456 Power Steering Trouble Shooting

Power Steering on the 456 does not work at any time.
Steering pump is new and bled properly.

Read this thread with interest: https://www.ferrarichat.com/forum/threads/steering-boost-comes-and-goes.500168/#post-144200836

Would like to know:
When disconnecting the power steering ECU, should one expect power steering assist be full ON or OFF?
When testing voltage @ speeds directly at the connector of the steering rack valve solenoid, is the test for voltage across the 2 terminals or to test terminals independently?

Thank you!

 

Full Off

 

Are you testing the voltage in parallel with the wires on the solenoid valve at the rack or pulling the plug off and checking voltage on the sockets of the solenoid valve plug (at the rack)? Without knowing how the Servotronic unit reacts to unexpected valve solenoid resistances, I would do the testing in parallel with the wires and I would test both wires (red-white and white-black) at the same time. One wire is signal positive, one is the signal earth (negative).

Testing the voltage at speeds? How are you testing the voltage at the rack with the car moving? Wouldn't it be better to do that inside the car? If you've carried out continuity checks between the Servotronic ECU plug and the solenoid, you can assume the wiring is ok.

I'm not 100% sure that the pinout for the Servotronic ECU plug diagram is correct in the 456 workshop manual as the labels on the real Servotronic unit don't seem to match. On the Servotronic ECU, "15" is the standard identification for ignition key volts. This should be connected to a green-white wire. The speed signal wire is white-yellow and should match the "V" pin on the ECU.

The diagram below should show how the car is set up (550 and 456) with repect to the power steering. If there are any variations (in the area of the Servotronics ECU), please let me know! Does the ECU base (plug) have small numbers on it?

ABS/ASR & Power Steering Wiring Diagram

 

The 456M, 550, and 575 have essentially the same hydraulic steering ECU, with minor programming differences. All three can be upgraded with the FHP steering ECU. The 456 cannot, so there is some difference in there from the other ones.

 

Thanks for the clarification, Terry. The workshop manual I was using appears to be for the 456M, so I still believe there is a minor error in the wiring diagrams for that car.

I'll try to remember, in future, to use "456M", when making quotes.

Cheers.

 

Thank you to everyone's replies.

Very helpful diagram! Looks like "15" supplies power to the ECU, which intercepts the speed signal "V" and regulates output voltage to the solenoid in order to vary hydraulic pressure.

As noted, 2 wires @ solenoid on Rack, one is Pos other is Neg. The solenoid itself looks like a series circuit.

Yes plan was to hook up voltage meter via wires to the Rack side of the harness, with the solenoid unplugged. I suppose that would be easier to do directly at the ECU. (Continuity may not rule out shorts upstream). Then drive at speeds and see if voltage varies with in range as indicated by link above. If it does, I'll assume the solenoid is faulty and vice versa, should be the ECU.

Will post back on test result and revert on pin out designations on connector side of ECU.

ECU's look different but imagine they work i much the same way?



Thanks Again!
-

 

Interesting. I wasn't expecting the ECUs to be so different. At least they are clearly labelled. Is it difficult to attach multimeter leads to those type plugs?

I wish they wouldn't use the word "tacho", "tachimetrico" and "tachofrequenz". It's definitely a speed signal, not engine rpm.

 

Would be pretty weird if the steering feel changed during the run-up in each gear.

 

Wow Im facing the same issues. I have bought a new pump and now a new filter. Not installed yet. However not looked at the electronic side of this. Will investigate further.

 

As some of you know, I provide reprogrammed steering ECU's for the 456M/550/575M cars, but have had to turn away 456GT owners due to the different ECU plug configuration, and possible different output, etc.

However, if there are any 456GT owners who have a bad steering ECU, and would be willing to send one to me, I'd like to do some research on it with an eye to reprogramming it, or upgrading the circuit to use the later ECU's.

If there's any interest in doing this, please PM me.

 

Continuity from Harness @ Rack to Servotronic ECU Harness side - 2 pin.
Voltage 0 across terminals - Harness @ Rack - with car at speeds.
Power/Ground present @ 3 pin Harness to Servotronic ECU.

Seems like the ECU is the issue. Cribbj - Just PM'd.

