Spun a bearing
10-18-2010, 07:34 AM
#16
Rennlist Member
Thread Starter
By Cheburator
I checked the factory spec from the GTS tech spec booklet, it quotes for radial play a range of 0.027 to 0.069 mm I went down the middle and converted to imperial and that comes out at less than 1.9 thou. That is tight for a race engine, or 1 thou less than optimal, it doesn't need much to pinch the bearing and game over.
A scenario could be, engine a little hot, oil viscosity breaks down and then metal to metal contact and then trouble.
I checked the factory spec from the GTS tech spec booklet, it quotes for radial play a range of 0.027 to 0.069 mm I went down the middle and converted to imperial and that comes out at less than 1.9 thou. That is tight for a race engine, or 1 thou less than optimal, it doesn't need much to pinch the bearing and game over.
A scenario could be, engine a little hot, oil viscosity breaks down and then metal to metal contact and then trouble.
Fair enough, 11 litres is a lot of oil, Greg Brown said he saw a 928 engine running on a dyno and it emptied the dry sump at certain revs, as soon as the revs came down the tank filled back up, hence my question. This also takes me to the quality of the oil supply to the pump. You got me thinking given the pressure fluctuations.
Maybe the fluctuations have to do with aerated oil, as the revs come down the aeration subsides. In another thread the Porsche 928 late and early pump were measured for capacity and it was stated that the pump (used in your car, i.e after 1980 MY) had a capacity of 3.95 gallons per 1000 rpm. I presumed US gallons were used when I converted to litres.
At 7,000 rpm it would be almost 1 3/4 litres per second, I know you may not be revving that hard, just making a point. Now if the dry sump pump is getting overwhelmed with all this oil the engine crankcase and heads will be getting fuller and fuller with frothy oil. So your oil tank no longer has 11 litres in it, it may only have a few and then it is getting deluged with lots of frothy oil and some of this is coming back to the engine oil pump. The scavenge pump needs a far greater capacity than the oil supply pump due to the mixing of the air and oil.
Maybe the fluctuations have to do with aerated oil, as the revs come down the aeration subsides. In another thread the Porsche 928 late and early pump were measured for capacity and it was stated that the pump (used in your car, i.e after 1980 MY) had a capacity of 3.95 gallons per 1000 rpm. I presumed US gallons were used when I converted to litres.
At 7,000 rpm it would be almost 1 3/4 litres per second, I know you may not be revving that hard, just making a point. Now if the dry sump pump is getting overwhelmed with all this oil the engine crankcase and heads will be getting fuller and fuller with frothy oil. So your oil tank no longer has 11 litres in it, it may only have a few and then it is getting deluged with lots of frothy oil and some of this is coming back to the engine oil pump. The scavenge pump needs a far greater capacity than the oil supply pump due to the mixing of the air and oil.
Have you got any pics to post of the tank and its method of removing the air? Also a dynamic oil level reading would be handy in this case which is the why I raised it initially. Just for additional info some pumps separate the air and oil before sending back to the tank. I would imagine for this very reason.
I doubt there will be metal - if there was I very much doubt that the conrod would still be in one piece and the engine would turn. But I hear you on cleaning everything just to be sure. Perhaps the time to get the VW Motorsport lifters in. Anyone got a PNr for them?
10-18-2010, 07:35 AM
#17
Intermediate
Join Date: Aug 2010
Location: New Zealand
Posts: 43
Likes: 0
Received 0 Likes
on
0 Posts
It really depends on your setup. The standard way is to use a stethoscope approach and identify changes of noise from the region of the pump as you increase the pump speed. Typically needs to be done with special recording equipment though as you're mainly interested in ultrasonic frequencies. This is made somewhat more difficult for engine oil pumps as they're usually bolted on to something much louder ... Removing and running the pump in isolation at various speed/temperature/backpressure would help identify the point at which cavitation happens.
If you're doing a post mortem, you can also do a close inspection of the surfaces on the pump for damage - it'll probably look something like corrosion. I can't recall the exact common points for cavitation on gear pumps - my guess would be the trailing edges of the gears and on the very tips of the gears (and wherever else you might get high-velocity "squirts" when running against a high backpressure). The inlet and outlets would be worth a close look too. Excessive bearing wear (in the pump) would be another symptom, though harder to quantify.
