scavenge pump flow rate

Status
Not open for further replies.
So as u can see on the plot, the lapping helped the old pump considerably,
adding on average close to 40% more flow at pump rpms near 400 corresponding to about 900 engine rpm. But the surprise was how more rpm hardly made any difference. My drill only went up to 1360 rpm, close to 3000 emgine rpm. But the flow didnt increase much. Two things; at the higher rpm the oil from the pump was raelly aerated, really foamey. Also during the higher speed runs only, all together about 2 1/2 teaspoons leaked from between the drive shaft and its bore. At this point i suspect this shaft to bore clearance is allowing too much air to enter and oil to leak out. For my test setup that more less mimics the engine installation, i calculate a couple psi of pressure at idle rpm. U have the foot of pressure head, 2.5” of 3/16” dia passage, and 24” of 0.25” dia hose to pass thru to the upper collector that represens the bikes tank. The governing formula for the two
passage restrictions to flow is: pressure = [8 x lenght of passage x viscosity x flow rate ] divided by [3.14 x passage radius to fourth power]
I did it the hard way and then finally found an online calculator to double check and they agreed well. If anybody is interested i can give a sample calc with all the proper units, etc in another post. Bottom line it looks like my pump has a low threshold where a little pressure developed by higher than idle rpm flows causes it to leak oil and suck air excessively. I need to take a more detailed look at the shaft to bore fit. Enuf for now but stay tuned. Hope i havent raised more questions than i answered. Tom V
 
Apols if slightly off-topic: When I first got my mk3, the amount of froth prevented reliably reading the oil level in the tank until an hour after a ride. I put one of the XS type pcv valves on the breather return and this problem has gone. It wasn't why I put the valve on, but it was a welcome side effect.
 
I just finished reading "Secrets of Speed", by Nicolas Swager, who suggests that in a dry sump oiling system the scavenge pump feed capacity should be about 1.5 times the feed pump capacity.
 
I just finished reading "Secrets of Speed", by Nicolas Swager, who suggests that in a dry sump oiling system the scavenge pump feed capacity should be about 1.5 times the feed pump capacity.
Well, the commando scavenge gears are 0.405” wide and the feed gears are 0.250 wide. Thats close to a ratio of 1.6.
 
At this point i suspect this shaft to bore clearance is allowing too much air to enter and oil to leak out.



AMR will modify your pump body to add O rings to the shafts, here I used X rings instead.


scavenge pump flow rate
 
Last edited:
I thought the leakage was through the gears. How does the o ring help?
Wouldn't a lip seal be better than an o ring in this application?
 
I thought the leakage was through the gears. How does the o ring help?
Wouldn't a lip seal be better than an o ring in this application?
No room for the lip seal but it would be the better solution if there was, but there is not. This is only a solution to the oil migrating from the feed to the return side along the shafts. You still need to remove the side clearance from the gears for the other route through the gears by reducing the width of the pump body, then add a spring and ball to the timing cover to stop oil coming out the feed outlet when not running and there is the other mod to change the route for the excess flow from the OPRV. There is an extensive thread on the AMR mod and all the different elements involved on the forum.
 
Last edited:
No room for the lip seal but it would be the better solution if there was, but there is not. This is only a solution to the oil migrating from the feed to the return side along the shafts. You still need to remove the side clearance from the gears for the other route through the gears by reducing the width of the pump body and there is the other mod to change the route for the excess flow from the OPRV. There is an extensive thread on the AMR mod and all the different elements involved on the forum.
But my leakage is between the drive spindle and the end plate bore on the low pressure scavenge side. And i can feel and see slight radial movement between them, AMRs orings are only on the pressure pump side. Dont think it wud help my problem. btw, i just ordered a new Andover pump today! I intend to bench test it before mounting to the engine.
 
But my leakage is between the drive spindle and the end plate bore on the low pressure scavenge side. And i can feel and see slight radial movement between them, AMRs orings are only on the pressure pump side. Dont think it wud help my problem.
You can also install an oil seal on the drive spindle in the endplate which can eliminate or minimize the shaft / bore leakage . I did this on one pump and will look up the details when I get to the shop.
 
But my leakage is between the drive spindle and the end plate bore on the low pressure scavenge side. And i can feel and see slight radial movement between them, AMRs orings are only on the pressure pump side. Dont think it wud help my problem. btw, i just ordered a new Andover pump today! I intend to bench test it before mounting to the engine.
Again thanks for sharing your work Tom . I too was surprised at how little the volume was changed as rpm in creased - certainly counter intuitive. I wonder if the frothing was /is a key factor in this as what was once an incompressible liquid had become air entrained and thus now is compressible.
I put a new Andover pump on my Commando when rebuilding.
They are supposed to come pre tested so I didn’t do anything in that vain but you are wise to do so . Will you run the same test as this one and check the volume/rpm relationship? That should prove interesting if you do.
Cheers - RT
 
But my leakage is between the drive spindle and the end plate bore on the low pressure scavenge side. And i can feel and see slight radial movement between them, AMRs orings are only on the pressure pump side. Dont think it wud help my problem.

