The Return of No. 83, a Hot Air Engine

[vtsteam]

Wow! That's industrial level. Very nice!

[shipto]

Wow! That's industrial level. Very nice!

Thank you run by a 2.2kw motor on an inverter can make scrap in no time at all.

[vtsteam]

Having a smaller lathe is no disadvantage when making scrap! 

[vtsteam]

Well the graphite arrived early so we're back on track. 

 

[vtsteam]

I was looking forward to working with the graphite this afternoon, but in measuring the cylinder to get a diameter for turning, I just didn't like what I found with a telescopic gage. I still had a small amount of oval, and taper. Also I didn't like the cross hatch finish, which was more appropriate to an IC engine. I tried lapping again with the lap I'd made earlier. But it was too small, and I didn't have enough adjustment to compensate.

I don't think I like an aluminum lap. The engine had been originally built with a lead lap that I'd cast onto an arbor, then turned to size. It made a great lap, and the abrasive bedded well, and it gave a really nice finish but I don't like working with lead any more. So I decided to try a wood lap.

I had a turned piece of oak that I'd earlier used as an slip fit arbor for thinning one of the displacer cylinders. I turned this down to just undersize -- a tight slip fit really on the power cylinder, and then charged it with abrasive and kerosene (parrafin oil). Here's the lap as used, it took off lots of fine brass:

[vtsteam]

I spent about an hour lapping and measuring. The oak lap did a really great job of straightening out the bore, and it also left a really fine finish, very unlike the aluminum lap. I used the same valve grinding compound with each as an abrasive. I can't account for the difference, but I'm definitely convinced oak makes the best lap on a cylinder of this size and type, and very easy to make. I haven't tried any other hardwoods. I don't think softwood would work as well, but haven't tried it.

I could see a reflection in the finish:

 

[vtsteam]

With the cylinder lapped, it was on to the graphite piston. I had to cut the rod in half so it would fit in the spindle behind the chuck. I faced the end and started turning. I think I like the quality of this latest graphite even better than the original piston. It seems finer grained and more slippery. It cuts like butter, and it was only three cuts until I was at final dimension -- a hard slip fit for the cylinder. I'll lap the piston to a sliding fit later with a scrap of newspaper - that's all it takes.

 

[vtsteam]

Then it was time to set up the vertical milling attachment, with a milling spindle to cut a cross slot in the base of the piston, where the connecting rod will go. It was getting late, and I hadn't set up a DTI for depth of cut yet, so I decided to wait until tomorrow to complete the operation.

 

[vtsteam]

This morning I measured everything carefully, and decided to lengthen the piston by 80 thou and move the wrist pin further forward to compensate -- mainly to reduce the tendency slip stick in the cylinder. I slotted it .5" wide using the mill spindle in the milling attachment. That went really well.

 

[vtsteam]

Time to drill the wrist pin journal. I didn't have a way (yet) to do that with the milling spindle. It's not yet possible to mount the milling attachment 90 degrees to the slots in the carriage. To do that I'll have to slot the present milling attachment's mounting holes to fit the table spacing.

But, no problem, we'll do it in the drill press.

I centered a chucked drill bit in a vee-block, then clamped a scrap of aluminum as a fence against the block. As long as the vee-block is tight to the fence, it will present its center to the drill bit. But it can slide along that fence. Then I clamped the graphite blank in the vee block, using a bit of cardboard under the clamp to protect the blank.

I checked the height of the piston slot to the drill table at both sides to ensure it was horizontal. Then I marked the top of the piston where I wanted to drill the journal hole the proper distance back from the end.

Then it was just a matter of sliding the Vee block along the fence until the mark was centered under the drill bit, and the hole was drilled.

 

[vtsteam]

I returned the blank to the lathe and parted off the new graphite piston. I probably shouldn't have pushed that all the way, and sawed the last little bit off with a hacksaw blade because it broke a chunk off. Luckily that came out of the blank instead of the piston, so it was easily filed off.

Final step was to ream the journal to size and install the con rod.

 

[vtsteam]

Lapped the piston with paper, reassembled the engine, and things are looking better! No 83 settled in at about 1300 RPM and hit a high of 1440 RPM. That's great as far as repairing the cylinder and fitting a new piston. But I'd actually hoped for better, since before the cylinder problem, I'd made some big changes to the displacer cylinder mounting, and added a copper end.

Well we're definitely back in business with a good baseline again. I'm going to think about the displacer cylinder change. I think definitely there is more cooling now, and it extends further along the cylinder with the heavy aluminum flange.

I think the problem may be that I've got too much cooling now compared to the amount of heat available, which is relatively low. I don't yet want to add more heat though I will in the future. I think my next change will be to make a displacer cylinder with a thinner flange thickness, and that will be made out of stainless or rather than aluminum. I'm not sure about the copper cap. Probably revert to stainless again just to compare with my earlier cylinder #2 which was that best performing one.

