Friday , July 25 2014
Recent Articles
Home > Engines > Building a Flame Eater (Vacuum) Engine – Part 1: The Cylinder

Building a Flame Eater (Vacuum) Engine – Part 1: The Cylinder

I’ve been working on Jan Ridders’ “Flame Eater” engine for over a year now. When I first started I didn’t have a mill (nor a plan as to how I would complete the project without a mill) but I started on the engine anyway. Eventually I turned just about every piece that could be turned and then hit a wall. Without a mill I could go no further. So the project got shelved for months while I searched for, purchased, and restored and Atlas MFC mill. A few weeks ago I finished my mill restoration and it was time to get back to my little Flame Eater.

Many of the pieces for this engine are easy to machine and require no explanation, but some are a bit more difficult. My intention here is to describe the more complicated pieces and the machining steps I used to complete them. My methods aren’t the only way to machine the engine, but if they worked for me, they should work for you as well. I’ll be documenting this project in several parts:

Part 1: The Cylinder
Part 2: The Piston, Valve, and Connecting Rod
Part 3: The Flywheel
Part 4: The Ball Bearing Support and Spirit Burner

So let’s get on with it.

Part 1: The Cylinder

Jan recommends using “pearlitic cast iron” or stainless steel for the cylinder, piston, and valve. I’d recommend trying to get cast iron if you’ve never worked with it before. It’s different than steel and very messy to turn, but I think that it’s properties lend itself well to the cylinder design. For one thing making the cylinder, piston, and valve from cast iron helps with lubrication because cast iron rubbing against cast iron is somewhat self-lubricating. With a flame eater engine you don’t want to use oil to lubricate the cylinder because it will eventually burn away and gum up the cylinder. Another benefit of cast iron is the fact that it was easy to cut the fins with a parting tool. The cast iron chips were short and broke away easily unlike steel, which often produces long springy chips that bind in between the two halves of the piece being parted off (or in this case between the fins).  In my opinion it would have been a little more difficult and taken more time to cut the fins in a steel cylinder. Whichever you choose, be sure to make all three parts out of the same material. If you make the valve and piston out of steel and the cylinder out of cast iron (or vice versa) they won’t expand and contract at the same rate leading to either a sloppy fit, or too tight a fit once the engine heats up during use.

Step 1a: Turn OD

Center drill the cylinder and support with a live center (or lubricated dead center). Note: other than lubricating the dead center, don’t use lubricant when cutting cast iron – it should be turned dry. Turn part to the desired OD but leave .001 extra to polish to size. I used 400, 800, 1200, and finally 2000 grit sand paper to get literally a mirror finish on the cylinder. Note: the picture below shows the part unsupported and with the cylinder hole already bored, you’ll be boring the hole next.

Step 1b: Drill/Bore/Ream ID

Remove the center drill and drill the hole to within .010 of final size, then bore to .001 or .002 of final size, and ream to final size (ream if you’ve got a reamer, if not you can always lap the cylinder instead). I drilled the hole in stages, about 1″ deep at a time. I drilled a .250 hole an inch deep, then I swapped the .250 for a .500 drill and pecked until that drill bottomed out at about an inch deep. Then I’d go back to the .250 drill and make the hole deeper by another inch or so. Drilling the hole in steps like this is safer than trying to drill the hole to depth in one plunge. If I tried to drill 3″ deep with a .250 drill bit the bit would likely break. But it’s fine in one inch increments. Remember to “peck” with the drill – plunge .050 or so and then retract the drill to clear away chips – then plunge another .050. The larger the bit the deeper you can plunge without clearing chips, but watch carefully. If chips stop coming out of the hole on their own they are building up inside and will eventually bind the drill causing it to spin in the chuck or break. Avoid that.

Once you’ve got the cylinder drilled out to within .010 of the final size for the entire depth of the cylinder, move on to boring. Boring will ensure that the hole is concentric and uniform in diameter across the entire depth. Take very light cuts with a quality boring bar. If you’ve got a reamer of the proper size use it. But don’t ream more than .001 or .002 and try to ream the hole all at once in one continuous plunge while turing the work at a slow speed. Don’t ream the hole multiple times or the hole will likely turn out oversized. Don’t ever run the lathe in reverse with a reamer in place – you’ll dull or chip the flutes on the reamer.

