Philip Duclos Fire Eater

[CFLBob]

I notice you are using a spotting drill to support the end of the tap handle.
It may be better instead to use a home-made centre from a bit of bar. Just put a length of 3/8" or whatever you have in the lathe chuck, turn the topslide round to 30 deg. and make your centre. Cut off and hold this in the collet or drill chuck on the mill; doesn't need to be hardened. It'll help to save the point of your drill.

Dave
The Emerald Isle

Good point, if you'll pardon the pun.

This is the first time I've ever used this wrench in a setup like this, and I used a small spotting drill to make the divot in the wrench's head. I just grabbed the first thing in my "bit drawer" that looked like it would fit.

 

[CFLBob]

Well, this is turning into a build thread, although that wasn't my intent at the start. The latest work was making the crankshaft support bracket.

For a couple of reasons, I didn't like the Duclos' approach and this seemed like a good part to redesign. Why didn't I like it? Two main reasons. First: he lets the flywheel shaft, 1/4" steel rod, just rotate on aluminum and one thing I've learned about these engines is it's worth it to remove as much friction as you can. I wanted to use ball bearings, which causes the uprights of the part to be bigger than Duclos used. Second: he makes this part from a solid 2-1/2x1x3/4 chuck of aluminum and turns most of it (about 65-70% by my calculations) into waste chips. So I decided to make it out of three pieces of 1/4" aluminum (all you're left with in his design). Two uprights and a bottom piece to hold them in place. Some time with CAD resulted in this drawing:

My CAD model on the left and Duclos' drawing on the right. You can see where the ball bearings sit, in the counterbored holes at the left. I got a set 1/4" shaft and 3/8" OD overall from Boca Bearings.

I then created a drawing with two of these brackets side by side, spaced far enough apart to put a 3/8" EM between them. A little time spent with both Rhino and Deskproto (my CAM SW) and porting it all to the mill ended up with this:

I won't bore you with the side story that popped up here, but I had to re-tram my Z-column. The spindle was perpendicular to the table, but the Z-column wasn't. Which made me do both.

So while these are smallish pieces, the center (deep blue in the 3D model) was 0.75 wide by 1.00" long, and a bit tedious to work on, it all went as designed and went together the way I planned it.

Would it help if I said the back looks just like this side?

30-some years ago, an older engineer told me that "engineering is the art of compromise". There are rarely ever perfect choices that everyone agrees are the best way to do anything and this design is like that. The three piece bracket does what I wanted. All the parts worked out just as modeled. The drawback is that it's three parts and they need to be aligned properly during assembly. When I first put it together, the flywheel spun, but also made the characteristic scuffing sound of something rubbing. The parts aren't perfectly aligned, but I improved that by loosening the four screws on the bottom and putting a shim between each side bracket and the flywheel hub (I used one of my old business cards - still useful after retirement!). When I retightened the screws and pulled out the shims, the scuffing was gone. The flywheel still wobbles which argues that the two sides aren't exactly the same height.

I think I need to come up with an alignment fixture to hold the base and the sides while I indicate the shaft to make sure it's level and not offset side to side.

If I give the flywheel a good flick, it spins for about 45 seconds. Is that a good number? I think it can be made better, but how long should it spin?

My next project is going to be the piston, and I have a small piece of 1" diameter graphite (1.030") I was thinking of using instead of the CRS bar I bought (longish story). Duclos' drawings call for hollowing out the piston until it's left with 1/32" thick walls. Is graphite strong enough for that? I can just envision it shattering as I hollow it out in the lathe chuck.

 

[Cogsy]

It looks like you might still have the shields in those bearings? If so, pulling them out and washing out the grease, then replacing with a very small amount of light oil will likely significantly increase your flywheel spin time. On Jan Ridders internal valve design he suggests a 2 minute free spinning flywheel is what to aim for. Of course this is going to be different for this design, but less friction is always better for flamelickers so I'd remove the grease.

The build is looking good btw!

 

[CFLBob]

It looks like you might still have the shields in those bearings? If so, pulling them out and washing out the grease, then replacing with a very small amount of light oil will likely significantly increase your flywheel spin time. On Jan Ridders internal valve design he suggests a 2 minute free spinning flywheel is what to aim for. Of course this is going to be different for this design, but less friction is always better for flamelickers so I'd remove the grease.

The build is looking good btw!

Thanks for the advice. The ball bearings are straight from the store and they are internally greased. I have no idea how to open them to clean them out but maybe some reading will lead me there.

 

[Cogsy]

Shielded bearings are reasonably easy to open. The metal shields do not contact the inner race so there is a slight gap there to put a probe in and simply pop the shield out. The smaller the bearing the more fiddly they are to do unfortunately.

 

[CFLBob]

he smaller the bearing the more fiddly they are to do unfortunately.

