Dry sump pump issue

[smithdl4]

I have an engine that uses a dry sump oiling system. The oil collection pan has three AN fittings to use for scavenging. Two of the fittings are approximately on the same level, the third is about 1/2 to 3/4 inch higher than the other two.

The system ran for years with no problem using a three stage pump (one pressure and two scavenge stages). The two scavenge stages were connected at times to any two of the three AN fittings on the tank. Recently I tried an upgrade to a four stage pump (one pressure and three scavenge stages). The pumps are belt driven by a toothed belt. The design also has the rear most scavenge section feeding the middle scavenge section which feeds all that plus whatever it pulls from the pan out of a common oil return port back to the tank.

The engine immediately began breaking belts. Pump was returned to the manufacturer twice and they could find nothing wrong. They recommended increasing the size of the common outlet for the scavenge sections from 5/8 inces to 7/8 inches. It still broke the belt. The pump seems to be hydrostatically locking itself.

The only explanation I can come up with is the height difference of the three pickup fittings. I think the higher fitting is not drawing oil and therefore the other two stages are pushing oil into the emptier section of the pump causing it to hydrostatically locking itself itself which stops the pump and breaks the belt. Is this a feasible explanation or am I out in left field? Any help would be appreciated.

Thanks all.

 

[Baluncore]

I assume the pumps are positive displacement pumps.
I would like to see a diagram of the system that breaks belts.
The height and path of the interconnections may be important.
Exactly what style of pump is being used in which places?
Manufacturer, pump models and part numbers?

If two positive displacement pumps are operated in series, driven by the same shaft, then there will need to be allowances made, to prevent a hydraulic lock.

 

[smithdl4]

The pump is basically an industry standard design automotive dry sump pump. Manufacturer is NRC Machining. If you are familar with an automobile oil pump which uses gears then this is just 4 of those on a common driven shaft that mounts outside the engine. One pump section provides pressure the other three suck the oil out of the oil pan and return it to the reservoir. I’ll gather the material you asked for and upload it as soon as I figure that out and get to my PC. Thank you for your help. I really appreciate it.

 

[smithdl4]

Baluncure, Below are the best examples I can find of the system I am dealing with in my query. Do these help?

 

[smithdl4]

Baluncore, First picture is schematic. Second is a duplicate I couldn’t get rid of. Third is a picture of typical pump construction. The forth and fifth show actual examples of connections. Mine are very similar. The difference is location of the return fitting on the pump and height of fittings on the oil pan. The scavenge out on my pump is located as in the 1st picture on the forward most section/stage of the pumps scavenge sections/stages. The scavenge stages are best viewed in pic 4. The difference is the fitting on the oil pan that connects to the forward most section/stage is elevated about 1/2 to 3/4 inches above the other two pan fittings. It means that section/stage might not be sucking as much or any oil as the pump begins operation thus the other two stages are overloading the output side of that section/stage. Could that hydrolock the pump. Thanks again

 

[Baluncore]

I think it comes down to the oil reservoir tank. There are three pumps filling that, but only one emptying it. If it was overfilled with oil, or not vented correctly, it could fill with compressed oil, compressed air, or an oil foam.

Can you operate the system with the top of the reservoir tank open, to see what happens?

 

[Lnewqban]

The design also has the rear most scavenge section feeding the middle scavenge section which feeds all that plus whatever it pulls from the pan out of a common oil return port back to the tank.

Could you please explain this part in more detail?

 

[Baluncore]

Could you please explain this part in more detail?

I took that to mean that all scavenge pumps in the pump block, share a common output cross-gallery, with one port, that feeds all scavenged oil to the reservoir tank.

I would expect a high overflow vent from the reservoir tank, that would feed excess air, with some overflow oil mist, back to the sump. That could be back down one of the scavenge inlet lines.

 

[Lnewqban]

I believe all these systems are controlled and protected by a spring type pressure valve.
If that valve exists, any hydro-locking must be happening between a point upstream that valve and the gear pump.

Could the increased needed torque for the added volume transferred by the new additional pump spinning at the same rate be enough to simply over-stress that unchanged belt?

We don't know the timing terms of failure ("The engine immediately began breaking belts"), or if the pump has been manually tested for free rotation when full of oil.

 

[smithdl4]

Baluncore, You are correct about the cross gallery in the pump block, sort of. The inner most scavenge pump’s outlet is directly into the output side of the middle scavenge pump whose outlet is directly into the output side of the forward most pump whose outlet returns the oil to the tank. So its sort of like you described it but sort of isn’t.
The tank in this system is a 5 gallon tank that is filled about 2/3 full of oil. There is a 5/8” diameter vent on top of the tank. The return line to the tank is designed so that incoming oil enters tangentially to the round tank thereby using centrifugal force to remove air from the incoming oil. The 5/8” vent line is clear/unclogged, and yes I have broken a belt with the tank cap removed so I don’t believe its the tank.
Lnewqban, I have attached a breakdown picture of the typical pump. In the picture item 1 is the drive shaft, item 2 the counter shaft. The Item fives are the gears, item 6 the pump body section. The right most three item 6 sections are the scavenge sections. They are separated by items 8 and 14. The two large vertical holes in the center of theses plates are where items 1 and 2 pass through. The third large hole offset right is the connecting passage between the sections of the pump. Does this help explain the operation?