Thanks!

 

Unfortunately a new factory Steering ECU did not remedy the problem. Scratching my head, thanks everyone!

• There is voltage @ the servotronic when Cycling the ignition switch, but drops to 0V after, and remains 0V even at speed.

The speedometer is working. Could a break in the speed signal to the servotronic ECU cause a loss of power steering?

• Where is the best place to probe the speed signal and what measurement should be expected?

Should there some be voltage at the servotronic, even under minimum power steering assist conditions (such as high speeds)?

 

What a pain. Do you mean there are 0 volts coming out of the Servotronic module to the steering rack on the red/white wire (and white-black wire)? i.e. on the two pin plug.
Or are you measuring the output volts at the Servotronic module using one pin and a local earth?

I'd say so. The system is probably designed to default to high speed/low assist whenever failures occur.

The Digitek unit in the instrument binnacle provides speed signals to various components on 2 different Digitek plug pins. One for the speedo, one for the other systems (Bilstein Suspension ECU, Servotronics, etc). The latter output uses splices to send the signals to different systems. If your suspension light isn't illuminated and your speedo ok, then the chances are that the Digitek unit is ok. If you're monitoring the speed signal (and your suspension system is ok), then the best place to do that is after the splices. Since the location of these is unkown, then you have to do it at the Servotronics unit.

If the Servotronics output signal is failing (and it's being measured at the Servotronics module), there can be other causes such as a Servotronics power failure, loss of earth and perhaps even a short in the steering rack. There is also a possiblity that the new part is faulty (may have been sitting on the shelves too long).
I'd say the Servotronics module is still the best place to carry out voltage checks.

Regarding power (on the green wire) to the Servotronics: The Bilstein Suspension Unit uses the same power, so it's unlikely there is a power failure (unless there's a problem with splices, or pins on the Servotronic connectors).
Regarding earths. I'm not sure where the earth is for the Servotronics unit. Can you follow the black wire?

There is a graph in the 456 and 55o manuals showing speed vs electrical current. Voltage is not shown. Max current is 60milliAmps at 15~20 mph (or less) and drops down to 0~50mA at 155 mph.

 

Qavion - Thank you for your reply.

Yes via test harness connected output of module to servotronics solenoid at the harness connector rack side. Measuring voltage across the two wires {red/white wire (and white-black wire)} with leads into cockpit, connected to multi meter measuring i volts .

Testing is conducted with the servotronics solenoid connector unplugged.

Brief reaction in voltage when the cycling ignition but no voltage after, neither at idle nor at speeds. Wires {red/white wire (and white-black wire)} between Servotronics Module and Servotronics Solenoid have continuity.

Yes safety feature that would make sense.

Is the Digitek unit located on the back of the speedo?

The speed signal according to diagram from link above is White/Yellow. It does look like it goes through another splice/system ("201" from wiring diagram).

The suspension light does come on/off intermittently but not often, may be 1/7 times the car is started/driven. No power steering even when dash warning light is off.

Speedo is OK.

I can try to measure at the servotronics module white yellow first. What type of signal would this be? Should be a wave form 0~2.5V? Not sure if my multimeter's sample rate can provide a reliable result...

Verified power & ground in harness going to servotronics ECU.

The old/new servotronics units yield same results - as far as voltage across terminals.
The test harness is testing everything upstream from the rack solenoid so if there was a short in the rack, it shouldn't affect things.

With Key on switched 12V

• green wire provides 12v
• ground continuity to chassis.

Did not do this test with engine running. Will double confirm this and revert.

I shall try to measure across terminals in mA and report back.

Thank you!

 

Thanks for the update

I can't say if this is good or not. Voltmeters present a high resistance to a circuit and if the ECU can't get enough current flowing to what it thinks is the solenoid, it might decide to stop sending signals (?)

I wonder if anyone has created a test box to feed an artificial voltage to the solenoid to check the power steering without the ECU being plugged in? We know the typical currents, if we know the resistance of the solenoid, we should be able to figure out what voltage to feed to the solenoid. A cheap potentiometer and a power supply may be able to generate some typical ECU output signals.