LOL. These things are kind of in the class that it's a marvel that they can be made industrially and sold for a few bucks. They shaft ID is 0.25 and the OD is .375, leaving about .062 on a side for the entire thing by .125 thick. I downloaded the 3D CAD file from Boca Bearings to try to look at the shields, and they're about 1/32 across. Fiddly is an understatement.

Boca Bearings has an app note about this. They basically say soak the bearings overnight in a solvent, let them dry (overnight is probably good again), then soak them in light oil for a few hours. I have plenty of time until I need to use these, so this is an easy experiment. A few minutes a day of actually doing something, and mostly waiting around, is easy.

Over the years I've found I'm pretty good at doing nothing.

 

[el gringo]

Boca Bearings has an app note about this. They basically say soak the bearings overnight in a solvent, let them dry (overnight is probably good again), then soak them in light oil for a few hours. I have plenty of time until I need to use these, so this is an easy experiment. A few minutes a day of actually doing something, and mostly waiting around, is easy.

Over the years I've found I'm pretty good at doing nothing.[/QUOTE]

Hi guys,
I have built three engines of this type including the Duclos 'fire eater'. Two have shielded bearings and the other has bushings made from bronze. All run very nicely. I have never bothered to remove shields or factory lube from the bearings.
Thirty seconds spin time would be a record for me.
I have used bearings from Boca for years and always very satisfied with their products and support.
Ray M

 

[CFLBob]

Boca Bearings has an app note about this. They basically say soak the bearings overnight in a solvent, let them dry (overnight is probably good again), then soak them in light oil for a few hours. I have plenty of time until I need to use these, so this is an easy experiment. A few minutes a day of actually doing something, and mostly waiting around, is easy.

Over the years I've found I'm pretty good at doing nothing.

Hi guys,
I have built three engines of this type including the Duclos 'fire eater'. Two have shielded bearings and the other has bushings made from bronze. All run very nicely. I have never bothered to remove shields or factory lube from the bearings.
Thirty seconds spin time would be a record for me.
I have used bearings from Boca for years and always very satisfied with their products and support.
Ray M[/QUOTE]

Ray, you have given me hope.

I just finished the connecting rod. One tiny, minor, glaringly huge error. The small end isn't 1/4" square on a side. I trimmed one side to the connecting bar. I doubt it will matter.

 

[Cogsy]

This design does run a lot better than the one I built, for that to run I needed everything as frictionless as I could get it. One day I might have a go at one of these, if I ever get any of my current projects finished...

 

[CFLBob]

My connecting rod. Like I say, one tiny, minor, glaringly huge mistake. Except I don't think it matters.

I think the wrist pin yoke is next. Since I'm cutting the piston to the fit the cylinder, I want to be able to use the connecting rod to push the piston all the way down the cylinder, and (more importantly), get it out of there.

 

[CFLBob]

On the topic of Graphite pistons, someone posted here that they made a graphite piston. Since a PM didn't return an answer, I thought I'd try to ask here before going to another forum and asking everyone else.

If you made a graphite piston, did you follow the drawing dimensions exactly? The wall thickness concerns me because I have no experience with graphite. In the drawing, the walls are 1/32 thick and the idea of cutting away almost 15/16" from a piece of graphite scares me. Is graphite strong enough for this?

Having never seen one of these engines in person, I don't know if there's a lot of room between the connecting rod and the edge of the piston when it's at the back of its travel, so maybe there's not enough room to let the walls be thicker.

 

[Brian Rupnow]

Make the walls 1/16" thick, otherwise it becomes just too fragile.--Brian rupnow

 

[CFLBob]

Thanks, Brian. In my mind, I could the see the graphite shattering when I hollowed out too much.

 

[el gringo]

Thanks, Brian. In my mind, I could the see the graphite shattering when I hollowed out too much.

Another way to go... I have simply slotted the graphite as a regular piston with a wrist pin, you need only to remove enough material for the conrod to function .
The only part you need to worry about thinning for conrod bottom skirt clearance is at the very bottom of the skirt, probably can get away with an internal skirt champher to clear if the piston is of proper length and rod cross section is correct.
Ray M

 

[Cogsy]

On the topic of Graphite pistons, someone posted here that they made a graphite piston. Since a PM didn't return an answer, I thought I'd try to ask here before going to another forum and asking everyone else.

I don't have the Jan Ridders plans in front of me so I'm not sure what thickness is called out, but I would have followed them exactly (it was my first engine build) and had no issues. Graphite, at least the stuff I used, holds up far better than I imagined is would.

About the PM - it wasn't me you sent it to was it? I had something weird happen with a PM yesterday and dismissed it as just a glitch but if others are having issues I'd like to know please. I certainly never received one from you.

 

[CFLBob]

I don't have the Jan Ridders plans in front of me so I'm not sure what thickness is called out, but I would have followed them exactly (it was my first engine build) and had no issues. Graphite, at least the stuff I used, holds up far better than I imagined is would.

About the PM - it wasn't me you sent it to was it? I had something weird happen with a PM yesterday and dismissed it as just a glitch but if others are having issues I'd like to know please. I certainly never received one from you.

