As i am waiting for my replacement PMD to arrive, this is very informative and looks like I will be keeping and rebuilding. Thank You for this very detailed video. ,
Good video. I was a control engineer and knew you could repair the transistors but never knew the details of the how to do it. You have covered it well. On my 6.5L 2001 van I bought from the Air Force I found the PMD to be without a transistor something most 6.5 owners would not think to check. I did not run it very long that way and it has been running well for the last 3500 miles. I also used your video on how to set the injector pump timing after I had the pump rebuilt it really helped. The van is very different since you actually cannot see the timing marks and actually have to use a mirror. It took me awhile to do it all working in 20deg Wisconsin weather but I did get it done finally. Thanks again for your efforts.
I really appreciate your videos, I am not in the electrical world but hate simply throwing something away that people used to test and repair! I recently tried my hand at this type of repair, a soldered in fuel pump relay in a fuse bus. I was very happy to pull it off for my client and admit I felt a bit like a bull in a China shop with my tools for the job. I used a Weller soldering gun and a compressed air blaster ( not sure what they are called right now) to get the solder off the pins. It worked!
I was just about to throw out a couple of bad ones I had, after watching your video I ordered some transistors and hope to repair them. Thanks for the in depth and informative videos you produce!
Nice job on the content. You covered everything well. My PMD has just started playing up (replaced 12 years ago, moved to behind the grill), so I'm waiting on a new new one. Once it turns up I'l pull the transistors from the dying one and test 'em to see if they're worth replacing.
Watched this a couple of times, having an intermittent problem with mine (2000 3500hd) and moved the PMD. Might try this to build a back up. Just wondered why you remove the screw head potting? Seems this is unnecessary- just remove potting on the solder contacts, and remove nuts on the transistor side, desolder, and install new. Also, did you test the old transistors? And do these fail intermittently leading up to a permanent failure or is there some other possible failure mode like a cold solder joint or broken circuit board connection?
Part of the failure analysis across some forums in the day was loss of fastener torque, leading to heat sink failure which kills the transistor internally. So you have to do more than what you suggest in order to completely remove both of the devices for replacement (and new heat sink materials, new fastener adhesive).
My experience of checking several of these PMD/FSD is that these power transistors (PT) are probably not the cause of stalling problems. None of the 10 I have checked had faults/breakdowns with these transistors. If you remove more of this epoxy, you will see some SMD on the printed circuit board. I suspect that it is in connection with these SMDs and their thin wiring paths that the errors occur. These components are much more sensitive to the large temperature variations than the powerful and well-built power transistors, which then create most of the heat. I have also tried taking out the 2 (PT) and putting them outside the PMD (on a separate cooling jacket), only connected to the PMD with wires. As a result, so much heat is not transferred. Now this is a rather impractical solution in relation to the installation of PMD, but a possible solution. By the way, I used a Dremel to remove the epoxy (with a light and steady hand) to avoid adding strong heat to the wiring paths and SMDs. And if you want to protect yourself from gases, you need something better than a good facemask, one with an ex charcoal filter.... I probably wonder if the constructions on these PMD boxes are actually physically well enough designed to also handle the large temperature fluctuations that occur. Their surface area should perhaps be larger as the heat from the 2 PTs will be concentrated and will affect the other and more sensitive components in the box to a large extent. And this epoxy, I think it is a bit unfavorable in terms of not releasing the heat. It will probably be able to protect the components, but in relation to heat, it may not be so favorable. Good Luck...
That was very informative. I hope you will forgive me for some criticisms. I would expect transistors bought from ebay or amazon to be counterfeit. Mouser and Digikey both have this part, and Digikey will provide free shipping if you prepay by check. Finned heat sinks are completely dependent on the velocity of air flow thru the fins. If the PMD were to be mounted so that the length of the heat sinks fins are parallel to the oncoming airflow when the truck is moving it would be more effective. So instead of being mounted behind the solid part of the bumper, mount it in the hole you are reaching into so air will flow thru (along the length of) the fins. If you have a thermocouple temperature meter with 2 input channels, you can measure the temp of the heat sink and the temp of the PMD case. This will tell you if there is poor thermal conductivity between the heat sink and the PMD case. I like to use the white aavid thermal grease instead of thermal pads.