Not really sure. I'm sure someone on the forum should know. Best I can do is show a diagram of the plugs going to the instruments:

Image Unavailable, Please Login

The Digitek plugs are 33F, 34F and 35F. For continuity checks between the instrument panel and the Servotronic ECU, you may be able to get better access to the wiring via the instrument panel plugs 36F, 37F and 38F. The specialists may have breakout boxes to read the signals with the plugs connected (or they may just probe the backs of the plugs). I think it's probably a better idea to just check the signal at the Servotronic ECU plug.

I assume the speed signal is similar to the signal going to the speedo, but the frequency may be divided by half (This is done on the F355). I can't recall if the 456 workshop manual shows a typical waveform for the signal going to the speedo? I bought one of those pocket digital oscilloscopes for stuff like this, but I can't say I've got my money's worth out of it yet.

https://www.adafruit.com/product/468

Don't forget that current has to be measured in-line. I think you'd have to hook up the ECU to the solenoid with a couple of wires (and put the meter inline with one of these wires).

Image Unavailable, Please Login

Good luck!

 

Qavion - Thank you for your advice. Finally getting around to spending time on this issue.

Retested inline as in diagram with multi meter spliced into one of the wires. Is it correct to assume that it does not matter which wire as both are part of same circuit?
- Result, No mA nor DCV.

This quote is from user Mink from the link in original post : I took Brian's advice above and measured the voltage at the rack at various speeds. It varied between 4 and 8 volts depending on speed and was very consistent. This led me to read up on how servotronic steering works and I decided to look at the Servotronic torque converter on the rack. My research found the part number on it to be the same as many 1990s BMWs. I replaced it and et voilà, I have power steering again!

Some online posts from BMW owners providing the solenoid with 12V to regain power steering. there's mention of utilizing a 9V battery to test the solenoid.

BMW mechanic said they used to bypass the ECU by plugging the 2 ends of the harness together to give 12V directly to the servotronic solenoid... tried that but no change in steering response.


Image Unavailable, Please Login

 

These two results are hard to explain unless the solenoid is shorted, but you would expect smoke to be coming from the solenoid if this was the case

I think I would have gone for a small 9 volts battery as a first step. That way, you limit the current going to the solenoid. Have you measured the resistance of the solenoid?

 

I believe the Steering ECU outputs a PWM signal, rather than a variable DC to the Servotronics spool valve, however I don't know the magnitude of the PWM pulses.

I would "never" connect a constant +12 to that valve.....

Someone with a scope could easily confirm whether it's PWM or not.

 

It's Asynchronous PWM as proven by Dale O'Harra with his scope when I had a quarter bet with both Dale and John, both of whom are EE's and I'm just a grease monkey but I won the bet !!

 

Dave, your memory is working much better than mine.

I was pretty sure it was PWM but couldn't remember why I thought so, and it was because I lost a quarter!

 

I got a mind like a steel trap.........................that's been left out in the rain for years and years.

 

So the inability of DrivenHK to detect an ouput from the controller may depend on the type of meter and what settings he is using. But, assuming the correct polarity of voltage was applied to the valve, wouldn't a fixed dc voltage simply keep the valve open, providing maximum power assistance? In effect, a 100% duty cycle?

 

I know the feeling only too well, my friend......

One would think so, but we would need to determine what voltage can be safely applied without toasting the coil on the little spool valve. I could be wrong, but I don't think a continuous 12v is safe for this control valve that is accustomed to seeing a PWM signal, in fact, looking at the tag on the valve in the illustration above, it appears to say "6v max, 0.8A 100% ED"

 

Your power steering doesn’t need a speed signal to activate the rack. At idle, or revs, and the vehicle stationary ie: no speed signal, power steering is at max assist. Are you sure your rack isn’t buggered? You said you fitted a new pump, and they pretty much bleed themselves, so it’s not leaving much else.

 

I'd try replacing that little spool valve or torque converter as BMW calls it

If you've replace the steering ECU and confirmed that it's getting power from the electrical system, then the only things left that could be faulty are the wiring from the ECU to the torque converter, and the torque converter itself. The rack itself is mechanical, and you've said you can still steer the car OK, so it's obviously not locked up - you're just not getting any power assist.