Thermal Pads and Bad IPM's: What if it's NOT Heat?

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • Jonny Accelerant
    Banned
    • Aug 2014
    • 205
    • USA

    #1

    Thermal Pads and Bad IPM's: What if it's NOT Heat?

    Working on a Maxent MX-42HPM20 and had a thought that I think merits a separate thread.

    My thought process is as follows:

    My Plasma TV seems to have a bad ZSUS IPM (unconfirmed) and maybe a bad YSUS also (no reason to think this yet), and while doing research on these IPM's I get the impression that it's a common failure, and while there isn't much consensus on what causes these expensive IPM's to go bad, most people assume that it's heat, as they run hot, and there's this (11mm thick) thermal pad behind them. You'd think it was there for a reason. They get hot.

    When I pulled my ZSUS board off, there was an obvious trace of liquid emanating from the pad, and some of the material from the pad stuck to the IPM, and it was almost "glued" to the IPM. I interpret all of this as a breakdown of the thermal pad material. So, while doing research on replacing this pad, maybe with a larger and more effective pad (it's too small, IMO, 19mm x 41mm x 11mm (thick), and it could easily be twice that large, or even larger), I note that a lot of the thermal pads listed by Digi-Key contain the words "Aluminum Oxide" in the description, which seems appropriate because my bad is unusually heavy, dark grey and it might have some aluminum oxide in it, and then I had the idea.

    The idea is that, what if these pads work okay when they are new, and are not conductive, or capacitive, or whatever, but after (in my case) 7+ years the material breaks down (seeping liquid, glues itself to the pc board and material rips away when pad is removed), the electrical characteristics of these pads change and THAT is what is killing-off these IPM's. I've read at least one anecdote that the ZSUS IPM fails after replacement, and that would make sense if it is the pad that is killing them off.

    The idea of forcing the pins of a (small) mass of electrical components into a solid material made out of aluminum oxide seems highly counter-intuitive to me. You wouldn't mash a flat piece of metal into the back-side of a pc board; so why would you assume a spongy material filled with aluminum oxide is not going to short, or allow some conduction, or act as a capacitor, or whatever. Cause problems, is what I mean here.

    Those are my thoughts. I'd be interested in whether or not I might be on to something, or (due to lack of knowledge) I'm missing something.
    Last edited by Jonny Accelerant; 09-24-2014, 10:25 AM.
  • Jonny Accelerant
    Banned
    • Aug 2014
    • 205
    • USA

    #2
    Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

    Also (I'm giving this a separate post because it's crazy), I noticed that the back of this ZSUS pc board had a film of hard, crusty dried liquid on it. I assumed it's some kind of flux material, or maybe some kind of corrosion-inhibitor.

    So, WHAT IF the pad is fine, and it's not degrading due to heat, but instead it's degrading and chemically reacting with this brown stuff? At the factory. So in some lab somewhere, the brown stuff doesn't harm anything, the pad doesn't harm anything, but after they are assembled maybe the two do some kind of chemical reaction that turns a non-conductive pad into something that 1) bonds to the pc board 2) creates a liquid that seeps out 3) changes the electrical characteristics of the thermal pad that 4) kills-off ZSUS IPM's?

    Whazzup? Huh? Do I get a parade? Or a Federal Investigation?
    (cause this is my grand conspiracy theory, & I'm either a hero or a nutjob)

    Comment

    • SteveNielsen
      Retired Tech
      • Jun 2012
      • 2327
      • USA

      #3
      Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

      How about some pictures?

      Some heros are nutjobs, I don't think they are mutually exclusive terms.

      Comment

      • budm
        Badcaps Legend
        • Feb 2010
        • 40746
        • USA

        #4
        Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

        Aluminum oxide is insulator:
        http://accuratus.com/alumox.html
        http://www-ferp.ucsd.edu/LIB/PROPS/PANOS/al2o3.html

        Thermal compound (Thermal paste for heatsink) also contain Aluminum oxide, or Zinc oxide (white stuff).