Thanks. I would characterize heat sink location and airflow concerns as a secondary, or even tertiary factor for this particular PMD design. Airflow alone is not what primarily affects the transistors lifespan (MTBF) here. They actually fall victim to two other factors regarding heat dissipation, at least with the grey Stanadyne model. It's not lack of _cooling in motion_ alone that will result in premature failure. It's when the vehicle is at rest that we have to look primarily, at how to dissipate the heat. The _post drive_ heat soak, after the engine is off and the vehicle is no longer in motion, and then the repeated heat-cold cycling that follows normal vehicle usage is where the root causes appear. After over a decade of analyzing failures, including purchasing DOA PMDs off of eBay, I have a good idea of the ways they fail that can be repaired (some small % also fail in non-repairable ways too). Various combinations of lead-free solder joint fracturing at the TO-3 transistor lead(s), loss of torque at the TO-3 fastener(s), and failure of the transistor die itself internally due to repeated excessive thermal exposures all cut the lifespan down considerably. Typically component failure of the PMD sets in around the third year or so of use. Once the vehicle is in the garage after a drive, no PMD mounting location will save the PMD as there is no airflow involved. Nor any fuel flow for the factory IP mount point. The PMD will fail over a 3 - 4 year period of time due to the _vehicle at rest_ factors I mention without improving the heat dissipation at rest. Over the next few years I hope to find that the use of _leaded_ solder, superior thermal pads at both the transistor (as shown in the video), and at the heat sink mounting (not shown), as well as superior fastener torque protection can extend the designs lifespan (MTBF). Improving the thermal dissipation is the only lever we can tweak when the vehicle is at rest. So the better pads and better contact with the steel of the bumper, or frame is the focus. Only time will tell over the next few years whether these matter or if the fundamental flaws are simply too great to overcome to get 10 years of life out of one. With Stanadyne production having ceased in 2015, the day is fast approaching where rebuilding a grey PMD is going to be the only source for a working one. Hence this project that I took he time to share with you all.
@@DrShock Thank you for your detailed reply. My background is cooling of electronic equipment. Bellville washers under the nuts holding down the transistors will maintain clamping force. A fan directly on the heatsink as shown in a product on Amazon called " AAVID CPU Cooling Fan & HEATSINK, 50mm Sq. Fan, 50mm X 52mm X 43mm Silver Heatsink & Fan, TX3 3PIN Connector " will provide cooling when the truck is not moving but it and the pmd would have to be mounted inside the cab for weather protection. The power to the fan would be controlled by measuring pmd temp.. Finally, the transistor leads could be extended out of the back of the pmd case and mounted on two new heatsinks. These heatsinks would be electrically insulated from each other and the cab of the truck. In this way thermal grease can be used instead of an electrically insulating thermal pad.
Great Video, what if one wanted to build their own extension harness .. do you know the type of connectors needed to buy from an outfit like Digikey same with the heat sink. Thanks!
I don't know the Aptiv part numbers for those, but would have to presume they are available as the extension harnesses do exist. I would expect though that the cost of purchasing only two connectors, plus the pins and crimping tool, plus shipping, would mean it's cheaper to purchase an aftermarket extension harness already assembled and tested. The heat sink though could likely be sourced cheaper than pre-made, though you would have to drill and tap the aluminum for the PMD special torx screws and sand smooth for the thermal mounting pad.
@@DrShock OK ... didn't know if you knew the connector number or style .. I'll do a little more research on it and may post a video on making one up. As far as having the the tools I do have those already and the auto wire. From what I have priced for the kit .. heat sink and harness I thought was a bit pricey compared to doing your make up. Thanks again on a great video ...
Thank you very much for your videos I just happen to have four dead PMD’s sitting right here that I’ve already started to take apart after watching your videos I have started working on dismantling number three, and I’m gonna be ordering transistors for them..The transistors are definitely testing bad The one PMD was interesting, with sort of a translucent jelly type stuff, some kind of heatsink I’m sure? . Do you have a test procedure for the PMD it’s self while it’s still on the vehicle? other than just knowing that it’s not starting? Maybe a way to connect to it while it’s plugged in? I am liking all of your 6.5 videos👍🏻 Cheers
Thanks. The potting material you describe, the _translucent jelly_ type sounds like an aftermarket PMD like FlightSystems or D-Tech. Stanadyne never used that type of potting material. Neither GM nor Stanadyne published any sort of complete PMD test. There are some basic voltage checks in a Stanadyne service bulletin, but those are just completely dead or not type tests. They don't find anything in between.
