A/C diode failed...took out a few things...

Scorpio / Scorpio Cosworth Discussions - Questions, problem resolution, general talk, technical tips and modifications.
Post Reply
brokencase
Level 3
Level 3
Posts: 258
Joined: Sun Dec 06, 2015 12:18 pm
Location: PA

A/C diode failed...took out a few things...

Post by brokencase » Thu Jun 21, 2018 7:18 pm

What a pain - Scorpio Bliss interrupted..

About two weeks ago I had noticed what appeared to be EEC-IV resets when I had the A/C running. I kind of suspected the A/C diode so I had turned off the A/C.

Took the car out in the evening a week and a half ago over to a fast food joint. Accidentally turned on the A/C. Noticed on the way back that the turn signals were not working.

Got home and studied the EVTM. Verified that the 15 amp fuse was blown. Changed the fuse and it promptly blew again.

What else is on the circuit? "Ohhhh..my speed sensor adapter circuit to use the Ranger speed sensor is also on that fuse!"
Crawl under car and disconnect it. Install new fuse, turn signals now work.
Hmmm. That thing has been running fine for months. What up?

Next day I start the Scorpio and in runs horribly. Pull codes - There is the "power reset" memory code 72. Nothing else.

Logic coalesces - "I better disconnect that other circuit I made to manage the temperature gauge."
Car now starts and runs OK. "Dang it took that out too.." This circuit has been running fine for almost 2 years!

Clearly I had not put adequate protection in these add-on circuits. But what to do?

Time to go talk to EEC-IV "Chief"...

Image

Ed Lijewski
Level 8
Posts: 7022
Joined: Thu Sep 04, 2003 5:53 pm
Location: The Belly of The Beast

Re: A/C diode failed...took out a few things...

Post by Ed Lijewski » Thu Jun 21, 2018 7:49 pm

Possible faulty ground circuit, probably in the wiring into the ECT circuit for the coolant temp gauge. The ECU doesn't read signals from the VSS or report info from it to anywhere in the car. But the ECU is constantly monitoring the ECT's status. Having the ECT signal serve two purposes instead of a single dedicated one may have overloaded your add-on wiring, or vibration wore through wire insulation naking acshort circuit back into the harness.

Speculation;...but that's what first occurs to me.

[Wondering who the EEC-IV Chief is...]

YMMV
Descartes: "Cogito Ergo Sum"
Lijewski: "Sum Ergo Drive-O. Mucho!

brokencase
Level 3
Level 3
Posts: 258
Joined: Sun Dec 06, 2015 12:18 pm
Location: PA

Re: A/C diode failed...took out a few things...

Post by brokencase » Fri Jun 22, 2018 12:39 am

The EEC-IV "Chief" said I did not have the right stuff for protection of my "add-on" circuits. As you might have suspected, the automotive electrical environment is a harsh mistress. In this case when the A/C diode went out my circuits suffered the back EMF surge of the A/C clutch disengagement,. It's a nasty ~80 volt spike.

But there are worse cases. Like alternator "Load Dump" where the battery is momentarily disconnected. You can search these items in google to find out more.

It was not that I was naive to these conditions, I had used a 15v zener, but it was only a small glass encapsulated variety. So the failure was most likely this part shorting to ground.

In the case of the ECT circuit it was probably a little more involved. The line connecting the little arduino to the ECT is normally a high impeadance input. It went dead short so the EEC-IV read the ECT as high temperature, when in fact the engine was cold. Hence the rough running. A soon as I disconnected it the car started fine.

Ford used what appears to be about a 5 watt zener on their circuits.. It is a decent size device. In addition a 20 Volt MOV (Varistor) is placed in parallel with the zener diode. Either by itself is insufficient. They work together. Unprotected by the zener the MOV would have a limited lifetime. Without the MOV, the zener would suffer on the big voltage spikes.

It looks like this...

Image

So I have rebuilt everything with the added protection. Couldn't repair, I potted the originals in epoxy! I also cannibalized an old power supply for a really tough diode to replace the A/C clutch diode. I am also going to put inline fuses on both of these items. Hate having to do this over again but it's "Live and Learn"...

brokencase
Level 3
Level 3
Posts: 258
Joined: Sun Dec 06, 2015 12:18 pm
Location: PA

Re: A/C diode failed...took out a few things...

