IC engine valve troubles

[RonGinger]

Many years ago I built a 4 cylinder engine, mostly following the PANTHER PUP design. I made it water cooled, so the block and head are completely my design.

It eventually ran, but I had a tough time getting the valves to seal, and I dont think I ever got all 4 cylinders to fire. I ran it at shows for a few years, but it eventually refused to start.

Today I decided its time had come, so I tore it apart

I first tried to measure compression. I made an adapter to fit my car compression tester, but all I got was a needle flicker, no pressure at all. I suspect the total volume of the air in the hose and gauge is much more than the displacement of the engine, so it doesnt read.

Has anyone found a way to measure compression of a 1" bore cylinder?

My valves all show a shiny ring that looks like a reasonable seat, and no signs of any leak spots.

The valves are tight fit in the valve guides. I have read that loosening that fit can make a valve seat better. Anyone have experience with that?

Should I try to lap the valves in, as with big engines? I have heard this causes more problems with small valves because the seat becomes to wide and the effective pressure is to low to seal/

Im thinking about installing the head and using shop air through the spark plug hole to pressurize each cylinder an see if there are any leaks. Has anyone tried this?

Id like to get this thing running again.

 

[stevehuckss396]

How wide are the seats. To get a tiny valve to seal the seat needs to be as thin as .010 wide. If you need to lapp the valves use a very high grit number compound. Try 5000 or something. You are correct that normal compound will only make it worse. Brian Rupnow built the Kerzel hit and miss and had a terrible time with the valves. Take a look at his thread for do's and donts.


As far as the compression tester goes, I have seen articles and I think on this board where a check valve was installed between the cylinder and tester. The volume of the mini cylinder is the problem. The check valve will allow you to "pump up" the tester to get a more accurate test.

 

[deverett]

Jan Ridders designed a simple compression tester.

http://modelengineeringwebsite.com/Compression_tester.html

Dave
The Emerald Isle

 

[Brian Rupnow]

Stevehucks is correct. I had a terrible trial with my Kerzel build, but eventually I did get the valves to seal. I think the final solution was to build a valve seating tool which was posted by one of this forums talented members. There is information in my thread about the tool, and a link to the tool builders original thread.---Brian

 

[gbritnell]

Hi Ron,
The only reliable way to check valve leakage is with vacuum. That being said here's what I do. Make some type of adapter to fit over your port or a tube that will taper fit into your port. Once done just suck on it. If it will not hold a vacuum then you have a problem. Generally the problem will be with the seat area. As Steve said, use a very fine grit lapping compound and just slowly work the valve against the seat. To do that I chuck the stem end of the valve with my small hand chuck and after applying a small amount of compound to the seat area I spin the valve with the chuck (by hand). You want to rotate it back and forth a few times then lift the valve from the seat and rotate it to a new position and repeat so that you are hitting a different high spot.
One thing you don't want to do is enlarge the valve guide hole. If you do you will draw air through there and it will make it almost impossible to check. One way that will help if your guide hole is too big is to insert the valve and then put a ring of grease around the top of the stem and the valve guide. This will seal it up enough to be able to apply vacuum to it.
After lapping it for a short time, pull the valve out, wipe all the grit off the seat and the valve and look to see if there is a nice uniform gray color to both surfaces. If so then try the vacuum test again. Believe me when you get it right you will not be able to suck through the port.
As far as a compression testing device, I have tried several different methods but not with much success.
The better automotive type testing devices have a check valve down near the spark plug adapter. The cheap ones don't. The reason to have it as close to the adapter as possible is that you are not adding the volume of the hose to the volume of the cylinder. On an automotive engine the additional volume is negligible but for small engines, motorcycles, lawnmowers etc. it does make a difference. This is almost impossible to do with our small engines.
George

 

[Rustkolector]

Rather than compression, I prefer using vacuum to test valve seating. I use a simple brake bleeding vacuum pump kit sold at Harbor Freight. It is done easiest with the head(s) removed. Lubricate the valve stem with heavy oil. With the valve springs installed, pulll vacuum on the intake or exhaust port. You should be able to pull down to about 25" hg on the gauge. If the gauge doesn't drop below 20" in 10 seconds, you have about the best seal you can ever expect, but that degree of seal is not needed for good operation. As Steve has said, a narrow seat is preferred and will seal easiest. I use 800 grit grinding compound. If that doesn't seat a valve in a few tries, the seat is probably not concentric with the guide bore and will need to be recut. Specially made seat cutters that use a pilot are best.

Jeff

 

[Lakc]

For a compression tester I used the smallest pressure gauge I could find, and made the shortest 1/4-32 to 1/8 pipe adaptor that could clear the head, worked like a charm without a check valve. It wasnt hard to tell where the needle peaked even as it was swinging. You are correct, volume is the enemy.

I used regular permatex valve lapping compound, but I wouldnt recommend it for most model engines. I have ground tool steel seat inserts and dual valve springs so stiff I can barely compress them with my thumb. Its not a common model setup.

I made a leakdown tester and tested the seal with compressed air. That worked fine for large leaks but not so well when they got smaller.

The best leak tester I used was my finger. If you cover the exhaust port and it leaks you will feel it suck against your finger on the intake stroke. Cover the intake port and you feel it blow on the exhaust stroke. I knew I had it right when I felt the compression at the exhaust port as the exhaust valve opened.

 

[RonGinger]

Thanks for all the response. I have also re-read Brians kerzel post, and looked at Georges seat cutting tool.

