This is an extract from rennlist, well documented by Jim Corenman:
The PSD pumps are impossible to fix, you're better off just sending it to me
and I'll toss it out for you :) :)
OK, seriously: It's just a pump, but the parts are tiny, the clearances are precise, and nothing is replaceable. If you are able to work carefully and not misplace any parts then there is nothing to lose.
You've got the pump off the motor, correct? And you found a drive-disc and a mysterious pump assembly with no visible way to get it apart. Correct?
The thing you are looking at in the pump assembly is a plate. It has small marks where it appears to be staked in place, but it will lift out with a small/strong pick tool. Work carefully and don't bugger it. It is basically held in place by the screen, which in turn is trapped in place by the motor.
This is what it looks like:
The secret stuff is all underneath that plate, and just (carefully, gently)
lifts out. Here are the next steps. Again, let me say that this is not a
beginners job. If you lose a part, you are done. If you don't get things clean,
you'll have to do it again. There is no shame in turning around here.
OK, if you insist on plunging recklessly onward, here is what you see with the plate removed:
The center hub lifts off the (stationary) center shaft:
The balls and "pistons" push out from the center hole with an "L" shaped dental pick (this is a good place for a cheap HF set, which can't scratch anything):
Here are the parts cleaned up and ready for assembly. You will need new
o-rings of course, just match what is there. I used calipers and McMaster Carr.
Did I mention "clean"? This is hi-pressure low-volume hydraulics, and the clearances are really tight. The shadow from a piece of lint will bind things up.
This one was not corroded but had a nasty white crystalline substance everywhere, jamming the pump. So everything came apart for cleaning, including the valve assembly, flush the lines, flush out the accumulator repeatedly, check the slave (was clean), then reassemble and bench-test.
Jim & Sue Corenman
You mentioned bench testing. Can you please walk me through that? I'm guessing you energize the pump for 15-20 seconds and make sure the accumulator holds pressure. And then you bleed it off at the bleeder valve on top of the locking solenoid valve. Or do you also energize the solenoid valve? I bought a spare a couple months ago but haven't played with it yet.
I didn't do anything special, just exercised it a bit before installing. I
first ran the pump with the valve-body bleeder open, maybe 30 seconds or so, to
flush clean brake fluid from the reservoir through the pump. Then I repeated
that with the solenoid valve powered, to flush fluid through the valve. Then I
put a multimeter on the pressure switch and ran it again to build pressure just
until the pressure-switch opened. (You could also connect the motor in series
with the pressure switch, per the factory wiring). I let it sit for a few
minutes, checking for leaks and checking to make sure the switch stayed open. I
didn't want to run the motor without monitoring the pressure somehow, and just
watching the pressure switch was simple.
That all checked out, so I then installed it back in the car and went through the bleeding process to the slave cylinder. Everything was clean there.
Here are some pic's of the crap that I found inside this pump (our '93 GTS which had been sitting for a while). I didn't find anything that would simply dissolve it. I should have paid more attention in chemistry class...
This is where the pressure switch goes, passages in the casting were also
clogged but the switch itself was OK. I took apart as much as I could, and
squirted brake-cleaner through all of the passages.
This is where the accumulator attaches, crud here and also inside of the accumulator (squirt in brake-cleaner, agitate vigorously, drain, repeat until clean).
Jim & Sue Corenman
Worth mentioning....these pumps are very tough to prime and begin the
actual pumping process.....because of the high pressure pump design. The pump
displaces tiny amounts of fluid...which is the only way they could get the
required pressure with that small 12 volt motor.
I've frequently had to add pressure to the top of the reservoir and disconnect the line to the accumulator to get them to even begin pumping. Even doing this, it can take many minutes for the pump to prime. Starting and stopping the motor seems to help more than just continuous running.
To be sure, At what end of the small cylinders do the balls
seat? on the "dark" one or on the "chromed" one?
I believe that you have it correct-- that the two balls seat on the shiny end of the small pistons (see pic below). The shiny end should be a ground/polished surface in order to seal (although it doesn't look that great in this pic... ) while the "dark" end is rough. Hopefully Greg can confirm.
Originally Posted by GTSR
Regarding the plate, It has a cropped sector in the central hole. Does it have a specific position or could it be mounted freely?
On ours, with the pressure switch at 12:00 (straight up), the cutout sector was opposite at approx at 7:00 (see pic). I am not sure it matters but I reassembled in the same orientation. The cutout is there to help fluid to flow into the "intake" side of the pump, but the whole outer circumference is the intake chamber.
Note Greg's comments above about priming, you want to run brake fluid through
the pump before firing up the motor. I just took some tubing and made a siphon
from the can, IV-style.
Jim & Sue Corenman