@@DrShock Thank you, for getting back to me so fast, I just search the model number on it and it came back as “Dorman” it is a-904-104 I remember it not lasting very long and it looks a little bit different on the inside, that jelly stuff looks very toxic handling it very carefully! The stock potting material is a lot different, The first one was a good learning experience to whittle a way at even though I think I did OK, but now I know where things are located can avoid some areas, I was hoping that this day would come along where I can repair these seemed like such a waste just to chuck them, I guess that’s why I held onto them for so long. The same with the truck 2000 gmc 3500 4x4 dually, it’s been my daily work truck for the last 12 years been very reliable hauling huge loads all the time! I also have a 1993 1 ton 4x4 same truck still going strong right now it’s the running one, lol anyhow cheers and thanks again for keeping these motors alive! 👍🏻
Thanks. That Dorman would be just throwing their name on a D-Tech type (has a notch in the top black plastic casting). Yeah that material is definitely not anything you want to breath in fumes of if heating, definitely a chemical mask op.
That's pretty bad that you are going through PMDs every 3 years. My dad's truck (95) had the pump replaced once under warranty when the PMD failed, and then it failed again one time after the warranty was up, and he moved it to the front behind the license plate then and it's never been an issue since. Seeing as these are just regular BJT transistors and nothing really special, I'd look for some different parts that have higher current ratings (I'm not showing much doing a quick search at mouser though). Do you put thermal paste between the back of the PMD and the heatsink when you mount it? I'd at least use some regular white thermal paste (or you can also use penaten diaper cream - it's basically the same stuff - zinc oxide paste - I've actually used penaten diaper cream on a CPU heatsink when I couldn't find my jar of thermal paste - never had an issue and that machine was my main computer for many years after)
also it would be interesting to completely de-pot one and look at the circuit design to see if there's something that could be improved to prevent the transistor failure - it's like there's some flyback diodes or something missing from the design...
The ones failing here are not my rebuilt ones, but rather the factory original Stanadyne ones. They are discontinued now, so Stanadyne no longer cares to figure out the remaining design issues. I do not expect my rebuilt ones to fail like the originals. ;-) I also am running a Standadyne injection pump replaced once under GM extended warranty back in 2004 or so. It's only the PMDs that have a very abbreviated lifespan.
I used the black one until the gray came out. Same frequency of failure replacement. Part of this is likely the application, the diesel 4WD Suburban is the most cramped/high heat layout of all the C/K trucks imo, even behind the bumper. The other caveats are that this has been a south Florida truck, daily highway driver baking in the scorching sun never to experience a garage or winter until recently. It's climate where that bumper easily gets hot enough to fry an egg every day. It's still on the road 7 days a week, for multiple hours each day where it leads a life more like a delivery van than a typical 6.5L truck. I'm sure some or all of these factors contribute to aggravating the PMD design flaws more so than the average truck. Hence my side project to rebuild these and improve the parts and materials used in the design to see if I can get a longer PMD lifespan with this same vehicle.
I where do the wires? From that harness go to on the motor I had a shop saying, they repaired mine. Put a new one in it and somehow, the truck was running when I got home it wouldn't start again when I pulled the harness. It was cut completely all the wires and 2 places and they take it back in there. So I couldn't see it underneath the battery. So now I gotta run a new harness. And I'm not sure where it goes. From the p m d box to the motor
As i am waiting for my replacement PMD to arrive, this is very informative and looks like I will be keeping and rebuilding. Thank You for this very detailed video.
,
Good video. I was a control engineer and knew you could repair the transistors but never knew the details of the how to do it. You have covered it well. On my 6.5L 2001 van I bought from the Air Force I found the PMD to be without a transistor something most 6.5 owners would not think to check. I did not run it very long that way and it has been running well for the last 3500 miles. I also used your video on how to set the injector pump timing after I had the pump rebuilt it really helped. The van is very different since you actually cannot see the timing marks and actually have to use a mirror. It took me awhile to do it all working in 20deg Wisconsin weather but I did get it done finally. Thanks again for your efforts.
