NoVector

When I first turn the key on, my PSD pump will do a cycle of:run for 1 second - off for 1 second - run for 1 second - off for 2 seconds - run for 1 second - off for 5 seconds - run for 1 second - off for 10 seconds, etc. This goes on for about a minute with the pump running cycle occurring less and less and after a ~minute, it stops. When I turn the key off and on, the cycle immediately starts over.

I’ve done the PSD flush procedures many times over the past 2 weeks cycling about 2 quarts of Dot-4 brake fluid thru it (and to the slave cylinder) with no improved results. No bubbles at all…

In my thinking, it’s probably not an ABS sensor because it eventually stops. I don't think it's a pressure issue as the cycle starts all over again as soon as I turn the key off and on. No leaks anywhere—to include checking internal slave cylinder. The pressure switch seems to be doing its job as the pump does eventually stop.

Could it be the ABS/PSD controller board? Even used, they’re expensive and hate to buy one unnecessarily. Or, does everyone’s PSD cycle like this??

FredR

What you are describing does not sound correct. The high pressure reservoir operates at somewhere around 160 barg AFAIK and the pressure is controlled by what I perceive to be a two position switch. If the low pressure setting is not satisfied the pump kicks in when you switch on the ignition and runs until the high pressure setting is reached and then switches off. The reservoir should not lose any pressure no matter how long it sits there.

As to why your system is doing what it seemingly does one would probably need special high pressure test gear to monitor what is or is not happening to facilitate analysis. The most obvious candidate for a fault would be the pressure switch or so I would tend to think.

NoVector

Hi Fred - Thank you. My thinking might be flawed, but I'm thinking if I was losing pressure, it wouldn't stop after a minute. Said another way, it either leaks or it doesn't. Like you, I'm leaning towards the pressure switch as the problem seems to be electronic, based on some controller not being happy. As I understand it after reading many rennlist threads, the pressure switch is either: 1) static, 2) low pressure, or 3) high pressure.What hurts my head is, why does it eventually stop after a minute(?) I.e., what controller says after a minute "okay, you win - but next time you turn me off and on, I'm going to do it all over again."

FredR

It could something as simple as a faulty connection or a cable. There is obviously a link between the ignition switch being turned on and the "logic" recognising a demand and presumably the system is fired by bus 15 voltage. i assume that is controlled via the ABS computer but I suppose such could be independent with a simple link/closed loop affair to operate the pressure pump. Assuming control is initiated by the ABS computer telling the motor to start and stop probably helps define the analytical approach needed here. Seems to me you really need an "Alan" input to best define the analytic steps here.

The "smart bit" is how the situation is "analysed" by the PSD control algorithm and a solution applied via the system hydraulics. Contrary to what some seem to think, although ancient age wise, it is still pretty rapid technology with response times in the millisecond region or so I understand.

Alan

The pump relay is controlled by the ABS/PSD brain - I think it likely gives up after a while if its losing pressure consistently, and you should get a warning message in that case of course. The pump takes quite a current - it is supposed to be initial charge and then very low duty cycle intermittent usage only when running. I'd suspect a pressure leak down or pressure switch is faulty.

Alan

NoVector

Thanks Alan - When the pressure switch is happy with the pressure it sees, does it send a signal back to the ABS/PSD board and the board tells the pump to shut off; or, does the pressure switch send a signal to the relay and it kills power to the pump? I saw a previous post of yours to Kevin in Atlanta describing the 5 pressure switch wires, but I don't remember reading whether it's the pressure switch or the ABS/PSD brain that controlled turninig off the pump(?)

PS - It definitely could NOT be the relay, right?? (Jim Bailey's relay - relay - relay is going thru my brain <lol> )

FredR

A leak does not necessarily imply fluid being dumped out of the system- it could be fluid falsely bleeding back to the reservoir if the valve that locks the pressure is passing. However if that were to be the case then logic says the cycle would not stop unless and until some kind of distress was to be detected and then one would expect to see some kind of alarm.

TheoJ

Hello,
it does not sound like a problem with an ABS sensor. With the ABS sensor problem, the PSD keeps activating, making a tock tock tock sound.
The ABS/PSD ecu activates the PSD relay to run the pump and build up pressure until the operating-pressure is reached. Then the relay gets inactive so the pump stops. If the pressure drops the operating-pressure switch closes and the relay will be activated again.
There is also a max pressure switch, which is closed by default and opens when there is too much pressure in the system. This interrupts the relay power so the relay gets deactivated. A safety measure.
Your problem could be caused by any of these things:
- pressure valve leaks, allowing fluid to return into the reservoir (most likely I think, and this is what Fred also indicated)
- operating-pressure switch does not engage normally but opens and closes when close to the pressure level (not simply on/off)
- fluid leaks somewhere (leaking pressure reservoir? I had that once.... )

regards, Theo
1992 928GTS

NoVector

Thanks, Fred and TheoJ - I really do appreciate it as this has me stumped… Today when I have everything connected, the pump did its cycling thing. But, when I disconnected the 2-wire barrel connector between the lock solenoid and the control unit, it acts as it should. I.e., when that connector is disconnected, the pump does not do its cycling thing.

