Class team build project - Duclos Fire Eater

[kvom]

For my final machining class at the local votech school, a team of 6 people will attempt to build 20+ examples of the Duclos "Fire Eater" Stirling engine. The goal is for each of us to end up with an engine, and then to have lots of extras to give to local highschools as recuitment aids for the machining courses.

Here's a HMEM thread showing a running engine: http://www.homemodelenginemachinist.com/index.php?topic=3568.0

We are divided into three groups of two and each group will be responsible for different parts as we progress. The parts will be made using the Haas CNC lathes and mills, with CAD/CAM done on MasterCam. For the most part we will try to make the parts as drawn. One learning objective of the course is to be able to design the necessary fixturing to produce multiple parts efficiently. Another is to develop teamwork so that mating parts do go together properly. I seem to be the only one in the class who has ever built a model engine, so I'm aware that getting separate parts to work flawlessly right off the machines can be quite a challenge. So the fact that we will be using precision CNC gear will be interesting to me to see how well it works. My teammate is another old retired guy like me who hasn't taken any of the CNC classes, so I will be doing all the Cad/CAM work for our parts.

Our first part is the the cylinder pedestal. While I spent the first class period on the computer, he cut 24 pieces of aluminum stock 3/4 x 2 x 3 and deburred them. To make the pedestal I needed 3 drawings. The first machines the profile on one side leaving 1/16" for holding in the milling vise. The second drawing is for cutting soft jaws for the vise to hold the parts from step one, and the third drawing machines the reverse side of the pedestal plus the holes for the cylinder and timing shaft. It will still need tapped holes in the base for mounting, but I'm not sure yet how we'll do that. Duclos specifies a press fit between the pedestal and the cylinder, but we are thinking about doing a shrink fit.

Our class meets Monday and Wednesday evening from 5:30 to 9:30. I hope to get the first side of the pedestal machined on Wednesday, and produce some pictures as well. Another team is making the cylinders on the lathe, so we might be able to see some other parts too. Team three is starting with the bearing bracket.

 

[mklotz]

examples of the Duclos "Fire Eater" Stirling engine

Dons pedantic hat...

It's not a Stirling engine. Properly, it's an atmospheric engine since it's the atmospheric overpressure that supplies the power stroke.

 

[Stan]

A fire eater requires a very free running piston in the cylinder so I would be very careful doing a shrink fit of the thin cylinder wall into the pedestal.

 

[kvom]

A fire eater requires a very free running piston in the cylinder so I would be very careful doing a shrink fit of the thin cylinder wall into the pedestal.

I did mention this as a possible issue, so we'll see on the first pair. The pedestal clamps onto the front 3/8", so it's possible that some polishing will restore the opening. We didn't think that the part of the bore covering the piston stroke would be affected. Our class instructor is a very good CNC machinist, and it's his advice to shrink fit, so initially the bore in the pedestal will be made .003" undersize.

The guys making the cylinder though .0005 undersize with some loctite would be fine.

 

[kustomkb]

Mine was made so the cylinder was a tight sliding fit which could be assembled by hand. Loc-tite would be good.

I honed the cylinder for a nice sliding fit and ran it in with graphite powder put in through the intake port.

If my engine was put away, hot, with the piston at top dead centre, when it cooled the piston(aluminum nickel bronze) would seize. -something to watch out for.

I had a pretty hard time getting 'er going at first. but now it runs quite well.

All the best with your team build!

 

[b.lindsey]

I have to agree with kustomKB. A light press is all that's needed and a .003" shrink fit could easily distort the bore. On mine at least the bottom end of the piston does extend about 1/8" beyong the bore when at bottom dead center, so coming back and truing up the bore to the tolerance needed could be problematic.

Bill

 

[kvom]

Thanks for the feedback. The cylinder team and I have decided to go for the -.0005 press fit with loctite.

Yesterday evening was spent setting up the machine. I loaded up the tooling in the changer for the 3 separate steps plus milling the soft jaws, then tweaking the settings by milling the first two blanks for the first face. By the time we had to leave I thought we would be ready to mill the remaining 22 blanks today. However, the instructor decided we needed to have rounded edges on all 4 sides of the pedestal's foot, so it took another hour or two adjusting the g-code and machine to use a 1/16" corner rounding bit.

Here's the result:

Once tested out and using a moderate feed rate, it takes ~10 minutes a piece for this face. I managed to do 10 pieces before I had to leave for the day.