No, it wasn't you.

As long as I'm here, there's a side story. I have a piece of graphite of about the right size, but I don't know anything about it.

10 year ago, when I was just getting started, my wife and I went to NAMES. I found out about Jerry Howell's engine plans and bought his kit for a flame eater. All of this got put aside and moved from room to room a couple of times. It included full drawings, two sets of ball bearings (one set would have been fine for the bracket I redesigned) and a piece of graphite pretty much perfectly sized to make the piston. No markings, and no details on the graphite.

I stumbled across the box with these parts the day after I ordered my ball bearings. Thankfully, I hadn't ordered a big graphite rod.

 

[CFLBob]

It has been a slow week due to some problems threading on my Little Machine Shop LMS 3540 lathe. The most involved part I made was the wrist pin yoke, which is a small part (5/16 OD) with a 3/16 long stud threaded 6-32 on the end. For about three full turns.

Briefly, I've only done a couple of threading jobs on this lathe, 1/4-28, and some practice 1/4-20 pieces. When I went to thread the part, I changed over to the gears for 32TPI and the gears mechanically interfered. I couldn't see how I could thread anything finer than 28 TPI on the lathe, but they say it can do 7 to 80 TPI. LMS was of no help but staring at the setup until my eyes glazed over eventually got me an answer that worked. Whether it's right or not, I don't know. But maybe if it works it's right.

I eventually found the work around, which is to reposition the swinging lever that holds the two intermediate gears (called B & C) outside the driving gear. It makes the lathe less safe to use - there's a spinning gear about four inches from where you hold your hand, but an impromptu cover makes it less likely to grab you. You'd have to work at getting something into the gear.

Once that was done, I did some practice pieces to make sure I could thread them safely a few times in a row and then my work piece.

This week's contribution to the parts pile:

The wrist pin yoke with the wrist pin (a 1/8" piece of steel with a 3/32 hole along its axis), a custom collar and washer (the washer is made from the head of a nail), the flywheel shaft, and the practice pieces I threaded three turns of 6-32 on.

Next it's on to the piston, which will be graphite.

 

[Cogsy]

It's been a while since I did any threading on a mini lathe but that doesn't look right to me. The gears can be 'double stacked' on each point if I remember correctly, and the bottom pivot can be moved around quite a bit on it's bracket to allow the different gear sizes to mesh correctly without going outside the gear cover. I'm guessing there's either an adjustment you've missed or your gear train should be 'staggered'.

 

[CFLBob]

It's been a while since I did any threading on a mini lathe but that doesn't look right to me. The gears can be 'double stacked' on each point if I remember correctly, and the bottom pivot can be moved around quite a bit on it's bracket to allow the different gear sizes to mesh correctly without going outside the gear cover. I'm guessing there's either an adjustment you've missed or your gear train should be 'staggered'.

This is a long story that I didn't want to spend lots of time covering. I made a video of the problem where I demonstrate the interference issues to send to a friend who's much more experienced than I am for advise.

Don't mistake this as confidence that I'm sure I'm right. This drawing should help with the names for everything:

Gears B/C (same hub) and D are mounted along a swinging, adjustable length arm. The hub D is mounted on is the leadscrew for the carriage and so its hub is the pivot that the arm rotates around.

If you compare this layout to the one above, it looks like the gears on C and D have been swapped; this looks like the gearing for normal running, which is the power feed. What was happening was that with any gear on D bigger than the size used for 28 TPI, the B/C gears get pushed farther out the lever and the C gear (being bigger) hits the bushing on the shaft of the driving gear - which is unlabeled, to the left of A.

Gears A-C are the same for threads from 28 to 40 TPI. Only D changes size.

I went from 32 TPI to 40 by changing gear D from 80 to 100 tooth and that larger diameter pushed B/C so far out that they hit that enclosure at upper left. I don't think B got within two inches of A, which is supposed to drive it. I took it all apart and put the lever to the right of that enclosure, where it would have to be, and then couldn't get the parts in there.

I searched everywhere I could think of for help with threading on this lathe. The thing that stood out was that the manufacturer, Sieg said it would cut 8-24 TPI. Axminster, the big tool dealer in the UK said the same thing. The only people who said this could do up to 80 TPI is LMS. I asked them for help, they said something about that lever being movable (which anyone who has changed gears would know) and when I sent them the picture of the lever arm out there to the right of the enclosure asking "is this the way it's supposed to be?", they never answered.

If I'm wrong and there's a less awkward way to put this together, I'd love to know, but I simply couldn't make it fit.

 

[Cogsy]

Interesting - I certainly see your issue. I only have experience with the SC3 and it seems the SC4 is clearly different (though I didn't realise you had an SC4 from the pictures). My SC3 is a metric version so again it's different, but I've cut 0.5mm threads on it which is equivalent to almost 51 TPI, so I can confirm the smaller lathes can handle finer threads. Of course that doesn't help you at all, but I like your work around, way to think outside the box!