        "act as a capacitor" When you have two conductors separated by dielectric (I.E. insulator), you form capacitor.
        Last edited by budm; 09-24-2014, 11:41 AM.
        Never stop learning
        Basic LCD TV and Monitor troubleshooting guides.
        http://www.badcaps.net/forum/showthr...956#post305956

        Voltage Regulator (LDO) testing:
        http://www.badcaps.net/forum/showthr...999#post300999

        Inverter testing using old CFL:
        http://www.badcaps.net/forum/showthr...er+testing+cfl

        Tear down pictures : Hit the ">" Show Albums and stories" on the left side
        http://s807.photobucket.com/user/budm/library/

        TV Factory reset codes listing:
        http://www.badcaps.net/forum/showthread.php?t=24809

        Comment

        • Alastair E
          Chief Womble
          • Mar 2013
          • 1963
          • U.K.

          #5
          Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

          Its thermal cycling, the RoHS Directive (removal of lead) that causes them to fail. They just Wear out.
          Heating/cooling of dissimilar expansion materials causes mechanical failure of the bonding between the semiconductor and the substrate/base material.

          --NASA and various other bodies have done much research on the subject of 'Lead Free' electronics--Google it, there's vast reams of info--Good reading for a boring Winter's night!

          IPM's seem to last around 15 to 20 Thousand Hours--Not bad for something in the Consumer Electronics arena.....

          Worth also googling-- Planned Obsolecence......
          TELEFIX

          How PLASMA SCREENS WORK, X-SUS and Y-SUS what they do--
          http://www.irf.com/technical-info/appnotes/an-1088.pdf
          PLEASE DO NOT EMAIL ME PRIVATELY FOR REPAIR ADVICE. QUESTIONS BELONG ON THE FORUM!

          Comment

          • Jonny Accelerant
            Banned
            • Aug 2014
            • 205
            • USA

            #6
            Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

            Originally posted by SteveNielsen
            How about some pictures?
            Here is the pad (front and back). I have to figure out how to get attachments in one post so show up in another.

            Here's the thread/post that shows the ZSUS board with the leaking liquid and the scrap of remaining material stuck to the back of the pc board.


            https://www.badcaps.net/forum/showpo...50&postcount=6
            Attached Files

            Comment

            • Alastair E
              Chief Womble
              • Mar 2013
              • 1963
              • U.K.

              #7
              Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

              The stuff that comes off the dead-toad pad is only silicone oil, so nothing to make issue over.
              The flux coating left on the board doesn't cause the trouble either.....

              Blame the Greenies and their Damnable RoHS Directive!
              TELEFIX

              How PLASMA SCREENS WORK, X-SUS and Y-SUS what they do--
              http://www.irf.com/technical-info/appnotes/an-1088.pdf
              PLEASE DO NOT EMAIL ME PRIVATELY FOR REPAIR ADVICE. QUESTIONS BELONG ON THE FORUM!

              Comment

              • Jonny Accelerant
                Banned
                • Aug 2014
                • 205
                • USA

                #8
                Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

                Originally posted by Alastair E
                The stuff that comes off the dead-toad pad is only silicone oil, so nothing to make issue over.
                The flux coating left on the board doesn't cause the trouble either.
                Why do you call it a "dead toad pad"?

                I did a Google search and you are the only person on the entire planet that puts those 3 words together in that order. The only other hit besides this thread, is a post you did here:

                I think this is what causes the odd failure--not of the actual chip, but of that rotten lead-free solder holding it on.
                They basically suffer metal-fatigue on the solder-balls and connections break between the chip and the board.
                --Some makers make the situation worse--by trying to cool the chip from the other side of the board, using a dead-toad pad to the chassis.
                Which is exactly what's happening here, and you say the pad makes it "worse", where in my original thread, someone else says the IPM would burn up immediately without that pad. So which is it? Worse, or necessary?