If you don’t have a reamer or your hole ends up slightly under or over the desired size, don’t worry. You’ll be turning the piston and valve to fit.

Step 2: Cut the Fins

This step really spooked me. So much so that I set the part aside for weeks avoiding this step. Don’t let it scare you -especially if you’re using cast iron. It’s not as bad as parting off because you’re not plunging the parting tool all the way to the center of the part. I turned the part at a fairly fast RPM, 500 maybe? I can’t remember for sure. And I plunged the parting tool very slowly into the part to the desired depth. I made sure that the parting tool didn’t stick out any further than it had to for clearance – thus avoiding unnecessary overhang. I used a dial indicator with a mag base to accurately space the fins. Be sure to support the part with a center to help with chatter. Once you’ve got the fins cut, part off the cylinder leaving an extra .005 or so of length so that you can face and polish the newly parted end.

Step 3: Mill the Large Flat

Support the part in the mill vise and mill the flat. I have a horizontal mill, so I’m using a vise within a vise to support the part parallel to the end mill. Remember to use some sort of packing. Aluminum shim or card stock (I’m using an old business card). Take light cuts of .005 to .010 – deeper if you’re using a larger end mill and have a more robust mill. But be careful not to put to much cutting force on the part – it could cause it to shift or rotate spoiling all your work. Take your time.

I used the knee to keep track of the depth of the flat, but you can also use a multi-anvil micrometer (with the cylindrical anvil inserted) to measure from the ID of the hole to the flat to double check your knee measurements.

Step 4: Cut the Two Small Flats

This step is pretty simple, but I thought it was worth mentioning the use of a square to ensure that the two small flats are at a 90 degree angle to the larger flat. The angle isn’t critical (89 or 91 would be fine) but it’s nice to have a way to get it very close.

Here’s a finished flat.

Step 5: Drill and Tap for 2x Cylinder Supports

You want to be very careful on this step not to drill too deep. If you drill into the cylinder bore, you’ve scrapped your part. One way to make sure you don’t drill too deeply is to use the knee on your mill. I’ll show you another method using a drill press and a plug gage.

Position the part so that the drill bit is just touching the surface of the flat (note: you should have already center drilled). Then use a plug gage with a diameter that matches the depth of hole you wish to drill. Position the plug gage between the nut and the stop on your drill press depth stop and turn the nut until it’s contacting the plug gage. Then remove the plug gage. Now when you plunge your hole the nut will contact the stop after plunging to the depth set by the plug gage. Now tap your holes using a plug and bottoming tap.

Step 6: Mill the Intake Slot

This is pretty self explanatory, but I would recommend using an undersized end mill. If you use a 3mm end mill from the start, your slot will be 3mm+ and ugly. I used a 7/64″ end mill and plunged the slot to depth and length using multiple passes along the centerline of the slot. Then I widened the slot from .109 to the 3mm (.118) width by taking a bit off of the top and bottom of the slot.

Here’s another view.

Step 7: Drill for the Valve Push Rod

This was another step that concerned me – but drilling the 4mm hole turned out to be a piece of cake because each gap between fins allowed for chip clearance. I could have plunged the entire hole all at once, but I didn’t.

Step 8: Bronze Bushings

Making the bronze bushings was easy. Turn the speed up to a fast speed appropriate for bronze of that diameter and use a sharp tool. I was able to turn the bushings unsupported using light cuts (.005) and a sharp tangential tool holder from Eccentric Engineering. I’ve worked with bronze in the past when I replaced the bearings on my Atlas mill and the bronze chipped away in dusty flakes. With my new tangential tool holder the bronze made long spiral chips and had a beautiful finish. Coincidence? No. Get yourself a tangential tool holder. You won’t regret it.

Once you’ve turned the OD and drilled the ID, part the bushins off and press them in. They are small enough to be pressed with a quality vise. However, as a general rule I’d recommend you avoid using your vise as an arbor press. It’s a bad habit that can lead to a busted vise casting. Add an arbor press to your tool list. Harbor Freight sells them and they are much cheaper than a new AngLock-style vise.