Familiar with boats? Think of sticking a small propellor onto a really powerful engine. Give it a small amount of gas, and it'll still work nicely. Give it enough power and it'll just sit there in it's own little pocket of "air" (actually water vapor) doing next to nothing. Back off the throttle enough and the bubbles all collapse, allowing the propellor/pump to be pumping fluids again instead of vapor.
Sounds exactly like what I would expect from an industrial pump experiencing cavitation. It's generally much easier to cavitate a liquid when it's hot than when it's cold. Dissolved air or bubbles also help.
Of course, I'm still one TB/WP job away from doing my first TB/WP job on a 928, so there's probably a list a mile long of things I'm overlooking
If you're doing a post mortem, you can also do a close inspection of the surfaces on the pump for damage - it'll probably look something like corrosion. I can't recall the exact common points for cavitation on gear pumps - my guess would be the trailing edges of the gears and on the very tips of the gears (and wherever else you might get high-velocity "squirts" when running against a high backpressure). The inlet and outlets would be worth a close look too. Excessive bearing wear (in the pump) would be another symptom, though harder to quantify.
Someone with a bit more knowledge of fluid dynamics should be able to explain why pressure would rise on lift off.
When the engine was cold, pressure always has been at 5 bar as soon as you go past 2rpm. It would only drop off as it got hot
Of course, I'm still one TB/WP job away from doing my first TB/WP job on a 928, so there's probably a list a mile long of things I'm overlooking
10-18-2010, 07:49 AM
#18
Rennlist Member
Sounds like cavitation to me too. What viscosity oil are you running? ... too low a viscosity for the temperature would make cavitation worse at high (track) oil temperatures and/or revs.
10-18-2010, 07:57 AM
#19
Rennlist Member
Thread Starter
If you're doing a post mortem, you can also do a close inspection of the surfaces on the pump for damage - it'll probably look something like corrosion. I can't recall the exact common points for cavitation on gear pumps - my guess would be the trailing edges of the gears and on the very tips of the gears (and wherever else you might get high-velocity "squirts" when running against a high backpressure). The inlet and outlets would be worth a close look too. Excessive bearing wear (in the pump) would be another symptom, though harder to quantify.
Familiar with boats? Think of sticking a small propellor onto a really powerful engine. Give it a small amount of gas, and it'll still work nicely. Give it enough power and it'll just sit there in it's own little pocket of "air" (actually water vapor) doing next to nothing. Back off the throttle enough and the bubbles all collapse, allowing the propellor/pump to be pumping fluids again instead of vapor.
10-18-2010, 08:00 AM
#20
Rennlist Member
Thread Starter
Castrol Edge 10w60 Specially formulated for ///M Engines - works well in my Z4M, which revs to 8000rpm and is built to much tighter tollerances than the GTS engine...
10-20-2010, 12:10 AM
#21
Rennlist Member
Sounds like oil/viscosity isn't likely to be the issue with the oil you're using.
10-20-2010, 01:27 AM
#22
Nordschleife Master
cavitation happens just under zero vacuum.
If you have more vacuum, or less vacuum it will not occur.
I would recommend a centrifugal style of pump at the drysump tank to the stock pump (electrically run), this will produce a very low head pressure to the oil pump and prevent cavitation.
If you have more vacuum, or less vacuum it will not occur.
I would recommend a centrifugal style of pump at the drysump tank to the stock pump (electrically run), this will produce a very low head pressure to the oil pump and prevent cavitation.
10-20-2010, 10:35 AM
#23
Rennlist Member
Thread Starter
cavitation happens just under zero vacuum.
If you have more vacuum, or less vacuum it will not occur.
I would recommend a centrifugal style of pump at the drysump tank to the stock pump (electrically run), this will produce a very low head pressure to the oil pump and prevent cavitation.
If you have more vacuum, or less vacuum it will not occur.
I would recommend a centrifugal style of pump at the drysump tank to the stock pump (electrically run), this will produce a very low head pressure to the oil pump and prevent cavitation.
10-20-2010, 02:03 PM
#24
Nordschleife Master
The length of the line is not the problem.
It is the diameter which creates the resriction.
You can either go from a -10 to a -16 to correct the problem.
But in addition to this you will need to modify the tank with the matching -16 outlet.
The reduction from -16 to the pump inlet will not be a problem if the rest of the pipe is -16
It is the diameter which creates the resriction.
You can either go from a -10 to a -16 to correct the problem.
But in addition to this you will need to modify the tank with the matching -16 outlet.