                Since you know so much about pads, etc.., do you know a source for larger pads? Isn't it safe to assume that bigger is better in this application?
                Last edited by Jonny Accelerant; 09-24-2014, 07:03 PM.

                Comment

                • mmartell
                  Badcaps Legend
                  • Oct 2013
                  • 3189
                  • Canada

                  #9
                  Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

                  I have seen these pads on ysus boards and buffer boards and never thought they were for heat transfer. Their locations didn't seem to facilitate that. In any event of the three dead zsus boards in my possesion from a similar lg set not one has this pad. The ipm I have lifted off my board is not sealed on the bottom and the components on it are fairly simple - the heat generated comes from the other side where the igbt's are packed and the large heatsink mates to.

                  @Alistair, you have a thread here from around the beginning of the year about repairing one of these models. Do you remember which module it was or from which set it came ?

                  Comment

                  • Jonny Accelerant
                    Banned
                    • Aug 2014
                    • 205
                    • USA

                    #10
                    Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

                    Originally posted by mmartell
                    In any event of the three dead zsus boards in my possesion from a similar lg set not one has this pad.
                    Maybe they died because they didn't have these pads?

                    Comment

                    • mmartell
                      Badcaps Legend
                      • Oct 2013
                      • 3189
                      • Canada

                      #11
                      Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

                      How did I know you would say that ? They may have lasted longer I have no way of knowing...

                      And there is no evidence they were ever there and then removed so if that is indeed their function I'm not sure why I'm missing them.

                      Comment

                      • Th3_uN1Qu3
                        Believe in
                        • Jul 2010
                        • 6031
                        • Romania

                        #12
                        Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

                        Those pads can only transfer a minute amount of heat. The bulk of heat transfer is done through the soldered connection of the IPM to the board. That's why they're such a bitch to remove.
                        Originally posted by PeteS in CA
                        Remember that by the time consequences of a short-sighted decision are experienced, the idiot who made the bad decision may have already been promoted or moved on to a better job at another company.
                        A working TV? How boring!

                        Comment

                        • tom66
                          EVs Rule
                          • Apr 2011
                          • 32560
                          • UK

                          #13
                          Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

                          The pads function is as an insulator, I think. They're usually mounted on the bottom of certain transistors and on parts of the board that may be more flexible, possibly leading to failure of the board during operation if the pad was not present?
                          Please do not PM me with questions! Questions via PM will not be answered. Post on the forums instead!
                          For service manual, schematic, boardview (board view), datasheet, cad - use our search.

                          Comment

                          • Alastair E
                            Chief Womble
                            • Mar 2013
                            • 1963
                            • U.K.

                            #14
                            Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

                            Dead-Toad....

                            Cold, Clammy--Just like those silicone pads......Summit Ive always called 'em.

                            I said-- IN SOME CASES--Where a chip is mounted On Top Of The board--like a large processor, and the toad is Under the board between the chassis metal and the board--The Unequal Expansion is greater due to the toad removing heat from underneath the board, the chip itself still getting quite hot, Ive measured 55 degrees on the top of a processor chip before now, and the chassis still cool at 18 degrees.

                            IF the chip is cooled by adding a heatsink to its Upper face, the problem is reduced--Note, Reduced, not cured, as you're taking the heat from the chip--not the board via the soldered pads etc. Failure of the through-hole soldering --like IPM--isnt such a problem as it is for the soldering of those processor-chips using solder-ball technology
                            For IPM its the Die Chips soldering to the substrate that fails, causes local overheating--as the thermal bond is less-- of the IGBT that eventually dies or the electrical connection between the chip die and the substrate fails. Result is the same--IPM goes tits up.
                            Must be remembered--This Usually happens at the 15000 to 20000 hours point, as thats when they seem to last to. Thats a Huge amount of time--For a Consumer Electronics item.