Step 9: Fixing a Rookie Mistake

By now you should be finished with your cylinder. But I wasn’t. I mentioned at the beginning of this post that I’ve been working on this project for over a year now and I turned most of the parts that I could make on the lathe months ago. Unfortunately I substituted a 5M thread for a 4M thread on the cylinder supports and completely forgot. Months later when it came time to make the cylinder I went ahead and tapped the holes for a 5m thread, not a 4m. The result can be seen below. Obviously the support legs didn’t fit. I fixed them by re-chucking them and drilling them out for a 5m set screw – which I inserted upside down and secured with lock-tite. It’s upside down because the support has a hole all the way through – so I’d have access to the set screw using a hex wrench if I ever needed a little extra force to remove the support legs from the cylinder.

I’m including my mistake so that you guys can learn from it.

Lesson learned: Never change a part without making a change to all mating parts on the blueprints.

But I also thought it would be an opportunity to show you that some parts can be made without using a die. I could have easily made these supports using a 5m tap and set screw from the start – totally avoiding the need to purchase a 5m die. Since I normally work in inches and not mm, this would have saved me a few dollars. I have a complete set of taps and dies up to 1″, but up until this project I didn’t own any metric taps or dies. So this little engine has actually been a bit more expensive than I had hoped. But that’s ok. And of course you can usually use the nearest inch equivalent when tapping holes of a non-critical dimension. I just didn’t want to. I wanted to use this project as a chance to practice working with metric measurements.

Well, that’s about it for the cylinder. Next I’ll be documenting the steps for making the piston, valve, and connecting rod.

 

[adrotate group="4"]

About Tyler

Tyler is a hobby machinist and 3D printing aficionado. He teaches computer programming and web development at Highline near Seattle. Tyler founded Projects In Metal in 2008 because he was frustrated by the lack of free plans available for hobby machinists.

38 comments

  1. fenichel said:

    Where did you find the ball bearings for the engine?  McMaster-Carr won't ship outside of the US.  Is there another plausible source?

     

      Also, I have brass on hand, but no bronze.  Would I be making a big mistake to make the bushings of brass?


    Hi Fenichel, sorry it has taken me a few days to respond. 

    I actually purchased the bearings from a local supplier (Grainger), but McMaster was my second option. You can look up Grainger Supply and see if they ship outside the US, or like Jerry said, try a hobby shop that specializes in RC equipment. 

    As for the bronze vs brass, I assume you're talking about the bushings for the wire that operates the valve, correct? If so, brass would be fine. It might not last as long, but you're also not going to be running this engine for hundreds of hours, so I don't think it will matter either way. 

  2. Tyler said:

    Did you try running it with a small amount of WD40 in the cylinder? That helped get mine running. I squirted a bit in the cylinder, and then turned the engine over a few times without a flame so that it would spit out any extra WD40 (hold a paper towel in front of the valve opening to catch whatever it spits out.

    . . . .
    I don't think the WD is turning the engine into an IC engine, I just think it's lubricating things and helping to make a tighter seal around the piston and valve.

      Tyler, do you have any other suggestions for getting a recalcitrant flame-eater to operate?  I just finished mine, and it doesn't work.

    • The piston & valve (on the second try) were a very tight fit, and they needed to be forced through the cylinder a few times before they would move freely.  I don't think they could be made any tighter.
    • I tried the WD-40 trick, without success.

    I suspect that the problem is related to adjustment of the Adjuster and/or to the extent to which the Fork is screwed in to the Piston.  As my engine is now,

    • the valve has about 6 mm of travel; about 2 mm of this is to the right of the point at which it covers the firehole.
    • at the leftmost point of the piston's travel, it about half covers the firehole.

    How do these figures compare to those of a properly operating flame-eater?

     

    Bob Fenichel

  3. Hi Bob, it's been a few days since you've posted this, so hopefully you've made progress?

    Sorry for my delayed response, but I've been in CA and Eastern WA for work meetings and I don't use my work laptop to access the forum. And tomorrow I'm heading back to Eastern WA for a camping trip and I'll be gone all weekend without internet access. 

    And unfortunately I can't answer your question about piston travel at the moment since my engine is disassembled.