The reduction from -16 to the pump inlet will not be a problem if the rest of the pipe is -16
10-20-2010, 04:18 PM
#25
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Somedays, I feel like I'm passing on my knowledge to rocks.
The last time Mark Anderson completely tore down his race car and freshened everything was over 10 years ago. His engine is very similar to this engine, in that it uses the stock oil pump to feed oil and uses a twin stage oil pump to scavange the oil, from the pan. He had obtained a different "proven" oil pan/scraper, which visually looked like a much better piece than the original pan. He had me put this pan onto the engine, which we fired up and tested, when the car was completed. Long story short, the oil pan/scraper would not return the oil from the engine at high RPM's. The engine would suck the oil tank dry. This resulted in a rod bearing failure.
Impossible...not hardly.
Alex:
Your post about oil quantity hit me like a brick. You stated that you had 11 liters in the system. That is not enough. You would need about 11-12 liters of oil in the 5 gallon tank, with the engine running. You would need about (depending on your hose sizes, cooler sizes, etc.) 18-20 liters of oil in the system.
Here's the quick math:
The engine when running low/mid rpms, will "hold" 3-4 liters of oil in the heads, windage, etc. When the rpm's increase, windage will increase, and you need to monitor how far your oil tank gets sucked down. A really good pan/scraper will put 2-3 additional liters of oil into the engine. A poor pan/scraper (like the example above) can put 16 more liters into the engine!
Add in your cooler, your filter, your oil lines and it is immediately obvious that with 11 liters of oil in the system, you are going to suck the tank dry at higher rpms (which is what you did).
A few other points, to keep you from doing this multiple times:
1. With the oil tank in the rear of the car, the distance to the pick-up for the oil pump is considerable. You will absolutely need a minimum of a -16 feed hose, from the bottom of the oil tank. If you use Brown/Miller hose, you will probably need -20 hose.
2. Be careful about what hose you use on this -16 hose. Suppliers of hose are now sourcing hose from China, which is generally very, very poor quality and will "suck" itself flat, when the oil pump increases demand. We have seen this dozens of times, on the dyno and have seen dozens of engine failures from cheap hose! It has gotten to the point where this, too, is almost funny (we have a large dyno facility within walking distance and I have a second business, which sells high performance hose. The poor people will walk over with a suction hose from the dyno, asking why their "economy" hose sucks flat when they rev the engine. The unlucky ones are those that put the engine into the car and can't see the hose suck itself flat.) Although you could use a really high quality steel or nylon covered rubber hose, for this application (like Goodridge 200/210 hose), I'd use a very high quality telfon lined hose for this suction line (which will never suck flat...even when hot), like 910 or 811 Goodridge hose. Goodrige is based in the UK.
3. Running a suction line from the rear of a 928 is hard. It starts out low and gets lower. The line should go downhill the entire way and it absolutely can't go above the suction level anywhere!! The "textbook" routing is for the line to gradually loose elevation, the entire length of the distance.
4. Did you use Porsche rod bearings or Glyco replacement bearings?
5. Use only the Porsche bearings! Depending on the size measurements of your new crankshaft (the one you have will probably be junk), you will probably want to use "yellow" bearings on everything except 2 and 6, which will be happier with "red" bearings.
6. Do yourself a favor and use a proven oil that others have used in 928 racing with success. Use the correct weight. I have no idea why you would even pick out a 10W-60 oil. Keep it simple and stay in Porsche's oil recomendations and use a 20w-50. Consider Amsoil (Read Mark Kibort on this forum). Consider Torco (my favorite). Consider Brad Penn.
__________________
greg brown
714 879 9072
GregBBRD@aol.com
Semi-retired, as of Feb 1, 2023.
The days of free technical advice are over.
Free consultations will no longer be available.
Will still be in the shop, isolated and exclusively working on project cars, developmental work and products, engines and transmissions.
Have fun with your 928's people!
greg brown
714 879 9072
GregBBRD@aol.com
Semi-retired, as of Feb 1, 2023.
The days of free technical advice are over.
Free consultations will no longer be available.
Will still be in the shop, isolated and exclusively working on project cars, developmental work and products, engines and transmissions.
Have fun with your 928's people!
10-20-2010, 06:56 PM
#26
Rennlist Member
Thread Starter
I kind of detect a slight note of sarcasm, but I will take on board some of the suggestions. Others have been implemented already.