                            Makers know that Toad is cheap, heatsinks are expensive--relatively. They don't care if the set fails 2 years or more down the road, and tell you to go buy a new one, hence the poor service-info availability for many parts of a Plasma chassis.

                            I disagree with the above statement--'Those pads can only transfer a minute amount of heat. The bulk of heat transfer is done through the soldered connection of the IPM to the board. That's why they're such a bitch to remove.'

                            Yes--IPM sinks to the board, the toad then heatsinks from board to the chassis....
                            Ask yourself why the toad is Always stone cold to the touch. Its conducting the heat outta your finger!

                            --Try an experiment. Put some toad onto ice-cube, touch it and see for yourself!

                            Its not only the semis that need cooling in some sets--The ER coils/caps often have toad located at their connections on the board to sink to chassis too, as well as the row of small ferrite-beads carrying the SUS currents....

                            The IPM I repaired was in a LG 42PC1DA set. It worked perfectly on my own wall for over 6 months, then I sold it when a Panasonic 42PZ70 came to me.
                            --A week later, it came back with the Y-SUS IPM failed, (I had repaired the Z-SUS IPM) so I changed both SUS boards for a good pair and haven't seen it since.....

                            Ive not got back to the boards, not needed to....
                            TELEFIX

                            How PLASMA SCREENS WORK, X-SUS and Y-SUS what they do--
                            http://www.irf.com/technical-info/appnotes/an-1088.pdf
                            PLEASE DO NOT EMAIL ME PRIVATELY FOR REPAIR ADVICE. QUESTIONS BELONG ON THE FORUM!

                            Comment

                            • Jonny Accelerant
                              Banned
                              • Aug 2014
                              • 205
                              • USA

                              #15
                              Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

                              Okay so then if toads are good, then bigger is better.
                              Right?
                              Does anyone know of a source for 11mm thick, thermal conducting silicon pads, that may or may not have aluminum oxide in them?

                              Comment

                              • Jonny Accelerant
                                Banned
                                • Aug 2014
                                • 205
                                • USA

                                #16
                                Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

                                Okay so then if toads are good, then bigger is better.
                                Right?
                                Does anyone know of a source for 11mm thick, thermal conducting silicon pads, that may or may not have aluminum oxide in them?

                                Comment

                                • budm
                                  Badcaps Legend
                                  • Feb 2010
                                  • 40746
                                  • USA

                                  #17
                                  Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

                                  http://www.aliexpress.com/item/200mm...012140472.html
                                  Never stop learning
                                  Basic LCD TV and Monitor troubleshooting guides.
                                  http://www.badcaps.net/forum/showthr...956#post305956

                                  Voltage Regulator (LDO) testing:
                                  http://www.badcaps.net/forum/showthr...999#post300999

                                  Inverter testing using old CFL:
                                  http://www.badcaps.net/forum/showthr...er+testing+cfl

                                  Tear down pictures : Hit the ">" Show Albums and stories" on the left side
                                  http://s807.photobucket.com/user/budm/library/

                                  TV Factory reset codes listing:
                                  http://www.badcaps.net/forum/showthread.php?t=24809

                                  Comment

                                  • Jonny Accelerant
                                    Banned
                                    • Aug 2014
                                    • 205
                                    • USA

                                    #18
                                    Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

                                    YIKES! $78 for a 200mm x 400mm pad.

                                    Comment

                                    • Alastair E
                                      Chief Womble
                                      • Mar 2013
                                      • 1963
                                      • U.K.

                                      #19
                                      Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

                                      What Ive done--Is fit a 12V Fan --old PC case-fan type or processor-fans, to the heatsinks of IPM's, but run 'em off the 5V rail.--Not all PC fans will start and run happily at only 5V, its the luck of the draw, but Ive found that most do.

                                      On the old LG PC1da type sets, the 5V rail is easy to find--Its on the small ceramic gold-plated fuse by the side of the main Vs multi-way plug.