 

[smithdl4]

Lnewqban, I have tried running the pump with a secondary drive system, an electric drill. It begins operation, but as the pump begins to build pressure it comes to a point where the drill will no longer turn the pump. The pressure gauge however never reaches the normal operating pressure the engine uses. This means the bypass pressure has not been reached. If I stop the drill, the pump slowly, very slowly reverses direction and releases pressure after which it can be spun with the drill for as long as I want, or immediately reinstall the belt and run the engine, and pressure is normal.

When this first happened, I immediately blamed the bypass system on the pressure side of the pump. At the recommendation of one of the pump manufactures, I pinned the internal regulator closed and installed an external regulator. Still broke the belt. A brand new belt I might add. My next thought was perhaps the pump was sucking the suction side connecting lines closed. I installed springs in those hoses to keep them from being sucked closed. Still broke the belt. This led me to my thought that the difference in height of the pan fittings, where two are sucking oil and the third is sucking air might be the problem and as I descibed in an earlier post the two sections sucking oil were inundating the third causing the pump to lock. To be fair, these pumps are designed so that when operating they remove all but a small amount of oil from the pan and would normally be drawing an oil-air mixture from the pan. It is just at start up that this situation with the two pan fittings underoil and the third not that this seems to occur.

 

[Baluncore]

Does the reservoir tank vent air back to the oil pan?
Or is the oil pan being operated at a partial vacuum?
If a vacuum is formed in the oil pan, that pressure is applied to all scavenge pump inlets, requiring a greater torque from the shaft.

 

[smithdl4]

There is a vent from top of the engine that connects to a “T”. The resevoir tank also connects to that “T”. The remaining branch of the T is connected to an empty 1gal overflow tank open to atmosphere. The scavenge side of the pump is supposed to create a very small vacum around the pan fitting area. I don’t think it creates anything large enough to matter the way its plumbed.

 

[Ranger Mike]

We ran a dry sump in our Cosworth Formula Car for ten years. Do you have cavitation problems with the oil? See air bubbles in t he tank?
We blew up many small block Mopars because the oil would foam up bad when it entered the rear of the block oil gallery, passed thru the valve lifters in the galleries and got agitated. Take a can of tomato juice and shake it vigorously. Lot of aid bubbles. We solved this by running a separate oil line to the front of the oil gallery and this slowed the oil running past the lifters. Both oil streams met in the middle to prevent getting agitated.

Pumping Cavitation: A restricted oil suction tube can cause the pump to draw in air instead of oil, leading to cavitation, which can put stress on the pump and drive belt.
We had an oil separator in the oil tank at oil return that separated the bubbles. Do you have this?

 

[smithdl4]

Ranger Mike, No problem with cavitation and the tank is baffled and designed to remove air. The entire system worked flawlessly until the switch from a three stage pump to a four stage pump. The extra stage is connected to a fiiting on the oil pan that is higher than the other two and feeds the front stage of the pump where the return line is connected. My question is will the other two stages pumping oil into the empty, pumping only air front stage cause the pump to hydrolock?

 

[jrmichler]

I have tried running the pump with a secondary drive system, an electric drill. It begins operation, but as the pump begins to build pressure it comes to a point where the drill will no longer turn the pump. The pressure gauge however never reaches the normal operating pressure the engine uses. This means the bypass pressure has not been reached. If I stop the drill, the pump slowly, very slowly reverses direction and releases pressure after which it can be spun with the drill for as long as I want, or immediately reinstall the belt and run the engine, and pressure is normal.

Assuming that the drill has enough power to develop the proper pressure, this points to the problem. There are two places to look when a positive displacement pump overloads. The first place is a blocked discharge, causing excessive pressure. If the blockage is inside the pump before the pressure gauge, the pressure gauge could show zero pressure. A discharge flow blockage could be after any one of the four pumps. The second place is a mechanical binding or blockage inside the pump.

If the pressure gauge is not showing pressure, the problem should be before the pressure gauge. To identify the cause, try disconnecting all hoses and fittings from the pump. Drive the pump with a drill. It should spin freely. Insert the entire pump into a can of oil. Drive it with a drill. If should spin almost as freely because the oil is flowing through the pump with no restrictions.

If it passes these tests, then connect the suction lines from the oil pan, and repeat the drill test. Then connect the line to the dry sump oil tank and repeat the drill test. Then connect the suction line from the dry sump tank to the pump and repeat the drill test. Then connect the line from the pump to the filter, disconnect all other lines from the filter, and repeat the drill test. Then connect the oil cooler lines, and repeat the drill test. Then connect the line from the filter to the engine block, and repeat the drill test. The overload should show up at some point.

 

[smithdl4]

Thank you everyone for your help. My solution for this is to return to my old reliable 3 stage pump, or at least one like it. If I ever find the problem I’ll be sure to come back here and reopen the thread. Again many thanks to all!

 

[Ranger Mike]

I would do the same. You know the 3 pump system and in racing, ifin it ain't broke, dont fix it

 

[tylerkelley1980]

Sounds like a classic case of air entrainment or possibly cavitation if the reservoir isn't feeding the scavenge stage correctly. I've seen similar symptoms when the pickup line wasn't fully primed or had a slight leak. Might be worth checking for aerated oil on return, or verifying that the pump's scavenge-to-pressure ratio matches your setup. Dry sump systems can be super sensitive to plumbing quirks.