Thanks for the advice though. Still did not answer my question why oil pressure stayed at 3bar and would only rise on lift off. It seems to me that the biggest problem was the restriction in the system, which would be removed....
The last time Mark Anderson completely tore down his race car and freshened everything was over 10 years ago. His engine is very similar to this engine, in that it uses the stock oil pump to feed oil and uses a twin stage oil pump to scavange the oil, from the pan. He had obtained a different "proven" oil pan/scraper, which visually looked like a much better piece than the original pan. He had me put this pan onto the engine, which we fired up and tested, when the car was completed. Long story short, the oil pan/scraper would not return the oil from the engine at high RPM's. The engine would suck the oil tank dry. This resulted in a rod bearing failure.
The pan is proven in so far that two other cars in the UK have run with it for 3 years without a problem
Impossible...not hardly.
Alex:
Your post about oil quantity hit me like a brick. You stated that you had 11 liters in the system. That is not enough. You would need about 11-12 liters of oil in the 5 gallon tank, with the engine running. You would need about (depending on your hose sizes, cooler sizes, etc.) 18-20 liters of oil in the system.
Here's the quick math:
The engine when running low/mid rpms, will "hold" 3-4 liters of oil in the heads, windage, etc. When the rpm's increase, windage will increase, and you need to monitor how far your oil tank gets sucked down. A really good pan/scraper will put 2-3 additional liters of oil into the engine. A poor pan/scraper (like the example above) can put 16 more liters into the engine!
Add in your cooler, your filter, your oil lines and it is immediately obvious that with 11 liters of oil in the system, you are going to suck the tank dry at higher rpms (which is what you did).
Point taken
A few other points, to keep you from doing this multiple times:
1. With the oil tank in the rear of the car, the distance to the pick-up for the oil pump is considerable. You will absolutely need a minimum of a -16 feed hose, from the bottom of the oil tank. If you use Brown/Miller hose, you will probably need -20 hose.
I think we used 16 hose, unfortunately there is a restriction in it just before the line goes back in the engine
2. Be careful about what hose you use on this -16 hose. Suppliers of hose are now sourcing hose from China, which is generally very, very poor quality and will "suck" itself flat, when the oil pump increases demand. We have seen this dozens of times, on the dyno and have seen dozens of engine failures from cheap hose! It has gotten to the point where this, too, is almost funny (we have a large dyno facility within walking distance and I have a second business, which sells high performance hose. The poor people will walk over with a suction hose from the dyno, asking why their "economy" hose sucks flat when they rev the engine. The unlucky ones are those that put the engine into the car and can't see the hose suck itself flat.) Although you could use a really high quality steel or nylon covered rubber hose, for this application (like Goodridge 200/210 hose), I'd use a very high quality telfon lined hose for this suction line (which will never suck flat...even when hot), like 910 or 811 Goodridge hose. Goodrige is based in the UK.
We use Goodrige SS braided hose
3. Running a suction line from the rear of a 928 is hard. It starts out low and gets lower. The line should go downhill the entire way and it absolutely can't go above the suction level anywhere!! The "textbook" routing is for the line to gradually loose elevation, the entire length of the distance.
Point taken, will check. It is easy enough
4. Did you use Porsche rod bearings or Glyco replacement bearings?
Glyco, which were plastigaged to ensure we were in the middle of the factory spec about 3 times on every journal
5. Use only the Porsche bearings! Depending on the size measurements of your new crankshaft (the one you have will probably be junk), you will probably want to use "yellow" bearings on everything except 2 and 6, which will be happier with "red" bearings.
Doubt the crank will be junk, given that I have seen seized cranks still being usable after a mild polish/grind. Will use Porsche bearings
6. Do yourself a favor and use a proven oil that others have used in 928 racing with success. Use the correct weight. I have no idea why you would even pick out a 10W-60 oil. Keep it simple and stay in Porsche's oil recomendations and use a 20w-50. Consider Amsoil (Read Mark Kibort on this forum). Consider Torco (my favorite). Consider Brad Penn.
If Castrol RS/TWS 10W60 is good enough for the Works BMWs in endurance racing, then it is good enough for me. I picked up 10W for the cold starts in the UK and I picked up 60W for the stupidly hot Summer days at the Nurburgring.
The pan is proven in so far that two other cars in the UK have run with it for 3 years without a problem
Impossible...not hardly.