[sorveltaja]

I built a Ringbom stirling engine in 2008, and it has kept good compression through the years.

Anyways, I can't remember the exact steps I used to hone the cylinder, but on the last stage, I used automotive polishing paste with - perhaps brass or aluminum rod (that's what I had at that time), against brass cylinder.

It was rather tedious process, as I did it manually. I didn't use split rod to lap - it was solid. It used to get stuck in the cylinder, when I went too far, while turning it. Good knock on the working end was enough to release it. Finally, it moved through the cylinder.

Then came the final stage - to fit graphite piston to the cylinder. I used the same procedure as with lapping. Solid graphite rod (part of it was then used as a piston) against lapped brass cylinder (as graphite wears itself relatively easily through the friction, when compared to metals).

That's just one possible way to minimize power piston/cylinder friction.

[vtsteam]

Hey sorveltaja!    Great to hear from you! How is your guitar? That was such an interesting project!

Also interesting to hear about your lapping experiences. For me the dowel worked surprisingly well, I really hadn't expected it to do better than the aluminum lap.

I'm thinking that "aluminum" is actually a too general term for me to make judgements about-- aluminum varies a lot in hardness, and so maybe I chose a harder variety out of the scrap box. this time

Pure aluminum is dead soft, and I bet that makes a better lap. I think it's important that the lap get embedded with the abrasive, rather than it moving around between the two surfaces too much. The lead lap I used years ago worked very well, and I know the abrasive embedded deeply there. Same thing for the wood lap this time.

I have a pretty good honing rig for larger cylinders in a larger engine, which I'm hoping to start at some point in the future. It will still need lapping, too as a hot air engine, but it should have a uniform bore after honing, so only will need to be lapped for a good finish. That's my hope anyway.

[vtsteam]

I ran the engine again tonight, and it is consistent at around 1300 RPM. I hear a rumbling sound and some slow beat volume changes at full speed. I think it might be the ball bearings.

It stands to reason that if the rest of the engine had rusted in a damp garage for 8 years that maybe the bearings might have some issues, too. Well, those are easily replaced, and I now have ABEC 9 bearings on hand. The ones in place were ABEC 7.

Another possible contributor is the main shaft. It's 5/16 (.3125") but the bearing takes an 8mm shaft, so there's a little play. I have some 8mm stainless steel now, so I'll make the switch when I change the bearings.

A third issue I noticed: I washed out the power cylinder before reassembling again tonight, and when giving the flywheel a spin, I saw a little moisture from not having dried it completely squeeze out from the edge of the cylinder where it seats in the base plate. This means a leak. Easily sealed with a little bearing compound -- Loctite 620, but good to know. It's probably always leaked, so that should help performance when fixed.

Fun stuff! 

[vtsteam]

Replaced ball bearings and crankshaft, and sealed the leak. I also found when replacing the bearings that the hub of the flywheel was contacting the bearing hub -- there was a high spot where a set screw had been tapped into the flywheel hub. Threading had slightly expanded the steel in the area, which was impossible to see, but I could faintly hear a contact sound when the new bearings and shaft were in place and I spun the flywheel by hand close to my ear. I think that was the sound I heard before when run at full speed, thinking it was bearing noise.

I put some marker on the flywheel hub, spun it some more. Taking it apart again, I was able to see the high spot where it had removed the marker. I filed the raised bit down with a small triangular file, and the noise. was gone.

[sorveltaja]

Hey sorveltaja!    Great to hear from you! How is your guitar? That was such an interesting project!

Yeah, that guitar is ok, except its tremolo turned out to have a flaw. Tension of the strings has caused string retainer to rotate(something I didn't expect), and therefore it isn't possible to tune it properly.

[vtsteam]

That's too bad, sorveltaja, but I bet, knowing your capabilities that you can change that, if you want to. I remember all of the amazing changes you made. Plus I was really impressed with the rust finish -- I know that sounds funny, considering all of the mechanics, but I thought that was really cool!

re. No. 83, I started on a fourth displacer cylinder today.

 

It features a thinner steel flange, compared to the last version's thick aluminum, and a reversion to a stainless steel end cap instead of a copper one. I wanted to try go back to the performance of my second cylinder, if possible, by removing some of the added variables.

So far, I've made 4 displacer cylinders, lined up here to show the differences. The first was just stainless steel tubing. Top RPM was around 800. The second had a thinned down wall thickness, and average speed was around 1400 RPM it did hit 1600 rpm on one run. The third cylinder had a changed mounting method. Instead of passing through the aluminum heat sink, it had a flange and mounted outside of it. It ran about 1300 RPM. An improvement in bearings and a sealed leak and removal of a frictional interference produced a max speed of 1440 RPM.

The 4th is a reversion to the second cylinder but with a flange mount.

 

[vtsteam]

Progress had stopped for the last 4 days as I waited while my electrical heating coil project dried out. This coil will allow me to get consistent heat input for testing changes made to No. 83. Sterno does not provide consistent heat as the can contents burn down.

