How to fit mismatched rings and piston and cyl bore??

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.


Well-Known Member
HMEM Supporting Member
Mar 8, 2019
Reaction score
Charlottesville, VA
Hello everyone. New guy here eager to learn and willing to take some educated risks.
I am collecting material and parts and pieces for a 52cc horizontal diesel with cross head. It is my first model engine build, but I am ready for the challenge and I hope I am up to the task. A lot of what needs to be done seems fairly intuitive, but there are questions that are begging for experienced answers to help me avoid sad mistakes. Any and all advice will be appreciated. The learning is a big part of the fun.

It seems that material availability is always an issue and many of the design decisions are driven by what you can get that is "close enough." I had planned on having a 1.260" bore, using 1.50 O.D. x .120 wall DOM 4130 normalized tube. Perfect for purchased 32mm piston rings. But the tube arrived with 1.275 I.D., well within the industry tolerances, smooth, round, and ready for honing. But that is 32.39mm, so I guess I need to go to 33mm rings (which I now have) and somehow fit them to the 32.39 bore. I feel that trying to bore the tube out to 33mm would leave me disappointed.

Question 1.
So, do I end file the rings to fit the tube I.D. and then lap them inside a piece of the tube to get them round? With minimal fixturing, it seems quite doable, but I am probably missing something. Is there a danger of ruining the O.D. of the rings with lapping? I can't tell if they have a taper, but the anti rotation notches definitely asign a top and bottom to the ring.

Question 2.
This will be a cross head engine, so no side thrust load on the piston. What material for the piston? I have some Nitronic 60 stainless which has extremely good anti-galling characteristics and should handle the diesel pressures and temperatures without a whimper. It will add some weight, but this is a cross head, so reciprocating weight is already thru the roof, so what's a few more grams. What about aluminum for the piston? I have no desire (nor skills) to cast something from old pistons, but is 6061 or 7075 or 2024 a better choice than the Nitronic 60. And honestly I am only 95% sure that the N60 is indeed that. There is a 5% chance that it might be 316 instead.

Question 3.
For calculating the fit of the piston to bore, I am planning to use a 500 degree F temperature differential between the piston and cylinder wall (water cooled) to make my clearance calculations. Is that a reasonable approach? And what should the final "at temperature" clearance be?

Thanking everyone in advance for their thoughts on this.
I'd use 7075 for the pistons and turn them for .002" clearance to the cylinder bore. It's easier to make your own rings to fit the cylinder bores after they've been honed. But if you don't want to get into this on your first engine, it would seem you have two choices. If it were me, I'd take one pass with a boring bar and open up the cylinders to within a couple thousandths of your rings and then hone the cylinders to fit them. You said your tubing was already round, but with no measurements I'm not sure what round meant to you. Tenths are important on ring fits. Practice with speeds and feeds on a piece of scrap with your sharp boring bar until you can get an acceptable surface finish. Avoid chucking the tube over the area of your material that will eventually become one of your cylinders in order to avoid distorting the tubing. If you can't make this work to your satisfaction you'll have to turn the o.d.'s of the rings (if there is enough material on them to do so). This would involve making a clamping fixture to hold them closed in the lathe chuck so you can turn (not lap) down their o.d.'s. You need to keep the corners sharp and maintain full contact with the cylinder wall which is easier guaranteed with turning. If you try to lap their o.d.'s with paper, you run the risk of rounding them over. Again, I'd be surprised if you have the extra .025" to remove from their o.d.'s. Some others will tell you to just use a Viton o-ring. For a first engine that might be good advice. I've never really tried it myself since rubber piston rings feel too much like training wheels on a bike, and making cast iron rings seems like a skill that a model engine builder would want to have in his toolbox.- Terry
Thanks very much Terry, 7075 it is. I studied and measured the tube and rings and then re-measured and thought some more. The tube I.D. was round within .0004 using a telescoping bore gage and mics, but honestly, I think I could feel the variation (with the telescoping gage) with more sensitivity than I could measure it.Your advice about boring and honing makes the most sense, but I will stick with the purchased rings... this time. I have had good luck boring the normalized 4130 tube in slightly smaller sizes when I paid attention to detail. So that is what must be done. The rings are .058 tall x .062 deep (gland depth) so you are right about making them flimsy if I try to modify them. I will have to search the forum for info on proper honing. The total honed tube length will be approx 3.5" x 33mm bore to fit the rings.
Thanks again,
Thank you Terry, your post number 428 was exactly what I needed, and several things popped out at me.
First off, your honesty about the effort and struggle to get 12 identically honed cylinders, reminded me (briefly at least) that you too are a mere mortal and that often there is no "magic" that can fix a problem. Thank you for letting us know that it doesn't always come out perfectly, and that it is sometimes necessary to adapt to the situation, or heaven forbid, actually scrap a part out.

Next was your comment buried in the post saying that the 1000 grit might have been overkill. Thank you for being honest about that too, because when I saw the pictures of the 3 cans of compound, I just thought to myself, the level of precision needed is really pushing the envelope, LOL.

The warning about chatter was sobering. That was part of my initial reluctance to boring the tube out, but I will probably bore out some sort of nest fixture with slots so that it can collapse like a collet onto the tube. That, with a more rigid boring bar set up should make it work out ok.

The barrel laps look quite doable. I have made small ones like that for tapering the land dia in air rifle barrels, so it should work out. Broken down into discrete steps as you documented, changes this process from something intimidating and daunting, to a challenge to be enjoyed.

