another Snow engine

[Chipmaster]

Hello Achim, stay with it, don't despair.
Andy

[Rustkolector]

Achim,
You are getting a lot more smoke than I would expect from a 50:1 camp fuel/MMO mix. This could be due to accumulated liquid fuel/oil in the intake manifold which creates a restriction to normal flow and symptoms similar those in your video. I would check the intake manifold for any accumulation. If you find any amount of liquid in there, you might want to build the dry log intake manifold that I mentioned in my "Improved Snow" post to see if it helps.

http://www.modelenginemaker.com/index.php/topic,7090.msg148157.html#msg148157

Also, with this much oil being ingested, the spark plugs will quickly foul. Keep them cleaned and set the spark gap at approx. .020". You are getting close!

Jeff

[zeeprogrammer]

I'd been gone for a while and still catching up.

Very glad to see this. You are so close! No doubt you'll get it. 

[fumopuc]

Andy and Zee (Carl)  thanks for your encouragement, glad to know you are both still following.

Jeff, thanks for all your suggestions and comments I appreciate it very much.
Of course I have followed your post “Improved Snow” with interest already in May.
Main reason for the smoke, should be more my heavy lubrication during the last reassembling.
The very last fuel mixture I have made is 70:1, which is used in the second part of the video.
The intake manifold was empty, when I have done my last visual inspection.
I think next to do, is to get a clear picture about the compression of each cylinder.
I do not own a pressure drop gage similar this Snap On product https://store.snapon.com/Cylinder-Leakage-Tester-Tester-Cylinder-Leakage-P650761.aspx, what in my understanding, would be the right instrument for measurement.
Instead of this I will make a new adapter for the spark plug hole, ready to fit  a small air tube.
I have one compressor with a proper industrial pressure regulator, so a carefully pressurizing of each cylinder should be possible and the instrument for comparison should be my  sense of hearing.
I will let you know the result as soon as possible.

Any further suggestions are always welcome.

[Stuart]

Hi
Regards testing the compression as you outlined that would be a blow by test and is normally done with a air line fitted with a small say 0.5mm restrictor , tee for the pressure gauge and an adapter for the spark plug or injector.

Now for the downside , the situation was a test on a full size Diesel engine v16 6.5 inch bore 9 inch stroke the Cat guys tried to do that very test , you could hear air leaking ,but the test returned result as good from the Cat manual  as I was shift eng for the bank I asked them to repeat the test without the rig on the engine , guess what the gauge read the same , so the test gear was useless.

So take care with the readings they can fool you

A cylinder comp tester is a gauge ( say 300 psi would be ok for you) with a non return valve between the gauge and cylinder do say five comp strokes and take a reading

Great build been following along but I only post if I have something to offer

Before it’s asked we had 6 cat v16 engine for standby power for the computer centre each rated at 1000hp or 750kw continuous rating

[Nick_G]

.
Bit of a random suggestion here.

I feel a little embarrassed at suggesting anything to someone who's skill level is far, far beyond my own. 

OK - The engine the way I see it is basically 2 engines combined. Would it be possible to isolate the 2 halves from each other. (removing parts / blocking inlets etc.) and see if one of the sections can be made to run.? Well sort of run.! as there would be big holes in the combustion strokes (2 not 4)

Just an idea to perhaps isolate the issue to which half of the engine needs a tweak.

Nick

[Roger B]

Hi Achim,

Can you put up a video of the valve gear moving when fully assembled? The ones I can see only show the lower (exhaust?) valve being operated. If you can show the piston rod movement as well even better  Single cam valve systems are very sensitive to valve clearance. I used to have Triumph Dolomite Sprint with this system 

[Rustkolector]

Achim,
I have never used an automotive compression tester or leak down tester on a model engine. They are not accurate due to the relatively small engine cylinder volume you are trying measure on a model engine. Model IC engines are relatively low compression by design, and I find if I have reasonable flywheel bounce back when rotating the crankshaft over compression and uniform bounce back between cylinders, that a model engine will run. The factors influencing this simple test are ring to cylinder fit, ring to piston side clearance, and valve and valve guide sealing. The first two factors are easy enough to accomplish with good machine practice and visual inspection. Valve sealing is more troublesome. In my opinion the best test for leaky model engine valves is an inexpensive hand vacuum pump made for bleeding automotive brake systems. If you can pump it down to 25" Hg and it leaks down slowly, the valve should be fine. If it leaks down quickly, it needs lapping. Always put grease or 90W oil on the valve guide to check only the valve seat seal. Cut very narrow valve seats in the valve cage. .010"-.015" wide is good.
Jeff

[fumopuc]

Hi
Regards testing the compression as you outlined that would be a blow by test and is normally done with a air line fitted with a small say 0.5mm restrictor , tee for the pressure gauge and an adapter for the spark plug or injector.

Now for the downside , the situation was a test on a full size Diesel engine v16 6.5 inch bore 9 inch stroke the Cat guys tried to do that very test , you could hear air leaking ,but the test returned result as good from the Cat manual  as I was shift eng for the bank I asked them to repeat the test without the rig on the engine , guess what the gauge read the same , so the test gear was useless.

