lathe and turned down one end to 1" diameter.  On the other end I turned down 1" in length to the required diameter of 1.65".  I then bored the interior to .75" for a depth of 1".  I don't have a good boring setup for the lathe as yet, so further boring will be done on the mill.

Having previously dialed the rotab in to work on the cam housing, I was able to chuck the raw crankcase without further ado.  I then used an endmill to enlarge the bore to just over 1".



Since all of the holes to be drilled and tapped into the interior will cause burrs, I plan to bore to the final diameter of 1.25" after all the drilling is done.  This will remove the burrs, which might be difficult to access otherwise.

Next the mounting holes for the cam housing were drilled.



With the crankcase still in the chuck, I tilted the rotab to the vertical position.  Here's my method of squaring it to the mill table:



The mounts for the cylinders were machined next.  I used a .75" endmill that was actually only .725", so each flat needed two extra passes to attain the necessary width.



I spot drilled the five holes needed in each flat, then drilled and reamed the center hole to .500".



Next there were 20 mounting holes to be drilled with a #50 drill.  I have two of them, but somehow I have lost or misplaced one of them.  Since there are a lot of holes this size to be drilled I hope I find the spare.



Given the angles of the mounting holes, I just tapped them 2-56 using the guide I had previously made to tap on the mill or lathe.  It turned out to be very easy to tap aluminum with this setup.  The weight of the chuck was enough to start the tap cutting, and the taps cut very well.  Less than minute for each hole.



Trial fit of some screws:



Unchucked the crankcase and put it on the tapping stand to tap the 5 holes for mounting the cam housing.



The next step will be to bore the interior to final dimension.  I will hold off parting off the crankcase as long as possible.

Maybe bogs can send me one of his cylinders for a test fit.   

[chuck foster]

very nice build 

i just didn't realize how small this engine was  (well small to me) 

keep up the good work (and photos) i will be following this build very close 

chuck

[sbwhart]

Good job  Nicely done and shown, great work with the RT 

Have fun

Stew

[spuddevans]

Looking good there  I'm enjoying watching this come together.

Tim

[Gerhard Olivier]

My mind is boggled


Amazing 

Gerhard