Performing a vacuum test on model IC engine

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.

Ken Brunskill

Well-Known Member
Joined
Feb 16, 2019
Messages
163
Reaction score
34
Location
Fremont, CA
Interested in how others have come up with means of testing the vacuum on their model engines.
Wondering if a Schrader valve would be useful & what it's proximity to the cylinder/cylinders ought to be.
 
Not as much as determining the intake integrity and timing. 1st application is on an Atkinson Differential model, then on to a Parcell & Weed model.
 
Interested in how others have come up with means of testing the vacuum on their model engines.
Wondering if a Schrader valve would be useful & what it's proximity to the cylinder/cylinders ought to be.
One of the issues that come to mind is: how much does the area/volume prior to any type of check device (before the gages) come into play? Seems like that would get both pressure and vacuum, as I'm considering checking at the sparkplug hole. Maybe I'm stating the obvious :rolleyes: & it can't be done there.
 
Ken,
Glad to see you're still working on the Atkinson which I still believe you'll get running eventually. I've had good success locating leaks using a stethoscope (or in reality a length of flexible plastic tubing with on end stuck in my ear and the other end probing the intake and exhaust), and compressed air in through a dummy hollow spark plug. Prior to this I tried using the plastic tubing as a manometer, with some water at the bottom of a tall "U" shape one end plugged and the other end connected to an intake or exhaust, but that never worked out very well, and more than once had the water sucked into the engine !, but the main problem is accidentally getting bubbles in the water which pretty much requires emptying and refilling and that just gets time consuming and annoying, at least if you're trying to make comparative measurements across multiple cylinders, but for a single cylinder engine the manometer might not be such a bad option.
Pete.
PS, the business end of my "stethoscope" has a length of empty ball point pen ink tube which can pin point small leaks (I found that some of the intake manifold screws on my Merlin were leaking this way)
PPS, if the water wets to the surface of the plastic tubing then it always microscopically runs down hill, so it always appears that your engine has a leak even if it doesn't. I suspect ordinary polyurethane tubing doesn't have the problem, but the stuff I was using did.
 
Not as much as determining the intake integrity and timing. 1st application is on an Atkinson Differential model, then on to a Parcell & Weed model.
What I do is use a rubber tip blow off nozzle set to low PSI on the air compressor to the spark plug hole. Turn over flywheel. You will know exactly when those valves open/ close and note any leakage. You may hear some piston ring pass by too but the valves hiss is a timed cycle event. Capture.PNG
 
Ken,
Glad to see you're still working on the Atkinson which I still believe you'll get running eventually. I've had good success locating leaks using a stethoscope (or in reality a length of flexible plastic tubing with on end stuck in my ear and the other end probing the intake and exhaust), and compressed air in through a dummy hollow spark plug. Prior to this I tried using the plastic tubing as a manometer, with some water at the bottom of a tall "U" shape one end plugged and the other end connected to an intake or exhaust, but that never worked out very well, and more than once had the water sucked into the engine !, but the main problem is accidentally getting bubbles in the water which pretty much requires emptying and refilling and that just gets time consuming and annoying, at least if you're trying to make comparative measurements across multiple cylinders, but for a single cylinder engine the manometer might not be such a bad option.
Pete.
PS, the business end of my "stethoscope" has a length of empty ball point pen ink tube which can pin point small leaks (I found that some of the intake manifold screws on my Merlin were leaking this way)
PPS, if the water wets to the surface of the plastic tubing then it always microscopically runs down hill, so it always appears that your engine has a leak even if it doesn't. I suspect ordinary polyurethane tubing doesn't have the problem, but the stuff I was using did.
Like your idea of a 'stethoscope' type device - way simpler than a machining a tool that won't work anyway (still enjoyed the designing time ;)). Haven't totally given up on that idea just yet, stubbornness, I guess.
I've come up with a SolidWorks design that might test compression and vacuum at the same time. I'm thinking that this will require (2) check valves close to the test port (1) for vacuum & (1) for pressure, before the gages. However, after serious thought, seems like that volume will just oscillate pressure and vacuum and negate the test.
 
What I do is use a rubber tip blow off nozzle set to low PSI on the air compressor to the spark plug hole. Turn over flywheel. You will know exactly when those valves open/ close and note any leakage. You may hear some piston ring pass by too but the valves hiss is a timed cycle event.View attachment 144738
Great idea, thanks!

I've come up with a 'Smoke Test' device by modifying an old CPAP machine (produces a really low pressure). I use incense in the water chamber to produce the smoke, and an adaptor to attach a fitting from the CPAP machine to the sparkplug hole.
 
Here is a confession that you might get a chuckle over, or rememeber doing the same or similar thing.