So, I’m thinking “Ah-ha! Fluid is definitely escaping past the lock solenoid and that’s why the pump keeps cycling!” In fact, I think it’s trying to pressurize the entire line up to and including the slave cylinder - and, the pressure switch doesn’t shut off ‘til all of it (to include the slave cylinder) is pressurized. But, if that’s the case, is it because of a faulty lock solenoid or is it something in the ABS/PSD controller telling the lock solenoid to open?

To try and solve that mystery, with ignition on and a volt meter on the 2-wire connector lock solenoid to the controller, there’s no power present. So, I’m interpreting that as whatever my problem is, it’s not the controller telling the valve to open. If it was, I think I’d see ~12v there at that connector.

Which takes me back to the problem, why would the pump run like this only when the 2-wire lock solenoid is connected? It appears nothing is telling the solenoid to open, yet, when the solenoid is disconnected, the unit seems to hold pressure just fine and everything is happy.

I’m not sure if it matters, but when I check the resistance on the lock solenoid, it’s ~3 ohms – which is in range of where it should be according to the WSM.

Taking the Sherlock Holmes alleviate the impossible approach, I think I can eliminate:
- PSD pump as it builds pressure
- PSD pressure switch as it tells the pump to shut off
- ABS/PSD controller & sensors as there’s nothing signaling the 12v to the locking solenoid to open to the slave cylinder
- PSD accumulator as it holds pressure

FredR

Hydraulic systems are non compressible hence why they are used. Porsche needed a passive reservoir of energy so that is why the pressure accumulator has a massive spring and a piston in it. The system is pressurised to a setpoint and when there is a demand from the system the pressure is partially maintained by the spring pushing the piston to compensate. Inevitably the pressure will drop but not instantaneously- thus if there is a containment leak the time to cycle will be controlled by the time taken to deplete- then the pump will kick in again. Thus a cyclic process will be set up and will continue to cycle as long as the ignition switch is turned on even if the motor is not running.

As I understood initially, the system cycles but with an increasing cycle time and then after several cycles it stops cycling. If that is the reality of the case then I am stumped as to what is going on.

Some background info about the PSD system: At the moment I have my PSD diff on the garage floor in bits pending final evaluation as to what to next as the clutch plates appear to be "foobarred" and will be posting in my thread about this shortly.. For anyone reading this- if you a PSD kitted model and have not had the slave cylinder out and dismantled for inspection- do so immediately, change out the seals rings and replace the isolation boot in the casing- also check the drain hole in the slave cylinder. I do not think there is a connection between No Vector's problem and this matter as the system only puts pressure on the slave cylinder when there is a computed demand for such but given my investigation it is clear to me that this part of the system needs intervention on a timely basis and it is not part of any routine service procedure I am aware. Changing the fluid every two or three years is for sure needed but there is more to it than that. Do let me know if there is anything I can check/observe if it will help your analysis in any way.

NoVector

Hi All –

The PSD is now working as it should. The problem was [drumroll, please…] the locking solenoid. Although I saw no stray electricity coming from the controller wire to the solenoid and the solenoid had the correct resistance, something seemed off.

If you look at the wire that comes from the controller and goes to the barrel connector, it appears to be multiple wires. Yet, when you look at the wires on the other side of the barrel connector to the lock solenoid there’s only 2 wires. I guess I wrongfully thought the whole circuit was only 2 wires. Afterall, when we follow the PSD flush instructions we simply temporarily wire the lock solenoid to the battery to open it and allow compressed fluid from the accumulator to travel to the slave cylinder.

At any rate, it occurred to me that the controller is doing more than releasing 12v thru that bundle of wires – and maybe that’s why a bench-test of the system isn’t showing any faults. Yet, when the PSD is actually connected to the car’s controller, the pump continuously cycled like it was building pressure. In the end, that’s exactly what it was doing – building pressure all the way to the slave cylinder!

I swapped the lock solenoid with one that I bought from Herman years ago, and viola – the PSD now runs like it should.

So, if you ever find yourself in a situation like me where the pump seems to continuously build pressure with an on/off cycle AND you don’t have a spare PSD or lock solenoid to swap, disconnect the pipe to the slave cylinder and put a bleeder hose on it. Then, power-up the PSD and see if fluid is flowing out the bleeder hose. If it is, the lock solenoid is likely toast.

FredR

Excellent- well done!