Yes, it probably would because the oil leaks from the feed side to the scavenge along the spindle bore and then out of the pump so it's oil from the feed side of the pump.
 
You can also install an oil seal on the drive spindle in the endplate which can eliminate or minimize the shaft / bore leakage . I did this on one pump and will look up the details when I get to the shop.
This is the seal I used. Eliminated spindle leakage. Tested up to 75psi.
scavenge pump flow rate
 
Yes, it probably would because the oil leaks from the feed side to the scavenge along the spindle bore and then out of the pump so it's oil from the feed side of the pump.
Appreciate your point, and not to beat it to death, but my problem is different. First, my prob is my scavenge pump wasnt keeping up while i was riding. Second, i already had all the AMR mods inclu the two orings on the feed side spindles. So with the feed pressure oil being kept from the scavenge side my scavenge pump apparently was sucking air thru the drv spindle/ bore gap at the suction area of the circumference, and also leaking scavenge oil back out that same gap but at the pressure side of the circumference. So my prob isnt the typical leak back to sump while the bike sits.
I hav the check vlv and orings for that, and the press relief reroute too.
 
So with the feed pressure oil being kept from the scavenge side my scavenge pump apparently was sucking air thru the drv spindle/ bore gap at the suction area of the circumference, and also leaking scavenge oil back out that same gap but at the pressure side of the circumference. So my prob isnt the typical leak back to sump while the bike sits.

Unless the additional timing case oil drain modification has been done then the oil pump spindle ought to be running submerged in oil.
 
Appreciate your point, and not to beat it to death, but my problem is different. First, my prob is my scavenge pump wasnt keeping up while i was riding. Second, i already had all the AMR mods inclu the two orings on the feed side spindles. So with the feed pressure oil being kept from the scavenge side my scavenge pump apparently was sucking air thru the drv spindle/ bore gap at the suction area of the circumference, and also leaking scavenge oil back out that same gap but at the pressure side of the circumference. So my prob isnt the typical leak back to sump while the bike sits.
I hav the check vlv and orings for that, and the press relief reroute too.
Forgive me if I am suggesting the obvious- it is apparent that you are very knowledgeable and diligent . Is it possible that the scavenge oilway from sump to pump is restricted in some way and hence the inability for the scavenge side to keep up ?
I don’t know how far your disassembly has progressed but in order to determine oilway clear path it seems splitting cases will be necessary.
 
Again thanks for sharing your work Tom . I too was surprised at how little the volume was changed as rpm in creased - certainly counter intuitive. I wonder if the frothing was /is a key factor in this as what was once an incompressible liquid had become air entrained and thus now is compressible.
I put a new Andover pump on my Commando when rebuilding.
They are supposed to come pre tested so I didn’t do anything in that vain but you are wise to do so . Will you run the same test as this one and check the volume/rpm relationship? That should prove interesting if you do.
Cheers - RT
Yes, i intend to. TomV
 
Forgive me if I am suggesting the obvious- it is apparent that you are very knowledgeable and diligent . Is it possible that the scavenge oilway from sump to pump is restricted in some way and hence the inability for the scavenge side to keep up ?
I don’t know how far your disassembly has progressed but in order to determine oilway clear path it seems splitting cases will be necessary.
Oh yea, i did the best i could to check from the screened drain plug in the sump thru the drilled passages that include a sharp turn upward before reaching the pump. A 5/16 grub screw on bottom of case just above the frame tube can be removed , if youre lucky, to provide better access to the passages. I found everything clear, but was only able to fish wires and zip ties thru the passages.
Did this before bench test of pump.
 
Unless the additional timing case oil drain modification has been done then the oil pump spindle ought to be running submerged in oil.
Oh! Yea i did overlook that. Well that at least partially blows my theory. What im saying may explain whats happening on my bench test but its different on the bike. I just went to check my case and sure cant find any holes drilled in a lower location.
But even with the pmp submerged in timing case oil a bad drvshft to bore leak would hurt the efficiency of the scav pmp and its ability to pull/lift oil from the sump. I have more work to do on my bench test setup so i can get a better apples to apples comparison of the old pmp woth the new one i have on order.
Thanks for pointing that out to me. I read the thread on the lower return hole and i’ll need to decide about that. Im a pretty easy rider so oil drag doesnt concern me. Nice to get all the oil out when changing oil but again not a big deal in my case. We’ll see. Thks, TomV
 
Im a pretty easy rider so oil drag doesnt concern me. Nice to get all the oil out when changing oil but again not a big deal in my case. We’ll see. Thks, TomV
I did no mods to my oilways and the engine spins up plenty fast enough for this old guy . Even with the old pump it obviously had enough suction to pull a 5/8 “ long piece of piston ring through three 90 degree bends in a 3/16 “ diameter oilway. I have no idea how that is even possible - but it happened.
 
Status
Not open for further replies.
Back
Top