I really appreciate your videos, I am not in the electrical world but hate simply throwing something away that people used to test and repair!
I recently tried my hand at this type of repair, a soldered in fuel pump relay in a fuse bus. I was very happy to pull it off for my client and admit I felt a bit like a bull in a China shop with my tools for the job. I used a Weller soldering gun and a compressed air blaster ( not sure what they are called right now) to get the solder off the pins.
It worked!
I was just about to throw out a couple of bad ones I had, after watching your video I ordered some transistors and hope to repair them. Thanks for the in depth and informative videos you produce!
Nice job on the content. You covered everything well. My PMD has just started playing up (replaced 12 years ago, moved to behind the grill), so I'm waiting on a new new one. Once it turns up I'l pull the transistors from the dying one and test 'em to see if they're worth replacing.
Wow, great video.
Great information.I think I'll be able to do this to the one I just replaced.
Thank you!
Amazing video! Finally I can rebuilt my 5 PMD collect on last few years 😂
Way to go, You are an amazing knowledgeable Man.
Thank you for learning and teach us this.
After all these years someone couldn't invent a better solution for this PMD?
You always have interesting videos. Thanks.
Watched this a couple of times, having an intermittent problem with mine (2000 3500hd) and moved the PMD. Might try this to build a back up. Just wondered why you remove the screw head potting? Seems this is unnecessary- just remove potting on the solder contacts, and remove nuts on the transistor side, desolder, and install new. Also, did you test the old transistors? And do these fail intermittently leading up to a permanent failure or is there some other possible failure mode like a cold solder joint or broken circuit board connection?
Part of the failure analysis across some forums in the day was loss of fastener torque, leading to heat sink failure which kills the transistor internally. So you have to do more than what you suggest in order to completely remove both of the devices for replacement (and new heat sink materials, new fastener adhesive).
My experience of checking several of these PMD/FSD is that these power transistors (PT) are probably not the cause of stalling problems. None of the 10 I have checked had faults/breakdowns with these transistors. If you remove more of this epoxy, you will see some SMD on the printed circuit board. I suspect that it is in connection with these SMDs and their thin wiring paths that the errors occur. These components are much more sensitive to the large temperature variations than the powerful and well-built power transistors, which then create most of the heat. I have also tried taking out the 2 (PT) and putting them outside the PMD (on a separate cooling jacket), only connected to the PMD with wires. As a result, so much heat is not transferred. Now this is a rather impractical solution in relation to the installation of PMD, but a possible solution. By the way, I used a Dremel to remove the epoxy (with a light and steady hand) to avoid adding strong heat to the wiring paths and SMDs.
And if you want to protect yourself from gases, you need something better than a good facemask, one with an ex charcoal filter....
I probably wonder if the constructions on these PMD boxes are actually physically well enough designed to also handle the large temperature fluctuations that occur. Their surface area should perhaps be larger as the heat from the 2 PTs will be concentrated and will affect the other and more sensitive components in the box to a large extent. And this epoxy, I think it is a bit unfavorable in terms of not releasing the heat. It will probably be able to protect the components, but in relation to heat, it may not be so favorable.
Good Luck...
That was very informative. I hope you will forgive me for some criticisms. I would expect transistors bought from ebay or amazon to be counterfeit. Mouser and Digikey both have this part, and Digikey will provide free shipping if you prepay by check.
Finned heat sinks are completely dependent on the velocity of air flow thru the fins. If the PMD were to be mounted so that the length of the heat sinks fins are parallel to the oncoming airflow when the truck is moving it would be more effective. So instead of being mounted behind the solid part of the bumper, mount it in the hole you are reaching into so air will flow thru (along the length of) the fins. If you have a thermocouple temperature meter with 2 input channels, you can measure the temp of the heat sink and the temp of the PMD case. This will tell you if there is poor thermal conductivity between the heat sink and the PMD case. I like to use the white aavid thermal grease instead of thermal pads.