I think I have a plan now, tomorrow Ill get on it.

thanks everyone

ron

 

[Swede]

There is a brand of lapping compounds called "Time Saver" that is engineered and guaranteed NOT to embed. Whenever I am concerned about being too aggressive with grinding compounds, I go for the Time Saver. It seems to be a lot more forgiving and gentle than typical ALOX or silicon carbide pastes.

I also agree that loosening up the valve stem to guide interface can work wonders. We tend to dislike doing this, but sometimes the reality is that looser tolerances work better. I think 0.0015" to 0.002" is good for typical model engine valve stems of 0.125" or 0.156".

Time Saver: http://www.ws2coating.com/timesaver.html

 

[RonGinger]

My head is a flat plate so I was able to take a light facing cut across it and remove all of the old valve seats from the cages, so I had nice square end brass cages.

I made a cutter like George suggested and cut a nice small seat. The cutter worked very well, it was a firm fit in the valve guide so I think the seats are good and concentric.

I lapped each seat with a 320 grit compound. I timed the first couple and was lapping for 1 minute. You can hear the sound change when the lapping stops cutting. I think I got a little hasty on the later seats.

I tried sucking on a plastic tube in the ports and thought I had a reasonable seal. When you pressed the valve to release it there was a clear snap to it.

I put it all back together, made a new head gasket and now I seem to have lower compression than I did before I started.

I think I also discovered why it stopped running- the carb and fuel tank was a mess of brown, sticky gunk from old fuel. Ive cleaned all that up now.

This morning I bought a hand vacuum pump kit with a bunch of tubing fittings. I have not tried it yet.

I think I may put it all back together, make a nice static display out of it, and go back to making steam engines.

 

[Lakc]

I put it all back together, made a new head gasket and now I seem to have lower compression than I did before I started.

You shaved the head, did you relash the valves?

 

[RonGinger]

Yes, I did re-set the valve clearance.

I have now tested with the vacuum gauge. The best ones will pull to about 20", then take 7-8 seconds to drop to zero. Some that were fast drop I have re-lapped to get to the 7-8 seconds drop.

I tested one while it still had lapping compound on it and it held solid at 20", but thats just the effect of the compound making a seal.

Is a 7-8 second drop good enough?

 

[Brian Rupnow]

Ron---I ended up using 400 grit followed by 600 grit when I hand lapped the Kerzel valves, after using the valve seat cutting tool.----Brian

 

[stevehuckss396]

I lapped each seat with a 320 grit compound.

Ron, Take another swipe, recut the seats, install the valves and start the darn thing. When you lapp you damage the seats. I never lapp anymore.

 

[Lakc]

Sorry Ron, I had to ask.

Like Steve said, run it and see how it goes.

 

[stevehuckss396]

how does lapping damage the seats?? ???

It scratches them all up. If you cut the seats and get a good finish, and do the same with the valves, then the 2 smooth surfaces will seal. That us if they are concentric with each other. When I made the V4 i kept lapping with everything under the sun. Nothing but trouble. New seats with a new boring bar, recut the valve surface with a brand new bit, install both and started right up. Works for me. Could work for him to.

 

[Swede]

I think it boils down to the grit used. Typical "Clover" brand, the stuff you'f sind in an auto supply store, is going to be way too coarse for these tiny valves and seats.

When you think about it, a polish like "simichrome" is a form of lapping compound. Anything with abrasive in it will cut metal. Good mild abrasives might be toothpaste, as mentioned, rubbing compound, rottenstone mixed with oil, jewelry polishes, diamond, etc. The thing to avoid is lapping the actual valve with the seat, as traditional abrasives embed; that's how they work. But you can do this with the Time Saver compounds, and it does indeed work very well.

Whatever works is good, but I think it's a disservice to say that "lapping won't work" and likewise "boring and turning alone won't work." Both have advantages and disadvantages.

 

[stevehuckss396]

but I think it's a disservice to say that "lapping won't work"

Didn't say it won't work. All I'm saying is if you get a good finish on the mating parts it's unnecessary.

 

[stevehuckss396]

steve could you let us know what you used (pictures please) and how you did what you did to get your valves to seal?

it sounds most interesting to make valves that seal without lapping.
i know i have never made a valve that would seal without lapping and i would like to know how cause lapping is messy and time consuming.

thanks
chuck

Here you go! Pages 19 - 22 shows the valve cages and valves being made. You will see the tiny boring bar used to cut the seats. It's razor sharp and left a REAL nice finish. No lapping, no testing, just put them in and started it up.

http://www.homemodelenginemachinist.com/index.php?topic=9609.270

 

[Steve_Withnell]

Hmmm.

I made the Nemett 15s and could not get the valves to seal. The advice was not to lap. I did in desperation, I used "brasso" which is a brass / copper polish in the UK. It did not improve the seal and it did cause circular scoring in the seats (as per advice not to lap).

I found an article from a ex Rolls Royce engineer, whose practice was to cut the seat at two different angles - 30 & 60 degrees. The aim is to get a concentric edge at the boundary of the two angles in the middle of the seat. So I did that (not too big a deal with only two seats to recut) and made I thought, a nice job. Still no compression.

So - (I've posted this previously) I put a small quantity of 20/50 engine oil in the cylinder. Spun the engine up, it fired a couple of times and then a load of compression appeared by magic. Its how my Dad taught me to get the drop on old diesels with low compression that wouldn't start. May be it worked maybe I just got lucky, but that's my story!