Post by brokencase » Sun Jun 24, 2018 9:26 pm

Well...

I rebuilt the speed sensor module with "the chief's protections and installed. All is OK. I changed it around this time. I still potted it in the 1" aluminum square tube. But this time I cut the square tube so there was a tab to use for attachment. I ended up drilling a hole in the gusset that the trans cross member attaches to on the body and used a big 1/4" sheet metal screw to hold the module in place. I also added a 5 amp inline fuse.

I put the new diode for the A/C clutch in by soldering it close to the connections right on the A/C. I left the old diode in place. Took it for a long spin with the A/C running and there were no EEC-IV glitches. So I guess it is OK.

The temperature sensor circuit it a different story. The Arduino Pro Mini boards that I bought off of ebay won't run on 12V like my original that I had originally used and had lasted for 2 years. I think I am going to have to redesign that circuit with a regulator.

I will post the revised schematic for each in their respective threads,.

It really sucks having to do all this work over again but at least it was a learning experience.

Ed Lijewski
Level 8
Posts: 7022
Joined: Thu Sep 04, 2003 5:53 pm
Location: The Belly of The Beast

Re: A/C diode failed...took out a few things...

Post by Ed Lijewski » Mon Jun 25, 2018 7:18 am

I don't fully understand why the original a/c diode failed...normal but uncommon failure, or related to your temp sensor and or VSS adaptations?

YMMV
Descartes: "Cogito Ergo Sum"
Lijewski: "Sum Ergo Drive-O. Mucho!

brokencase
Level 3
Level 3
Posts: 258
Joined: Sun Dec 06, 2015 12:18 pm
Location: PA

Re: A/C diode failed...took out a few things...

Post by brokencase » Mon Jun 25, 2018 7:22 pm

Totally unrelated. My temp gauge circuit had been running fine for 2 years straight. the speed sensor adapter circuit had been running fine for more than 6 months. Neither are associated with the A/C clutch circuit - not even on the same fuse.

What you have to realize is that the protections I added already exist in other modules in the vehicle, like the ECU, ABS computer, dash cluster etc... My add on circuits did not have this protection. This is why they failed. It is not the case where my circuits failed and took out the A/C diode. None the less is is quite possible that other components in the system could have potentially been damaged by the failed A/C diode. It is a bad thing to have happen. You better check yours.

There is a TSB regarding poor connections to the diode. I had cleaned the connections and tested the diode when I first got the car because I was aware of this TSB. The problem is that where the diode is situated it is exposed to the elements. So it may be that the diode is OK but the connections to it are not reliable.

My new diode is SOLDERED across the A/C clutch wires very close to the clutch. Being soldered there is no room for error regarding loose or corroded connections. I left the old diode in place. I don't care what it does or no longer does.

More importantly there are other types of transients that can occur that will also damage unprotected circuits. These events are just as rare as the failed A/C diode case. I now feel much better that the circuit changes I have made will help deal with these cases as well.

Ed Lijewski
Level 8
Posts: 7022
Joined: Thu Sep 04, 2003 5:53 pm
Location: The Belly of The Beast

Re: A/C diode failed...took out a few things...

Post by Ed Lijewski » Mon Jun 25, 2018 8:19 pm

The diode failure in your Scorpio is the only one I've read about, here or on the email lists (IMON and MCA), ever. Had you heard of other failures? Or of any Scorp owners citing the TSB and doing what it suggests to their a/c diodes as PM?

YMMV
Descartes: "Cogito Ergo Sum"
Lijewski: "Sum Ergo Drive-O. Mucho!

brokencase
Level 3
Level 3
Posts: 258
Joined: Sun Dec 06, 2015 12:18 pm
Location: PA

Re: A/C diode failed...took out a few things...

Post by brokencase » Wed Jun 27, 2018 7:51 pm

All I can say was that the connections to my diode were pretty crusty when I first got my car.
I can't be the only one if they had a TSB and rework regarding this.

The only symptoms for the rest of the Scorpio folks is the weird behavior of the EEC-IV resetting. What you notice is the check engine light blinks and the car bucks momentarily. Idle speed goes up. I had to pull over to the side of the road and shut the car off and re-start it to get the idle to drop back to normal. If you pull codes you may or may not get the code 72 which is the EEC-IV noting the power interruption.