Thanks. I would characterize heat sink location and airflow concerns as a secondary, or even tertiary factor for this particular PMD design. Airflow alone is not what primarily affects the transistors lifespan (MTBF) here. They actually fall victim to two other factors regarding heat dissipation, at least with the grey Stanadyne model. It's not lack of _cooling in motion_ alone that will result in premature failure. It's when the vehicle is at rest that we have to look primarily, at how to dissipate the heat.
The _post drive_ heat soak, after the engine is off and the vehicle is no longer in motion, and then the repeated heat-cold cycling that follows normal vehicle usage is where the root causes appear.
After over a decade of analyzing failures, including purchasing DOA PMDs off of eBay, I have a good idea of the ways they fail that can be repaired (some small % also fail in non-repairable ways too).
Various combinations of lead-free solder joint fracturing at the TO-3 transistor lead(s), loss of torque at the TO-3 fastener(s), and failure of the transistor die itself internally due to repeated excessive thermal exposures all cut the lifespan down considerably. Typically component failure of the PMD sets in around the third year or so of use.
Once the vehicle is in the garage after a drive, no PMD mounting location will save the PMD as there is no airflow involved. Nor any fuel flow for the factory IP mount point. The PMD will fail over a 3 - 4 year period of time due to the _vehicle at rest_ factors I mention without improving the heat dissipation at rest.
Over the next few years I hope to find that the use of _leaded_ solder, superior thermal pads at both the transistor (as shown in the video), and at the heat sink mounting (not shown), as well as superior fastener torque protection can extend the designs lifespan (MTBF). Improving the thermal dissipation is the only lever we can tweak when the vehicle is at rest. So the better pads and better contact with the steel of the bumper, or frame is the focus. Only time will tell over the next few years whether these matter or if the fundamental flaws are simply too great to overcome to get 10 years of life out of one.
With Stanadyne production having ceased in 2015, the day is fast approaching where rebuilding a grey PMD is going to be the only source for a working one. Hence this project that I took he time to share with you all.
@@DrShock Thank you for your detailed reply. My background is cooling of electronic equipment. Bellville washers under the nuts holding down the transistors will maintain clamping force. A fan directly on the heatsink as shown in a product on Amazon called " AAVID CPU Cooling Fan & HEATSINK, 50mm Sq. Fan, 50mm X 52mm X 43mm Silver Heatsink & Fan, TX3 3PIN Connector " will provide cooling when the truck is not moving but it and the pmd would have to be mounted inside the cab for weather protection. The power to the fan would be controlled by measuring pmd temp.. Finally, the transistor leads could be extended out of the back of the pmd case and mounted on two new heatsinks. These heatsinks would be electrically insulated from each other and the cab of the truck. In this way thermal grease can be used instead of an electrically insulating thermal pad.
Great Video, what if one wanted to build their own extension harness .. do you know the type of connectors needed to buy from an outfit like Digikey same with the heat sink. Thanks!
I don't know the Aptiv part numbers for those, but would have to presume they are available as the extension harnesses do exist. I would expect though that the cost of purchasing only two connectors, plus the pins and crimping tool, plus shipping, would mean it's cheaper to purchase an aftermarket extension harness already assembled and tested. The heat sink though could likely be sourced cheaper than pre-made, though you would have to drill and tap the aluminum for the PMD special torx screws and sand smooth for the thermal mounting pad.
@@DrShock OK ... didn't know if you knew the connector number or style .. I'll do a little more research on it and may post a video on making one up. As far as having the the tools I do have those already and the auto wire. From what I have priced for the kit .. heat sink and harness I thought was a bit pricey compared to doing your make up. Thanks again on a great video ...
Can you diode test the old ones before they're removed to see if they're bad?
You would have to desolder the leads to perform the diode test, due to the nature of the surrounding circuitry you would otherwise get false results.
Thank you very much for your videos I just happen to have four dead PMD’s sitting right here that I’ve already started to take apart after watching your videos I have started working on dismantling number three, and I’m gonna be ordering transistors for them..The transistors are definitely testing bad The one PMD was interesting, with sort of a translucent jelly type stuff, some kind of heatsink I’m sure? . Do you have a test procedure for the PMD it’s self while it’s still on the vehicle? other than just knowing that it’s not starting? Maybe a way to connect to it while it’s plugged in?
