PDA

View Full Version : Dead PMDs wanted..



Cowracer
07-16-2003, 11:45
Being an electronics kind of guy, I am real curious about the failure mode on the PMD. I know they fail at a awful rate, but I have never seen anything about what exactly is failing in their internals.

I'd like to have about 10 or so dead PMDs to tear apart and see what specifically is failing, and why. Of course, full details will be posted afterwards.

Unfortunatly, I can't pay for shipping, but if anyone can help by sending me their dead PMD, I would appreciate. Please include as much info as possible such as age of truck, age of PMD, codes set (if any), driveability symptoms, etc

Also, if anyone has a schematic of the PMD's internal circuits, that would help a lot too.

Thanks much!

Tim Streisel
2551 Lombard Ln
Imperial, MO 63052

[ 07-16-2003, 11:56 AM: Message edited by: Cowracer ]

rjwest
07-16-2003, 13:50
I also requested schmatic several months ago
No responce, Must be closely guarded secrect.

Good luck on removing Sealent, I tried melting it
off, grinding, etc, Gave up...

Info, one rebuild shop said they OHM them
to see if Good or Bad ?????
Probably just check Output Tran...

Probably why there are so many repeat fails....

From the Tran part that can be seen, it appears not
to be any Severe duty/ mill spec part that you
would expect in a component in High temp enviroment...

Good Luck....

rustypig
07-16-2003, 18:56
I second the "good luck" about getting them apart in any kind of "clean" manner. I tried once (cause I'm like that) and gave up fairly shortly because it appears they pour in a hard setting kryptonite filler around the internals.

I would be interested to see what you find though!

moondoggie
07-17-2003, 08:33
Good Day!

I had wished to do the same as you're trying to do, having once been an electronics tech.

I brought one to a chemist friend, & without giving him a few hundred $ to actually analyze the material, he suggested that most potting materials are thermoset's (as opposed to thermoplastics). Thermoplastics can be thermally and/or chemically melted; thermosets will burn before they will melt. There are also virtually no chemicals that will melt them in a way we'd like - there are probably things that will attack the potting, but those chemicals would probably melt everything, including what we wanted to see. (Please hold any mistakes in the above against me, not my chemist friend. I showed him my dead FSD a long time ago, & my memory - well, as gram used to say, "Of all the things I've lost, I miss my mind the most.")

I think you may find that there is more than driving the fuel solenoid going on here (see "DTC 35 - Help!). There may be some signal conditioning circuitry that sends the PCM information about when the solenoid opens & closes.

Blessings!

Brian Johnson, #6042

'82 6.2 1500 4X4 1/2T pickup, 4spd man w/ OD, bone stock, 335K+, "In Rust We Trust"
'89 6.2 4X4 1/2T pickup, bone stock, 145K+
'95 6.5TD 2500 4X4 heavy 3/4T pickup, Gear Vendors Aux. OD, 180K +
'95 6.5TD 1500 4X4 3/4T Suburban, Kennedy exhaust, 200K +

Dimsdale
07-17-2003, 11:46
I wonder if they could be split along the plane of the circuit board by "freeze fracturing" them after being dipped in liquid nitrogen? Other than that, and given the above information that the thermosets will die rather than reveal their secrets, maybe some creative grinding with a Dremel tool might reveal the contents in manner similar to what paleontologists use to dig up fossils. This is getting a little to intense though. Any information would be helpful though, particularly when these prone-to-fail parts are so expensive.

Bottom line: you will probably find something incredibly simple and possible to reproduce with $2 worth of parts from Radio Shack.

xseabee
07-17-2003, 12:15
What about x-ray? Would that show the internals without the need to remove whatever their embedded in?