                                      They run very quietly, and make a huge difference to the Heatsink and therefore, The IPM temperature.
                                      Alternative to sinking via the IPM pins--to a plasma-panel ally frame that itself gets warm.

                                      In that 42PC1DA, I noticed when I did this, the whole set ran a lot cooler, apart from the area above the PSU which was still fairly warm.

                                      Heat kills electronics, whether drying out caps or stressing semiconductors, so its gotta help.
                                      TELEFIX

                                      How PLASMA SCREENS WORK, X-SUS and Y-SUS what they do--
                                      http://www.irf.com/technical-info/appnotes/an-1088.pdf
                                      PLEASE DO NOT EMAIL ME PRIVATELY FOR REPAIR ADVICE. QUESTIONS BELONG ON THE FORUM!

                                      Comment

                                      • mmartell
                                        Badcaps Legend
                                        • Oct 2013
                                        • 3189
                                        • Canada

                                        #20
                                        Re: Thermal Pads and Bad IPM's: What if it's NOT Heat?

                                        That is exactly what I intend to do in a 42pc3dv soon as my module arrives and the board is fitted.

                                        Comment

                                        Related Topics

                                        Collapse

                                        • Document Archive
                                          Mounting and Handling Guidelines for TO220 TO220F TO247 Cases / Packages
                                          by Document Archive
                                          Mounting and Handling Guidelines for TO220 TO220F TO247 Cases / Packages

                                          The TO220, TO220F and TO247 are the popular packages for power devices because of their versatility and ability to dissipate moderate amounts of heat. This application note describes the basic guidelines for handling power MOSFETs in TO220, TO220 and TO247 packages shown in Figure 1. Please note that only mechanical and soldering guidelines are covered here. Additional precautions are required for isolating high voltage rated devices to meet safety regulations.

                                          Securing a good thermal interface between...
                                          10-07-2024, 04:35 AM
                                        • sancho76
                                          thickness of thermal pads for Alienware M15 R1
                                          by sancho76
                                          Hi, somebody knows the thickness of thermal pads for M15 R1? Or if someone have this Alienware model, can take measure for original thermal pads.
                                          02-01-2025, 03:19 AM
                                        • harp
                                          detecting heat with IR LED receiver
                                          by harp
                                          detecting heat with IR LED receiver

                                          I have some dark IR led, and want to know if it is be situable to detect pure heat (not visable glow ir radiation) in well lighted desk?

                                          So, if I put hand over it in near distance like few centimeters, that detect rising heat, not covering it from light.
                                          Something like termal camera, but in one spot, any simple circuit?...
                                          09-06-2023, 06:03 PM
                                        • dbuergi
                                          Restore solder pads for SMC on MacBook A1466
                                          by dbuergi
                                          Hi all,

                                          i got a A1466 logic board for spare parts but it looks like the SMC ist bad. I got several spares but somehow i was too fast while removing it and ripped some pads off. You can see the damage attached.
                                          Just for experience i would like to restore the lost pads but before i start (haven't done this before) do you have some hints or tutorials how to do it?
                                          As written i want to improve skills so it doesnt matter if it takes longer.
                                          12-29-2024, 10:24 AM
                                        • ludovik
                                          Moto G84 short after heat
                                          by ludovik
                                          Hello everyone,
                                          i was reballing the power manger of a moto g84, after applying heat i tried to power on and it wouldn’t.
                                          i injected 1.5v to some capacitor that seem in short because both side beep if tested.
                                          when i inject i feel the cpu burning;
                                          since the cpu is exactly on the other side of the board, is that possible that the heat from the reballing damaged the cpu? do you know where i can find schematics or what can i do?
                                          i know this is not enough to explain the situation so if you need any other info ask me please!
                                          Thank you very much in advance
                                          11-08-2024, 02:13 PM
                                        • Loading...
                                        • No more items.
                                        Working...