I am liking all of your 6.5 videos👍🏻 Cheers
Thanks. The potting material you describe, the _translucent jelly_ type sounds like an aftermarket PMD like FlightSystems or D-Tech. Stanadyne never used that type of potting material. Neither GM nor Stanadyne published any sort of complete PMD test. There are some basic voltage checks in a Stanadyne service bulletin, but those are just completely dead or not type tests. They don't find anything in between.
@@DrShock Thank you, for getting back to me so fast, I just search the model number on it and it came back as “Dorman” it is a-904-104 I remember it not lasting very long and it looks a little bit different on the inside, that jelly stuff looks very toxic handling it very carefully! The stock potting material is a lot different, The first one was a good learning experience to whittle a way at even though I think I did OK, but now I know where things are located can avoid some areas, I was hoping that this day would come along where I can repair these seemed like such a waste just to chuck them, I guess that’s why I held onto them for so long. The same with the truck 2000 gmc 3500 4x4 dually, it’s been my daily work truck for the last 12 years been very reliable hauling huge loads all the time! I also have a 1993 1 ton 4x4 same truck still going strong right now it’s the running one, lol anyhow cheers and thanks again for keeping these motors alive! 👍🏻
Thanks. That Dorman would be just throwing their name on a D-Tech type (has a notch in the top black plastic casting). Yeah that material is definitely not anything you want to breath in fumes of if heating, definitely a chemical mask op.
Great tutorial. If you own a scope, could you take a few traces for me? Looking to buy a 98 k3500 that stalled and won't restart, 700 miles from me.
That's pretty bad that you are going through PMDs every 3 years. My dad's truck (95) had the pump replaced once under warranty when the PMD failed, and then it failed again one time after the warranty was up, and he moved it to the front behind the license plate then and it's never been an issue since. Seeing as these are just regular BJT transistors and nothing really special, I'd look for some different parts that have higher current ratings (I'm not showing much doing a quick search at mouser though). Do you put thermal paste between the back of the PMD and the heatsink when you mount it? I'd at least use some regular white thermal paste (or you can also use penaten diaper cream - it's basically the same stuff - zinc oxide paste - I've actually used penaten diaper cream on a CPU heatsink when I couldn't find my jar of thermal paste - never had an issue and that machine was my main computer for many years after)
also it would be interesting to completely de-pot one and look at the circuit design to see if there's something that could be improved to prevent the transistor failure - it's like there's some flyback diodes or something missing from the design...
The ones failing here are not my rebuilt ones, but rather the factory original Stanadyne ones. They are discontinued now, so Stanadyne no longer cares to figure out the remaining design issues. I do not expect my rebuilt ones to fail like the originals. ;-)
I also am running a Standadyne injection pump replaced once under GM extended warranty back in 2004 or so. It's only the PMDs that have a very abbreviated lifespan.
@@DrShock I wonder if the difference is the "gray" PMD vs the "black" PMD - mine is definitely the black one.
I used the black one until the gray came out. Same frequency of failure replacement. Part of this is likely the application, the diesel 4WD Suburban is the most cramped/high heat layout of all the C/K trucks imo, even behind the bumper.
The other caveats are that this has been a south Florida truck, daily highway driver baking in the scorching sun never to experience a garage or winter until recently. It's
climate where that bumper easily gets hot enough to fry an egg every day. It's still on the road 7 days a week, for multiple hours each day where it leads a life more like a delivery van than a typical 6.5L truck.
I'm sure some or all of these factors contribute to aggravating the PMD design flaws more so than the average truck. Hence my side project to rebuild these and improve the parts and materials used in the design to see if I can get a longer PMD lifespan with this same vehicle.
What is the name of that piece?
You'll have to be more specific to answer.
Fantastic detail. It's called flux paste.
I where do the wires? From that harness go to on the motor I had a shop saying, they repaired mine. Put a new one in it and somehow, the truck was running when I got home it wouldn't start again when I pulled the harness. It was cut completely all the wires and 2 places and they take it back in there. So I couldn't see it underneath the battery. So now I gotta run a new harness. And I'm not sure where it goes. From the p m d box to the motor
99 S10 wiring diagram to the models and carberater
Fuel at filter no fuel at injector. No start