Thanks again.
Your own injection system or something adapted?
It will be my own design, which isn't down on paper yet. And a design review would be most appreciated when the time comes. I do have some experience designing & building 3,000 psi air valves & pumps for air guns but will need help with the subtleties of diesel.
Thank you, Lloyd
I am thinking that I should probably transition this discussion over to a new thread on "A Build in Progress," so that all of this advice can be accessible in one place. Thoughts on that from anyone?

Dieselpilot, you asked about direct injection, and that actually brings up a question that I need helpon before I can really start designing the injection pump and injector. That question is the volume of diesel fuel (or kerosene) that will be required per injection. If the calculation were strictly by a stoichiometric reaction, I believe that is a ratio (by weight) of 15 parts air to one part diesel fuel. If the single cylinder is 1.30 bore x 2.6 stroke (33x66mm) the swept volume is 56.5cc, that would be a total mass of air of .073 grams(edit- was .0073). A 15 to 1 ratio would require .00049 grams of fuel per injection. My understanding is that in actual operation the ration might need to vary between 15 (edit-was 51) to 1 and 100 to one.
So the fuel required could range from .0049 to .00073 grams.(edit- was .00049 and .000073)
If diesel fuel weighs .832 grams per cc, or 1.20 cc per gram(edit-was grams per cc) , that calculate out to a range of .00588cc to .00088cc per injection.(edit- was .000588 and .000088)
Now, IF those calculations are correct and I make an injection pump with a
Plunger of 1.5mm dia, the stroke would have to vary from 3.3mm to .50mm.
Or, using imperial measurements, a Plunger dia of .0625" with a stroke to vary from .115" to .022".

Knowing that calculated and real-world requirements can be quite different, and if I haven't messed up the math, I am wondering if those injection volumes sound at all in the ball park?
Thank you,
Last edited:
Similar to numbers I've come up with, though you have an extra decimal place in the fuel volumes. Compressibility may be a factor at these pressures. I haven't attempted to build anything yet. Surely, you've seen Find Hansen's work. If you haven't, watch the video. With the number of factors against miniature injected diesel, it's amazing his work so well. I don't think any one else has done this well.
dieselpilot, Thank you for catching that decimal place error in my post #10. I have checked over the math and marked the edits that I had to make. Despite the decimal place errors, the final results that I had for the actual plunger diameter and stroke length were correct.
For a swept volume of 56.5cc:
15 to 1 air fuel ration 5.85 cubic mm per injection.
100 to 1 air fuel ratio 0.88 cubic mm per injection.

Yes, I had indeed seen Find Hansen's videos of his beautiful diesels. They are actually what made me decide to really do this. I studied (for the umpteenth time) his A frame assembly video. He gives numerous hints and dimensions here and there, but seems to avoid giving any critical dimensions. But, by scaling off the computer display screen using dimensions that he mentioned (i.e., conn rod 6mm, small copper tubing 2mm dia, etc) I estimated that his injection pump has a plunger of about 1mm dia x a max stroke of about 1.6mm. I found it interesting that the tapered wedge that governor moves to adjust the stroke length allows for a stroke variation of maybe 3 to 1. So if the max stroke is 1.6mm, then the minimum might be about 0.5mm. So the maximum fuel per 1.0mm dia x 1.6mm injection stroke is 1.26 cubic mm.

Doing the stoichiometric 15 to 1 calculation for his 20 x 40 engine (12.6cc)(.0162 grams of air) yields a fuel volume of 1.31 cubic mm per injection. Frankly, I had a big grin on my face after I did that calculation and saw how closely the measured-from-the-screen estimate of 1.26 cubic mm compared to the calculated 15 to 1 of 1.31 cubic mm. Science is so amazing!

So, continuing in that vein, if his max fuel volume is 1.31 cubic mm, then his minimum volume would be 1/3 of that, or about 0.43 cubic mm. That would be an air fuel ratio of 45 to 1. The calculations I did in post 10 were from a ratio of 15 to 1, to 100 to 1. So it seems like his engines don't run as lean as 100 to one. So maybe I should adjust my minimum injection volume up a little from my original calcs. Having a max to min stroke ratio of 3 to 1 (as his tapered wedge provides) should be easier to build than something that must adjust by a factor of 6.

diesel pilot and terry, thank you for your help and encouragement. This is getting exciting!
I think relatively low pressures (for diesel injection) and other factors limit minimum injection. On his website it says the plunger is 0.9mm for the 10cc horizontal.
diesel pilot, Again, than you for drawing out this information. What you are saying about the self limiting action of the injection "system" makes perfectly good sense. Injecting against the compression pressure certainly reduces the pressure differential between the injector and the cylinder, and therefore the actual flow volume would be reduced. Also, elasticity in the high pressure side of the system, and maybe leakage past the check balls might be some of the "other factors" that you mention. I am a bit confused about his statement that the compression pressure is 49 bar (710psi). That is much greater than a simple 20 to 1 pressure increase(20 bar), but less than the 68 bar (986psi) that 100% adiabatic compression would yield. I am sure I am missing something.
But I think the bottom line for the pump design, is that the pump should be capable of producing over 1,000 psi, and that the max pump volume per stroke needs to be about 5.8 cubic mm.
I hope I haven't slipped any decimal points this time, LOL.
Yes, all that matters. I don't' know his background, but he must be quite knowledgeable. He's either measured pressure or is taking into account valve events and/or heat loss.

Latest posts