    But I'll still try to help. First, have you looked at Jan's site in the troubleshooting section for the engine? He has a lot of info on how the engine should act (ie how many revs it should make free-spinning with the valve removed, how quickly/freely the piston should drip through the cylinder, etc).

    You talked about the piston placement, but not the valve. Does the valve close over the inlet hole properly? If not, it needs to in order for the vacuum to form. The piston doesn't cover the hole if I remember correctly, it just pushes the valve back open to allow for more flame to be sucked in.

    It took me more than a week to get mine to run. It was a real finicky engine to get going. It was so bad that at one point my friend and I wondered if we'd been duped (meaning we wondered if Jan's videos concealed an electric motor somewhere that powered the engine). Rest assured, it can be made to run, but it's not easy!

    I also honed my cylinder smooth with a Sunnen honing machine, which might have helped, or at least didn't hurt. If you don't have access to a honing machine you could try lapping your cylinder to make it smooth. Mine was mirror smooth on the inside when I finished honing it. 

     

  4. Tyler said

    First, have you looked at Jan's site in the troubleshooting section for
    the engine? He has a lot of info on how the engine should act (ie how
    many revs it should make free-spinning with the valve removed, how
    quickly/freely the piston should drop through the cylinder, etc).

    No, I didn't know about that section.  Thanks for the lead.

    You talked about the piston placement, but not the valve. Does the valve
    close over the inlet hole properly? If not, it needs to in order for
    the vacuum to form.

    I did mention the valve, but perhaps I wasn't clear.  It moves about 6 mm in all.  The last 2 mm of its rightward travel comes after it has completely covered the firehole; this portion of its travel serves no obvious purpose, but is not obviously harmful, either.

    The piston doesn't cover the hole if I remember correctly, it just
    pushes the valve back open to allow for more flame to be sucked in.

    That the piston should even partially cover the hole is counter-intuitive to me, too, but I don't claim to have a well-developed intuition for this sort of engineering, so that's why I asked.  Perhaps someone else with an assembled flame-eater will chime in.

    It took me more than a week to get mine to run. It was a real finicky
    engine to get going. It was so bad that at one point my friend and I
    wondered if we'd been duped (meaning we wondered if Jan's videos
    concealed an electric motor somewhere that powered the engine). Rest
    assured, it can be made to run, but it's not easy!

    That's good (it says that I'm not necessarily incompetent), but also bad (considering the extent to which your experience obviously exceeds mine, it says that I may not be adequately competent to get the FE to run).

    I also honed my cylinder smooth with a Sunnen honing machine, which
    might have helped, or at least didn't hurt. If you don't have access to a
    honing machine you could try lapping your cylinder to make it smooth.
    Mine was mirror smooth on the inside when I finished honing it.

    When I first made the cylinder, I drilled it, bored it, and then honed it with an 18 mm SiC hone (http://www.kbctools.com/can/Na…..DFPage=607).  But then my piston & valve were too small, so I redid them.  They were then a little large, so I re-bored the cylinder but had no nice means of honing it to its new ID.  It is smooth now, but surely not mirror smooth.

     

      Again, thanks for your help.

  5.   I went to Jan's flame-eater site, found the troubleshooting page, and tried to print it out, but suddenly got “server not found” errors.  Now I can get back to the Dutch version of the page, but the putative link to the English version gets me

    File not found

             Firefox can't find the file at http://heetgasmodelbouw.ridder…..engels.htm.

      Check the file name for capitalization or other typing errors.
      Check to see if the file was moved, renamed or deleted

    Does anyone here have a saved copy of the page?

  6. Bob, I just clicked on your link and it took me right to Jan Ridder's site. Perhaps he was having server issues?

    Sammy

  7. sammy said:

    Bob, I just clicked on your link and it took me right to Jan Ridder's site. Perhaps he was having server issues?

    Sammy

     I just clicked on it and in Firefox I got the same non-result as before.  I found an old copy of of Internet Explorer and was able to bring up the page, but I couldn't print it; I captured it as text.  At the same time, my antivirus software (MS Security Essentials) reported that the page contains a JS/Obfuscator.BN virus, so I am going through a disinfection cycle.