I was desiring to measure the compression and vacuum on the Atkinson Differential I've built as well and then have a device for future builds. (Reference post of Feb. 16th.) I'd created several versions in Solid Works after completing one, then thinking 'hey that's too complicated' and then simplifying the design. Then on the 6th iteration I was interested in making sure the gages would not interfere with each other and went to McMaster Carr to get their Solidworks solid model of the gage(s), there I was surprised to find that one can obtain a single gage that measures both pressure and vacuum o_O. And for only $16.07+ Tax.
Admittedly, I've not had a lot of experience with gages and what was available, if I'd had that I most certainly would not have gone down the design my own path.

You can guess the rest;).
 
Ken, here's the result of doing the math in reverse that I tech-talked today,
PSI = 66
CR = 3.375 : 1
that's below the CR of 4.7 that someone quoted for the Model T,
but is around the CR for model turbo-jet engines that have been used in some RC planes (it should be noted that the exhaust gas temp for model jet engines is in the 500 to 600 Celsius range, and that's with mixing in lots of cool compressor air in with the combustion gas before turning the turbine).

so its not that an engine with this low compression cannot run, rather that its exhaust will necessarily be hotter (the lower the CR the less mechanical energy you're extracting from the combustion energy, leaving it hotter).

rather it would be helpful to know what PSI is observed in other running Atkinson Differential engines to see if you're in the right ballpark, using similar measurement gage, similar volume of pipes and fittings, and especially similar one-way valve, as these things will always introduce errors so best to compare by using the same errors :) !

my guess is that 66 PSI / 3.375 CR is too low, and it should be up above 100 PSI where folks that have measured their run-able engines seem to get (they're not seeing the theoretical 166 PSI for 6:1 CR because of "pumping losses", IE valve timing and manifold vacuum issues, plus heat lost through the head and cylinder walls, all of which is expected).

for the A-D engine I suspect that valve timing is an issue in addition to head room. if the valve timing is built into cylinder ports like in a 2-cycle model airplane engine then there's not much to be done except make a new one (unless its also a function of the weird crank pins and rods that I vaguely remember, or maybe the A-D doesn't have as weird a mechanism as the A-C engine does ?), but the head room might be easier to adjust and experiment with ???

I guess the good news is that getting 66 PSI means your rings are sealing, although that might still improve with more break-in.

Peter.
 
Last edited:
Ken, here's the result of doing the math in reverse that I tech-talked today,
PSI = 66
CR = 3.375 : 1
that's below the CR of 4.7 that someone quoted for the Model T,
but is around the CR for model turbo-jet engines that have been used in some RC planes (it should be noted that the exhaust gas temp for model jet engines is in the 500 to 600 Celsius range, and that's with mixing in lots of cool compressor air in with the combustion gas before turning the turbine).

so its not that an engine with this low compression cannot run, rather that its exhaust will necessarily be hotter (the lower the CR the less mechanical energy you're extracting from the combustion energy, leaving it hotter).

rather it would be helpful to know what PSI is observed in other running Atkinson Differential engines to see if you're in the right ballpark, using similar measurement gage, similar volume of pipes and fittings, and especially similar one-way valve, as these things will always introduce errors so best to compare by using the same errors :) !

my guess is that 66 PSI / 3.375 CR is too low, and it should be up above 100 PSI where folks that have measured their run-able engines seem to get (they're not seeing the theoretical 166 PSI for 6:1 CR because of "pumping losses", IE valve timing and manifold vacuum issues, plus heat lost through the head and cylinder walls, all of which is expected).

for the A-D engine I suspect that valve timing is an issue in addition to head room. if the valve timing is built into cylinder ports like in a 2-cycle model airplane engine then there's not much to be done except make a new one (unless its also a function of the weird crank pins and rods that I vaguely remember, or maybe the A-D doesn't have as weird a mechanism as the A-C engine does ?), but the head room might be easier to adjust and experiment with ???

I guess the good news is that getting 66 PSI means your rings are sealing, although that might still improve with more break-in.

Peter.
Peter, thanks for the supportive comments, I've sent the originator of the plans a note to see if he has any pressure or vacuum data on his running model.
I'll spend some more time making some careful measurements with various valve placements, the initial trial I reported was a 'WAG' for the valve placement. Where I was simply shooting for maximum pressure and vacuum. (Trying to seal off both just to get an idea of the ring seal.)
 
Ken, of course the first question should be are you getting any "pop", if not then its not compression, a proper fuel + air mixture should pop even at very low compression,
 
Ken, I too have been having trouble, my Merlin won't start any more, out of desperation having tried everything, I got some "starting fluid" (AKA ether) and sprayed a shot into my intake manifold, and I heard it pop on the first flip, and a couple flips later it started and ran. Would like to hear from you about trying that on your engine as well. Even if the engine isn't ready to run, it should at least give a pop which will confirm ignition and at least some compression, then you can focus on valve timing. Pete.
 
Last edited:
Back
Top