N6KB

I have a 1989 Subara GL 1800 SPFI no turbo. The oil presure has dropped and there is tappet noise due to the hydraulic valve lash adjusters not being properly pressurized. This engine has the oil pump on the front of the block, with the planetary gear pump rotors operating inside a cavity in the block. The cavity has worn deeper into the block, so that there is too much free space behind the rotors and very little oil pressure produced by the pump. Chilton manual says to rebuild the pump using over sized rotors. I have not found oversize rotors, nor a complete pump with oversized rotors, for sale anywhere. Replacing with a new standard pump would accomplish nothing, because the wear is in the cavity in the engine block. I can tell there is no significant wear in the rotors, by the machining marks which have not worn off. Where can I purchase an oversize rotor set, or a complete pump with oversize rotors? Or is there another solution, such has an insert shim or spacer to go inside the cavity in the engine block?

qldit

Good Afternoon N6KB, that kind of wear sounds odd, that really shouldn't happen under normal conditions.

If you can't locate new impellers I would be inclined to measure and make a set of shim plates to take up the end float.

A piece of brass shim might be suitable (of correct thickness) but it will take a bit of careful work to make it.

I imagine it would be like a figure eight shape.

Even with pump impeller endfloat loss I would expect that pump should supply sufficient pressure to feed the tappet lifters.

I would be suspicious of the bigends or pressure loss possibly somewhere else.

Others will have different ideas.

Cheers, qldit.

Basementgeek

Depending on where you live, maybe try a heavier weight oil. Maybe a single weight 30W.

BG

N6KB

The car has 263,000 miles with the engine never having been opened up, except the timing belt cover to replace water pump and timing belts. Has original clutch even. Oil pump rotors are hardened (I guess) steel and the engine block, in which the rotors turn, is aluminum. The cavity in the block, where the rotors work, is a cylindrical shape and has two D shaped holes that go to the sump (I think). Cannot just put a shim in there, because it would block the those D shaped holes, which are the pump inlet, I think. I'm using plain SAE 30. Don't need to worry about cold weather here. Once it got down in the 50s (Fahrenheit) and we had to close the windows.

qldit

Good Evening N6KB, for the pump to be so deficient the endfloat must be substantial, have you measured it?

Have you checked the pressure relief system, it is almost as though the spring has collapsed in it or it is stuck.

In view of the mileage and being aluminium I would be considering some kind of epoxy substance, but grinding it clean will be a problem.

I have never struck that kind of problem.

Cheers, qldit.

N6KB

No I did not measure the endfloat. Wish I had, but did not have the right information at the time I had it apart. Next time I open it up I want to be ready, so I can get the job done in one day. If I cannot find a source for oversized rotors, adding a spacer is the way I'll go. I may drill and tap holes in the block to attach a spacer onto that area between to two D shaped holes. The excess space may be as much a 1/4 inch thick, and countersunk screw heads would hold a piece of aluminum or brass in there.

Can you tell me more about operation of the pressure relief system. I cleaned the pump thoroughly using WD-40. I did not remove the nut that holds the spring in. It was very tight and I did not want to break anything. Is the pressure relief system a ball in a seat with a spring holding it there? If there were debris holding the ball/seat open, would this cause the problem? I thought I had it clean, but without undoing that nut, maybe I cannot be really sure.

qldit

Good Afternoon N6KB, I am not familiar with that engine system, but as a general idea the oil pump is usually a gear pump that develops flow, this flow is dependent on engine revolutions and thus varies the available pressure.

As you have correctly determined gear pumps can internally bypass oil with excessive end clearances in the gears.

Now in all systems that I know about there will be a pressure relieve valve usually in the area of the pump. (apart from a self pressuring pump like a wobble plate type)

In some cases it may be a simple ball and spring kind of thing where the oil pressure lifts the ball of a seat dependent on spring tension, and ports overpressure back to the sump.

These relief valves often may be more complicated where a piston type device modulates oil pressure instead of a ball.
The spring is naturally the pressure control basis, and sometimes the spring may deteriorate or the piston may stick in it's travel. Naturally these things only ever stick in their full bypass kind of mode so produce a similar kind of effect as you are experiencing.

On older engines I have sometimes used stronger springs to try to improve oil pressure but that usually had limited success because the problem most often relates to higher bypass in big end bearings etc.

So it may be worth bearing this in mind.

I wouldn't use screws in an area like that, I would consider a locktight type cyoanacrilate adhesive for any shim retension if you go that way.

I can't imagine how wear in a complete oil environment with no need for thrusting would be that excessive as to drop the oil pressure so substantially.

It may be an idea to obtain the manual for that engine or vehicle and have a close look at how that system works.

If you do make a spacer up it will have be the full circumference to be really effective and you may be able to locate it with "tangs" to use part of the existing hole space as anti-rotation lugs kind of thing.
It may be better to consider aluminium if this is the case and work to that idea, but make the profile perfectly, and then rub the piece with "soap" and heat it with a suitable flame to really "brown the soap" (not blacken) before bending the locating tangs, this is simply an annealing measure to lessen breakage for bends in aluminium. it only has temporary "time to work" useage.

Sorry no real help, others will have different ideas.

Cheers, qldit.

oldmn

Even if you could find bigger gears you would have to send the block in to have it machined to clean up the galling/ wear so the new gears would work.
What would likely happen is to clean up the bottom of the hole and then take the same amount off the outside of the case so that it would use normal pump.

N6KB

Cleaning up galling of the bottom of the hole may be necessary, although I am pretty sure the wear is very uniform and may not need to be cleaned up. Taking of any material from the outside of the case would result in the sprocket being offset from the rest of the sprockets that the timing belt runs on. Then I would have to do something to move the sprocket further out on the shaft to get it lined up with the belt. Oversized rotors would solve the problem, if I could find where to get them. Otherwise I need to fill in the space in the block which is now deeper due to 263,000 miles of wear. The difficulty is keeping any kind of spacer from rotating and blocking the D shaped inlet holes in the back of the pump cavity. I worry about using screws, into drilled a tapped holes, yet it seems less likely to fail than epoxy. I am not a machinist and do not have access to a mill or lathe. Whatever I fabricate will be done with hand tools or a electric drill.

qldit

Good Morning Gentlemen, it is cases like this where a decent Vernier Caliper tool with "step" is extremely useful and will allow rapid confirmation of lost material depth.

The interesting thing with these kinds of fixes is the incredible ingenuity that often appears and the simplest tools can achieve marvelous remedial components.

A decent work vice with jaw protection and a selection of fine files is worth having.

A piece of stiff cardbord and slight pressure helps get transfer of orifice info, just use slight pressure over the item.

Use care and the "get it right first time" principle and I can foresee a good fix.

Cheers, qldit.