I'm really happy to say the new mini electrical heating furnace works really well with the engine. And with less power than I thought it might take.

The engine starts and reaches a steady state at about 960 RPM at 80 watts @ 10.8V. That was on the porch outside my house at an initial test at a 53 degree (F) air temperature, and some fairly strong breezes.

I think it would probably run faster indoors, but I had to do this initial test outdoors to burn off the fumes from a small amount of tape and release agent used in originally making the coil.

Here's the engine running in a test:

 

[vtsteam]

I don't really have a good data baseline yet for a few  reasons, but definitely will be able to do so with the electrical heater. I want to do tests indoors, and I did notice the RPM kept increasing slowly over time, as the wattage actually decreased -- both probably due to the last bit of moisture gradually drying out of the heater, the low temperature outdoors and wind, and the plaster mass heating up slowly.

Nevertheless, I did achieve 1050 RPM at about 100 watts and that speed was increasing slowly, but by then I needed to stop and bring everything inside, because it started to rain.

I'm very, very happy that this heater worked out first try with the amount of wire and power estimated. It was a long wait while it dried. I can now really compare modifications in the engine.

The engine is running fairly loud lately -- I think it's worn big end bearings knocking now. I've got definite plans for improvement!

 

[vtsteam]

Well it's kind of disappointing to see the very small amount of interest most visitors to the site pay to people posting here. Other than a welcome few comments from old friends, it's as if we were alone.

Yet I'm seeing 2000 views a day on the progress of this little engine. and in the forum statistics on the front page right now, 500 visitors, and only three members signed in.

This is a rather sad state of affairs, and a forum like this will not continue unless others join, sign in, comment and, one hopes, even add interesting projects of their own. Certainly, now, I'm asking myself just who is it I want to communicate projects to? If it's just lurkers, bots, spiders and AI content reapers, then no thanks, I'd rather just work on my own.

Nope I'd rather talk with real people. Are there any out there in that group of 2000 views, or those 500 visitors today? Say something. Say hello. Say anything! Let us all who post here know that we have a reason to do so. Organic Intelligence. 

[nickle]

I’ve been on this forum for a very long time. I remember when forums were much more active and there was a lot of discussion. I don’t post much because I am not very active in my shed while life is busy. I also think YouTube has kind of trained us to be passive consumers of content rather than participants in discussions.

This is a very interesting project both technically and the back story. I’m really enjoying watching you problem solve and tweak this little engine. It gives me something to aspire to once I have a bit more space in my life again. For the time being though I am enjoying your build.

I’m checking in regularly.

Nick

[vtsteam]

Oh, that's great Nick! That gives me a real boost. Thank you!    I hope you will build an engine (if you want) that would be great to hear about. Very happy to hear this thread means something to you.

I hope people do want to communicate on fora again, I also see the lapse of the last few years, and yes, probably videos have a lot to do with it. But there is no substitute for group conversation. We're humans (mostly), we talk, and we want to be heard.

I know I'm really not writing this stuff for show, I'm here to communicate, and hear from others what they are doing and are interested in. And I really don't care at what level that is, could be expanding an ant farm. Or fixing a trumpet. Great! Bring it on!

Okay so, I've been working on a side project temporarily, trying to rehabilitate some small cnc equipment I built a long time ago, and which has been attacked by time rust and mice (TRM) in my further shed. Definite hazard around here. And trying to get my head around what to do about the disappearance of the parallel port as a viable means of CNC communication, since both of the now dead old laptops I used for that did so with the contraptions I once built.

That's beside the point of this thread, and so tonight, not to seem to have become sidetracked completely, I ran No.83 on electric power for two trials, one with displacer cylinder #3, and one with displacer cylinder #4, both at 100 watts heat exactly.

The #3 cylinder (thicker aluminum base) was slower to heat and run up to speed, but ultimately edged out the #4 cylinder at top speed by about only 10 more RPM. In  both cases the rod ends were annoyingly noisy, and I really want to replace the whole set ASAP with what I hope will be a big improvement all around. I think the #3 cylinder possibly peaked at RPM because of this mechanical mayhem, rather than actually reaching potential.

BTW the CNC machines are, I hope, going to give me a lost foam casting pattern for the heat exchangers cum base plates (hollow and water filled this time) for a high speed vertical shaft Rider variation on this engine -- using otherwise the same specs sizes and construction as No 83, for comparison testing.

It will be called No 84, of course.

Well okay one more thing -- I'm trying to rehabilitate my small melting furnace, too, after it got buried in snow this winter and spoiled the lining. So, that's four projects I'm doing at the same time right now. But that's a happy state of affairs for me...disorganization is life.

[tom osselton]

From what I see it really started declining after the like button failure people need feedback even just to see if someone out there is watching.

[shipto]

I agree the "like" button was good to show interest even if you had nothing to add to the conversation.