Electrovalve diodes

[white exec]

Good bit of discussion recently about failure of suspension electrovalves (Xantia, XM...), and the small diode built into the EV solenoid body.

Drawing from Citroen B7 Technician Training document

Finally got round to taking a failed one, and opening it up.
After unscrewing the two halves of the central hydraulic valve, the steel cover and solenoid brown sleeve were pulled off, and the brown casing sliced around the front face with the bandsaw...

Next to come off was the two-pin connector, which surprisingly revealed four terminals below. Clear that the outer brown plastic case was filled with white epoxy resin (potting compound), encapsulating all the innards - coil, diode, and internal connections.

The centre two brass pins were the externally visible + and - connections.
After a bit of shaving away of brown plastic, clear that the hidden outer pair carried the diode and coil connections.

The diode and coil wire ends were generously soldered to the brass outer terminals. There were some markings on the diode (about 1N4xxx size), but nothing which could readily identify it.
______________

The whole EV assembly is definitely something not meant to be taken apart. The two halves of the central hydraulic shuttle valve are thread-locked together, and require considerable force to break the seal. This gives access to the internal valve parts.

The resin-filled solenoid assembly (coil, diode, terminals) is not for opening up for repair!

Simon has written some useful guidelines for testing these units, without destroying the internal diode (raw 12v must not be connected to them). Additional external diodes will offer protection for the ECU, and enable proper solenoid action where the coil is still working.
______________

Out of curiosity, I wondered how fine the solenoid windings were. Some relays can have pretty hair-like wire.

Cut and sanded down part of the brown casing...

Found the winding to be full length and 3mm thick. Enamelled copper 0.4mm gauge, so surprisingly stout.
Given this, failure of the winding I would think unlikely, unless subjected to burn-out from a continuous 12v.

[electronmirror]

Some interesting picture here! I must track down the notes referred to for non destructive testing.

On a similar theme, I saw on another forum, but cannot for the life of me remember which one, comments which suggested the placement of external diodes similar to E-Crofting kit may be of limited effectiveness. Unfortunately there was no explanation of why that should be the case, my first thought was around the length of wire between diode and coil if the diode is placed in the wiring loom close to the suspension ECU.

I've seen the thread detailing the placement of said diodes in the wiring loom close to the ECU in the engine bay and may have to carry out the same mod myself once I've had a look at the electrovalves with an oscilloscope.

Placing the 'repair' diode as close to the solenoid coil of the electrovalve as possible is clearly desirable to suppress back emf spikes and prevent damage to the ECU but does anyone have direct experience or thoughts on this ?

[CitroJim]

Chris, most interesting Thanks for doing that for our ongoing education

I did not expect the diode to be that large, physically...

Interestingly, I have never known a 1N4000 series diode replacement to fail so I guess the OE encapsulated diode must be either poor quality or marginally speccd, especially in terms of PIV (Peak Inverse Voltage).

Another oddity is that in most cases diodes in rectifier service fail short-circuit. It is relatively unusual for them to fail open.

Placing the 'repair' diode as close to the solenoid coil of the electrovalve as possible is clearly desirable to suppress back emf spikes and prevent damage to the ECU but does anyone have direct experience or thoughts on this ?

This is indeed so... The nearer the better in theory but in practice, the usual location of the repair diodes works perfectly adequately.

[white exec]

I've only handled two EVs with failed diodes: both of them had gone short-circuit.

Agree that the best location for external protection diodes is close to the coil, but not a lot of fun installing there. Maybe an additional male-female 'brown' 2pin "through connector" containing a diode might do it, if there were room, and if connection was reliable. PSA-type push-on connectors don't have a great track record though.

Good few folk have fitted additional diodes at the ECU connector, but - I'm guessing here - as a preventative measure, in case of OE EV diode failure and possible damage to ECU output. Do wonder how many cases there were where the diodes were fitted there to cope with a failed OE EV diode . . . and did it cure it?

I've got a couple fitted at the ECU - 400V 6A, and taken to a good Gnd. At least the wiring from ECU to EVs is a decent gauge, otherwise fitting there might not be such a good idea.

[xantos]

I've done to my Activa something like this and it helped...



You could also put a diode near the connector to the hydractive valves... Here is some good information...
http://xantia007.free.fr/10_regulateur_fiche.php

[white exec]

How was it behaving before you fitted those? Was there a problem, or just fitted as a precaution?
Hopefully they usually have some insulation round them . . .

Important that solenoid protection diodes have an adequate working voltage rating. The reverse 'spike' when the coil switches off can be ten times the operating voltage. So in this case, minimum diode w.v. ought to be 200V (to be safe). A 400V diode costs more-or-less the same tiny sum.

[white exec]

Post above ^^^ edited to include some solenoid winding detail

[xantos]

Yeah of course they are insulated...

Well it was just that the hydractive block was not operational due to the duff diode, so that meant that the car was on "hard" mode all the time. After the modification the valves started to work as they should and the ride was significantly better. I remember that Mandrake (Simon) had quite a few very informational posts about diodes (I thin he made a phD out of it )

Here's one for you: https://www.frenchcarforum.co.uk/forum/ ... 1n#p374821

[white exec]

That's pretty clear confirmation that fitting 'additional' diodes reinstated normal working.

Didn't think for one moment that there would be bare wires there!!!

[Peter.N.]

I have fitted the diodes in the ECU and never had a problem, a relatively easy solution compared to dissecting the solenoid, I had thought about tapping in to the wiring but a bit messy, and knowing me I would probably connect to the wrong wires!

Peter

[white exec]

I only dissected the solenoid out of sheer curiosity, Peter. Had a duff one (shorted) on the bench, so wondered about both the diode (what sort?) and the coil windings (how sturdy?).

You're right - absolutely no real possibility of repair . . . unless you went in for a bit of "blind drilling" to tap into those outer hidden contacts!

Good to find sturdy windings, though. Wonder if there's a reliable way to "blow" a shorted diode open (without harming the coil) so that an external diode could do its job? Sounds like one for you or Jim....

[CitroJim]

Chris, given the sturdiness of the windings, I'd be very tempted to try, if you find another coil with a short-circuit diode, 'blowing' it open-circuit by applying a pulse of very high voltage to it...

I'd have to think of a good source but my first choice would be an old-fashioned 'Megger' insulation tester as used by electricians...

I did think of using the car ignition system but that would be a bit too much!

[xantia_v6]

Silicon planar diodes are very difficult to blow open once they have shorted, you would need to be brutal enough to blow the leads.

On the subject of whether the diode should be located near the coil or the controller, there is quite a lot of dissension on the subject.

My interpretation is that there are 2 different scenarios, the first is where you have something like a relay or A/C clutch which is either on or off, and only switched occasionally. For these applications the diode is best fitted close to the inductor to keep the inductive loop small.

For applications where the inductor is being constantly driven with a PWM signal (such as the electrovalve), the current through the inductor does not drop to zero between switching cycles, and it is better to have the diode at the controller end of the wire, to minimise the current transients through the cable. In either case, it is only the radio interference that suffers.

[CitroJim]

Silicon planar diodes are very difficult to blow open once they have shorted, you would need to be brutal enough to blow the leads.

Yes, agreed but in the case of a shorted one, it's worth a go to see if it can be rescued!

[white exec]

Would be interesting to see a scope trace of the current flowing through a good working EV...
- during the 0.5sec 12v pull in
- during PWM hold in
- at switch off
Is that something that you could do on the bench, Jim?