So take care with the readings they can fool you

A cylinder comp tester is a gauge ( say 300 psi would be ok for you) with a non return valve between the gauge and cylinder do say five comp strokes and take a reading

Great build been following along but I only post if I have something to offer

Before it’s asked we had 6 cat v16 engine for standby power for the computer centre each rated at 1000hp or 750kw continuous rating

Hi Stuart, thanks for your thoughts and your experience with the CAT guys. The engine analyze process I know from the past is 
•   first you do the compressing test like you mentioned with a pressure gage and a non return valve in the line, because this can be done quick and easy if the gage with a tube and tapered rubber sealing has to pressed into the spark plug thread. If all cylinders behave equal, they are all good or implausible, all bad.
•   If one value is different than the other cylinders, than a leakage test with the above mentioned tester could help to let you know where the leakage is and how much it is.
I can also remember, that it was useful to repeat leakage test up to three times with always cranking the engine between.
After a quick search in the net last night, I have found a Chinese clone of the Snap On tester at ebay for EUR 29,34, so I could not resist to order it.
It was told the tester is already in Germany and will be delivered until end of this week.
If no other experience is known here  about testing a model engine with this leakage tester, so I thought I would be time to get some.
I will let you know about any result soon.

[fumopuc]

.
Bit of a random suggestion here.

I feel a little embarrassed at suggesting anything to someone who's skill level is far, far beyond my own. 

OK - The engine the way I see it is basically 2 engines combined. Would it be possible to isolate the 2 halves from each other. (removing parts / blocking inlets etc.) and see if one of the sections can be made to run.? Well sort of run.! as there would be big holes in the combustion strokes (2 not 4)

Just an idea to perhaps isolate the issue to which half of the engine needs a tweak.

Nick

Hi Nick, first of all I am not sure that we have a different skill level, but I am sure we are sharing the same hobby and this is the most important thing, so many thanks for your suggestion.
Theoretical it should be possible to split the engine in two halves, but it will be difficult to adjust the timing and ignition, so I would like to avoid this effort. It should be possible to detect the failure also without this method. I will let you know about the progress.

[fumopuc]

Hi Achim,

Can you put up a video of the valve gear moving when fully assembled? The ones I can see only show the lower (exhaust?) valve being operated. If you can show the piston rod movement as well even better  Single cam valve systems are very sensitive to valve clearance. I used to have Triumph Dolomite Sprint with this system 

Hi Roger, yes the video with the slowly driven engine shows the exhaust valve only. I will make a video of the full valve motion at the next opportunity.
The adjusted valve clearance here is 0,15 to 0,2 mm, normally you will get some setting after a some run time, but there was no real long run until today.

[fumopuc]

Achim,
I have never used an automotive compression tester or leak down tester on a model engine. They are not accurate due to the relatively small engine cylinder volume you are trying measure on a model engine. Model IC engines are relatively low compression by design, and I find if I have reasonable flywheel bounce back when rotating the crankshaft over compression and uniform bounce back between cylinders, that a model engine will run. The factors influencing this simple test are ring to cylinder fit, ring to piston side clearance, and valve and valve guide sealing. The first two factors are easy enough to accomplish with good machine practice and visual inspection. Valve sealing is more troublesome. In my opinion the best test for leaky model engine valves is an inexpensive hand vacuum pump made for bleeding automotive brake systems. If you can pump it down to 25" Hg and it leaks down slowly, the valve should be fine. If it leaks down quickly, it needs lapping. Always put grease or 90W oil on the valve guide to check only the valve seat seal. Cut very narrow valve seats in the valve cage. .010"-.015" wide is good.
Jeff

Hi Jeff, good to know that you are still following and you are also sharing your thoughts.
With the valves I am nearly sure that this is sorted out, have at look at reply #401, please.
http://www.modelenginemaker.com/index.php/topic,4060.390.html
As mentioned above, I will try to get a better feeling about the relationship between the bouncing flywheel and a possible measured value with a leakage tester in combination with audible blow by, wherever it will be. So I will let you know immediately after this experiment was done.

[Rustkolector]

Achim,
When my Snow was first running, Doug tested my compression using a small compression gauge he made up that had very little internal volume between the engine connection and the gauge. All my cylinders were about 22-24 psi. His were a bit higher, but had more run hours. I was very disappointed. However, when the flywheel was spun by hand, the flywheel would stop on compression and rock backwards noticeably. The engine would start by hand and run well so I never paid much attention to compression readings since. I will agree that this is by no means a definitive test of cylinder compression. It is more a GO - NO GO test for slow speed heavy flywheel engines like the Snow.

I read your post regarding valve sealing. I make my valves by brazing SS heads on what we call "drill rod". The drill rod is quite uniform in diameter and stem sealing is never a problem. The valve faces and valve seats are cut on the lathe using the same 45 degree compound angle setting and a pot chuck to hold the valve. I also bore before reaming the valve guide hole in the same setup. Concentricity is everything when it comes to valves. I have found that a very narrow valve seat ring is much easier to seal than a wide seat.

I usually assemble the valve and valve cage, add the valve spring assembly to hold the valve shut, and snap it open and closed a few times then vacuum test it. My own test standard for valves is to pump down to 25" Hg and time the pressure increase to 20" Hg. If it takes more than 5-7 seconds, it is probably a very good valve seat. I usually go for 10 seconds. Many valves will go for 30 seconds, or more. In all fairness, I think my standard is more than is probably needed for a good running engine, but good sealing valves are a great asset to have when attempting to start a model engine. 
Jeff

[90LX_Notch]

Stay with it Achim.  I look forward to watching it run once you sort it out.


-Bob

[b.lindsey]

It looks like you are very close Achim. Stay with it and you will have it running soon and be doing the happy dance!!

Bill