In addition to the above, and in my case only, it revealed the weakness in the surge protection of my added on modules for the speed sensor and temp gauge.

Anyhow - here are the relevant docs, decide for yourself.

First the TSB...

Image

Then they did a rework for all the 88 cars (actually built in 87) "M57".
They probably fixed the diode issue in later cars.
Item K below...

Image

Image

Ed Lijewski
Level 8
Posts: 7022
Joined: Thu Sep 04, 2003 5:53 pm
Location: The Belly of The Beast

Re: A/C diode failed...took out a few things...

Post by Ed Lijewski » Wed Jun 27, 2018 8:07 pm

Many thanks. That was really interesting and informative.

I appreciate and value your efforts in describing the event failure, its resolution, and this post script.

YMMV
Descartes: "Cogito Ergo Sum"
Lijewski: "Sum Ergo Drive-O. Mucho!

brokencase
Level 3
Level 3
Posts: 258
Joined: Sun Dec 06, 2015 12:18 pm
Location: PA

Re: A/C diode failed...took out a few things...

Post by brokencase » Wed Jun 27, 2018 9:36 pm

The "M57" fixes are revealing. Busted speedo gear, power door locks cycling, radiator side tanks leaking, etc... The rest of these short comings you probably read about elsewhere and is probably why one would be better off finding a pristine '89 Scorpio!

Otherwise the owners of '88's will just have to come up with their own innovative solutions... I hope I never run into the power door lock cycling issue.

It is a shame about the Scorpio US marketing fiasco. They really were putting the effort in to address all the issues. But you know deep in your soul that there had to have been that conference call between the dealers and the factory ,where the dealers were screaming "These fixes are costing us BIG TIME. Make this go away!"...and so it did...and you know what? I don't think many cars actually got the M57 fixes. I think most dealers and or owners took the hit and flipped the cars. Everybody just bailed out of it.

Ed Lijewski
Level 8
Posts: 7022
Joined: Thu Sep 04, 2003 5:53 pm
Location: The Belly of The Beast

Re: A/C diode failed...took out a few things...

Post by Ed Lijewski » Thu Jun 28, 2018 7:46 am

Do you know after what build date the M57 fixes mght have been made on 1988 models?

(I eperienced the door lock cycling syndrome once...only once, thankfully. A sunny winter morning but only around 25"-40* from home to destination, after driving for 30 minutes and parking, on turning the key in the door lock all locks cycled repeatedly. Fortunately on removing the key from the DS lock, opening and closing that door again, on retrying to lock the car that issue didn't recur, then or since.)

YMMV
Descartes: "Cogito Ergo Sum"
Lijewski: "Sum Ergo Drive-O. Mucho!

Ed Lijewski
Level 8
Posts: 7022
Joined: Thu Sep 04, 2003 5:53 pm
Location: The Belly of The Beast

Re: A/C diode failed...took out a few things...

Post by Ed Lijewski » Sat Jun 30, 2018 12:57 pm

So, I'm still confused about this:

"In this case when the A/C diode went out my circuits suffered the back EMF surge of the A/C clutch disengagement,. It's a nasty ~80 volt spike."

Where do the ~80 volts come from when the alternator limits output voltage to ~max 14.4/5v in a 12v system?

YMMV
Descartes: "Cogito Ergo Sum"
Lijewski: "Sum Ergo Drive-O. Mucho!

brokencase
Level 3
Level 3
Posts: 258
Joined: Sun Dec 06, 2015 12:18 pm
Location: PA

Re: A/C diode failed...took out a few things...

Post by brokencase » Thu Jul 05, 2018 12:40 pm

The back emf voltage spike of the A/C clutch is a well understood issue. It is why they put the diode across the A/C clutch in the first place.

This page shows an interesting chart that is often used to explain the harsh automotive electrical environment...
http://www.analog.com/en/technical-arti ... esign.html

But to best answer your question... The alternator regulator is designed to regulate the voltage output of the alternator based on what voltage is "sees" in the system load. But the inductive spike of the A/C clutch is located some distance from the alternator. Even if the Alternator regulator could manage to regulate this spike, the spike must still propagate to the regulator. In the course of it's travel the spike still exists to blow out anything not designed to handle it.

https://progeny.co.uk/back-emf-suppression/

Post Reply