Alex:
Your post about oil quantity hit me like a brick. You stated that you had 11 liters in the system. That is not enough. You would need about 11-12 liters of oil in the 5 gallon tank, with the engine running. You would need about (depending on your hose sizes, cooler sizes, etc.) 18-20 liters of oil in the system.
Here's the quick math:
The engine when running low/mid rpms, will "hold" 3-4 liters of oil in the heads, windage, etc. When the rpm's increase, windage will increase, and you need to monitor how far your oil tank gets sucked down. A really good pan/scraper will put 2-3 additional liters of oil into the engine. A poor pan/scraper (like the example above) can put 16 more liters into the engine!
Add in your cooler, your filter, your oil lines and it is immediately obvious that with 11 liters of oil in the system, you are going to suck the tank dry at higher rpms (which is what you did).
Point taken
A few other points, to keep you from doing this multiple times:
1. With the oil tank in the rear of the car, the distance to the pick-up for the oil pump is considerable. You will absolutely need a minimum of a -16 feed hose, from the bottom of the oil tank. If you use Brown/Miller hose, you will probably need -20 hose.
I think we used 16 hose, unfortunately there is a restriction in it just before the line goes back in the engine
2. Be careful about what hose you use on this -16 hose. Suppliers of hose are now sourcing hose from China, which is generally very, very poor quality and will "suck" itself flat, when the oil pump increases demand. We have seen this dozens of times, on the dyno and have seen dozens of engine failures from cheap hose! It has gotten to the point where this, too, is almost funny (we have a large dyno facility within walking distance and I have a second business, which sells high performance hose. The poor people will walk over with a suction hose from the dyno, asking why their "economy" hose sucks flat when they rev the engine. The unlucky ones are those that put the engine into the car and can't see the hose suck itself flat.) Although you could use a really high quality steel or nylon covered rubber hose, for this application (like Goodridge 200/210 hose), I'd use a very high quality telfon lined hose for this suction line (which will never suck flat...even when hot), like 910 or 811 Goodridge hose. Goodrige is based in the UK.
We use Goodrige SS braided hose
3. Running a suction line from the rear of a 928 is hard. It starts out low and gets lower. The line should go downhill the entire way and it absolutely can't go above the suction level anywhere!! The "textbook" routing is for the line to gradually loose elevation, the entire length of the distance.
Point taken, will check. It is easy enough
4. Did you use Porsche rod bearings or Glyco replacement bearings?
Glyco, which were plastigaged to ensure we were in the middle of the factory spec about 3 times on every journal
5. Use only the Porsche bearings! Depending on the size measurements of your new crankshaft (the one you have will probably be junk), you will probably want to use "yellow" bearings on everything except 2 and 6, which will be happier with "red" bearings.
Doubt the crank will be junk, given that I have seen seized cranks still being usable after a mild polish/grind. Will use Porsche bearings
6. Do yourself a favor and use a proven oil that others have used in 928 racing with success. Use the correct weight. I have no idea why you would even pick out a 10W-60 oil. Keep it simple and stay in Porsche's oil recomendations and use a 20w-50. Consider Amsoil (Read Mark Kibort on this forum). Consider Torco (my favorite). Consider Brad Penn.
If Castrol RS/TWS 10W60 is good enough for the Works BMWs in endurance racing, then it is good enough for me. I picked up 10W for the cold starts in the UK and I picked up 60W for the stupidly hot Summer days at the Nurburgring.
10-20-2010, 07:21 PM
#27
Rennlist Member
Greg, Perhaps you should write a book: "Race prepping the 928". Mark could stock it, and charge $20.00 or something. Because, generally people listen better to advice that they have to pay for.
10-20-2010, 08:46 PM
#28
Rennlist
Basic Site Sponsor
Basic Site Sponsor
No sarcasm. I just don't understand how people can still be tossing rod bearings out of these engines. We solved all of these problems 10-15 years ago, openly discuss and actively try to help people understand the problems, and yet people keep turning expensive metal into junk.
Seems almost futile, at times.
10-20-2010, 08:52 PM
#29
Rennlist
Basic Site Sponsor
Basic Site Sponsor
10-20-2010, 09:31 PM
#30
Rennlist Member
I agree ... it's all about the pump being deprived of inlet flow ... which can be as a result of a an overall too small a diameter line for the flow rate of the pump, a line which is too long for its diameter (which kills flow rate), or a stricture at some point in the line (which is like valving it down). The line can be as long as you like as long as the diameter is (more than) adequate for the flow, and there are no throttle points of smaller diameter.