Once this is finished, phase II will be to drill and tap the two blind mounting holes on the bottom 8-32. Rigid tapping is pretty amazing when it works - 600 RPM, and too fast to see with the naked eye. of course, if we break the tap it will be back to hand tapping. Next class is Monday night.

 

[kvom]

After 4 milling operations, here's the first "complete" pedestal.

This one is actually a "setup" piece. I still need to fine-tune some settings before running "production" on the remaining 20. The hole is still too small for a sliding fit by at least .001". We will probably just set the cylinder nose diameter rather than tweaking the mill, as that is easier.

 

[njl]

Nice looking piece kvom, good see how it was crafted from the blank.

Nick

 

[Stan]

I know nothing about CNC setup so is that the reason for having exposed mounting screws in the pedestal?

 

[kvom]

The depth setting for the mounting holes needed to be corrected, so that's why this piece is a "setup" piece. The good pieces are not drilled through the top of the base. So given that blemish, this piece was used to adjust the last milling pass.

In all, we cut 28 pieces of stock. Three were trashed because a parallel moved in the vise. One was trashed when an end mill broke (used a non-centercutting endmill by mistake). Four had blemishes and are setup pieces for the final run, including this one. So we have 20 pieces that have the first two passes done, and hopefully will all survive pass 3.

 

[kvom]

After finishing 20 pedestals, my next part was the crank arm. The CAD/CAM was pretty easy as I just had to cut a few round profile operations followed by the drilling/reaming/tapping. The mill needed 11 tools in the magazine, as each hole needed two operations plus center drilling. Setting up the machine actually took quite a while. Using very conservative speeds&feeds each part cycle took 4 minutes.

Yesterday I finished up the CNC by doing a simple program to drill and tap the grubscrew hole in the side.

I then started on the cam fork, which will require 4 separate operations. I got only part of the first op completed, so more to do tomorrow. The other teams are doing well too. The bearing supports and crankshafts are done as well as the pushrods. The cylinders are turned and bored, waiting a free mill to complete.

 

[stevehuckss396]

For my final machining class at the local votech school, a team of 6 people will attempt to build 20+ examples of the Duclos "Fire Eater" Stirling engine. The goal is for each of us to end up with an engine.

20 seems a little short. There is 5648 of us and then you need a few for yourselfs. Sounds like you need to turn up the speeds and feeds to cut that cycle time down.

Haahahahhaha!!!!!!!

Nice work BTW, It's fun when everything goes good.

 

[kvom]

I have to get only one to work. The rest are other ppls problems. ;D

 

[kvom]

A bit of "homework" this afternoon. Turn 20+ brass counterweights from 3/8" brass rod; these will be pressed into the ends of the crank arms. Given that these are 1/4" in length, I'm glad to have a collet chuck on the lathe.

 

[kvom]

Although I haven't posted for a while, we've been busy working on the parts for the engine. Last night we had most of what was needed to start assembling a few. Here's some parts:

A plethora of pedestals:

A bunch of bearings:

A batch of bases:

A collection of cylinders:

And a flock of flywheels:

The burners are fabricated from a brass doorknob, with the spout being s .45 caliber automag shell casing and the wick being from a Tiki torch.

To attach the cylinder to the pedestal, we heated the pedestal ring with a torch, then slid the cylinder lip into the expanded opening while using a square to ensure that the milled face is perpendicular to the base.

Starting the assembly:

And the first trial fit with all parts in place:

At the end of the class, I still didn't have the valve and valve spring ready, so took it home as-is. I need to remake the little screw that holds the cam follower as well as finish the valve. The instructor has gotten a couple of these running, as he took parts home last week. He has made a video that I'll post on youtube once he sends it to me.

I'm hoping to get mine to run by Monday's class. Otherwise it will have to wait until I get back from my trip in November.

CNC parts I made:

Pedestal
Crank arm
Cam fork
Valve
Milling ops on the cylinder

This has been an interesting project.

 

[njl]

Thats very nice work Kvom, all the bits look very nicely made. Hope you get it running before your trip, my money is you beating your deadline. Thm:

Nick

 

[kvom]

Here's a short video of the first engine that ran (not mine):

[ame]http://www.youtube.com/watch?v=Nllxscz9zFw[/ame]

 

[ChooChooMike]

A very nice soon-to-be flock of fireeaters (continuing with your theme !!)

Great job

 

[GailInNM]

Or maybe a gaggle of gulpers. ;D

Looking Gr-r-r-r--REAT. And you can all say that with a streak of accomplishment.

Gail in NM