Mark

JMZ
07-17-2003, 16:35
I think there are a lot of 6.5`s out there that never have a pmd failure in years. And some of us keep knocking them out in less than a year. We are in minority. That is why there is not an aftermarket option. :mad:

ucdavis
07-17-2003, 17:51
I've disected two of the gizmo's, cuz I'm like 'that.'
I'll ditto in on the general indestructablity of the potting. Part of the circuit board wants to come with the potting. Possibly a whole lotta patience w/a fine & sharp chisel & small hammer w/light taps could produce a sufficently clean dissection to be of use.
What you can get out of an old fsd tho, is a connector for a remote harness. Place fsd in vice, connector up, & hacksaw vertically down each side of connector, till refusal (you run into SS screw holding circuit bd. to metal base).
Then rechuck fsd in vise flatwise, cranking down tight (w/just the metal bottom plate in the grip of vice). Using cold chisel, attempt vigorously to separate top from metal bottom plate; use sharp chisel & don't spare hammer. You won't get if apart, but will loosen up the connector. After you have nearly taken your vice off the workbench, and made a small mark in the fsd, address your chisel vertically downward along the seam between the connector & the field of the fsd top (the side w/the white stenciled lettering); go a bit lighter here w/the hammer; the blade of chisel will be parallel to long dim. of the fsd. After you get a cut started, you can angle the chisel so force is toward the connector. Eventually connector pops off leaving six short square metal ends that correspond to the six pins in the connector; the ends are at right angle to the pins. It takes a soldering job to put new wire on the ends (length to your desire); use on old harness connector for the other end of the remote loom. This results in a factory (more or less) end at the IP wiring harness that you just banged out of an old fsd, and an old harness end from a prior harness on the fsd-cooler end, with your new wire in between.
Now if I could only learn how to solder.

ucdavis
07-17-2003, 17:53
Or, you could just buy brand new dead FSDs from Stanadyne ;)

JMZ
07-17-2003, 18:09
I think there are a lot of 6.5`s out there that never have a pmd failure in years. And some of us keep knocking them out in less than a year. We are in minority. That is why there is not an aftermarket option. :mad:

rl22
07-18-2003, 06:59
I'd also add from my experience: my PMD was fine for years but started showing classic symptoms several months ago; however, I have always revived the thing by either waiting for it to cool or pouring cold water over it which gets it to working fine again for many days or even weeks. So, if many PMDs are like mine, self correcting, if you tear one apart assuming it is bad, it may show as normal without applying heat to it. What I'd like to find out is if the PMD can be bench tested...see my post of 7/17.

moondoggie
07-18-2003, 07:40
Good Day!

xseabee: X-ray would work, to a degree. We did that once in awhile when I worked at the VA Hospital here. It really only gives shapes, however. Pretty hard to get component values, etc.

I repeat my caution about chemistry: What I wrote was a chemist's opinion, based on visual observation. If I had the time to engineer an FSD replacement, I'd have paid him the considerable money to find out for sure what this stuff is.

ucdavis: Have you ever built bridges or big buildings? I've only heard the term "refusal" used once before, in connection with driving piles for a bridge.

Blessings!

Brian Johnson, #6042

rjwest
07-18-2003, 09:07
I have gotten two dead ones ( free ) from standyne
dealers. Ask they will probably save some for you...

tom.mcinerney
07-28-2003, 23:10
Gents: If do search for "FSD" in 'subject' ('any date') see posts from 11/25/'02, and 11/28/'02; one virtually identical to this. {5/6'03} also good. I believe it was last year about thanksgiving time that JC?GMC65? made a post with a subject similar to 'Tease', wherein he gave an outline of the FSD circuit. Elsewhere he's posted his observation/conclusion that a major failure mode involves the driver-triggering circuit, where connections fatigue fail from thermal expansion/contraction , esp subsequent running-hot, or hot-soak . Previous to this i found myself dreaming of massive igfet/mosfet arrays....One member has observed that some rigs get in a rut where they fry fsds annually/30Kmi. Others go 100Kmi+.
A thorough search for {fsd or pmd} in text might bring up the 'tease' post--then reply to it(maybe copy and re-post it first), then when it's current, we can move a bit forward on this biddie!