Announcement

Collapse
No announcement yet.

Peavey CS800S "Tempsistor" testing?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    What you could do is lift up the output xstrs' emitter resistors R187, 188, 195 thru 201, as well as R185, and the sense resistors R189 thru 194) at the emitter connections. That then leaves the base-emitter junctions open per xstr. The base-collector junctions are still all in parallel. So, if there is even one transistor shorted on the pos or neg side, all will read the same. Usually, when an output xstr fails, it fails with a C-E short as well as B-E and B-C short. And, unless the emitter resistor for that part failed open, it would pull the power supply down, and blow the mains fuse, unless there were power supply fuses per amp channel.

    You need to check the emitter resistors in the output stage first, as if you find any that are open, it's associated power xstr is likely bad.

    In how to do that (removing power xstrs), I had moved on from Soldapulit style solder extractors to Pace Desoldering irons with a full set of tips. Also a Weller Desoldering iron for their WTCP line. And, Solder Wick.

    Now, when CR140/CR141 were in circuit, and they tested short, but upon lifting, measured ok, that tells me you have Collector/Emitter shorts on both NPN and PNP sides of the output stage. That also suggests there are power supply fuses on the amp boards.

    In looking at the parts placement guide, super-imposed over the back foil pattern, I don't even see foil for the power supply rails bussing the collectors of all the power xstrs together. Nor have I found where CR140/CR141 are. I didn't see any fuses, but, I'm also not looking at the actual boards as you have.

    So, yes, you do need to find what has failed in the output stage, to yield the shorted readings you got on CR140/CR141.
    Logic is an organized way of going wrong with confidence

    Comment


    • #17
      Thanks again nevetlab,

      I will lift the resistors you suggest and check. I'm pretty convinced all of the power xstrs are cooked, but maybe I'll get lucky. I'm thinking that at some point, either the other person that was in it, or me, hooked up the PS wires wrong or something. Maybe that's why he gave up and got rid of it. But of course I'd rather not put $50 worth of xstrs in if I don't have to.

      I've attached a copy of the PCB image with CR140 & CR141 circled.

      Click image for larger version

Name:	PCB with CR140 & CR141 circled.jpg
Views:	1
Size:	5.33 MB
ID:	851909,

      Comment


      • #18
        Ah.....I must have had blinders on this morning. Also, from the 'blinders' effect, I see the power supply 'rails'. Each aluminum plate being used as the heat sink/heat spreader IS electrically connected to the TO-3 cases. No Insulator between the TO-3's and the heat sink. My bad.

        You might just have one pair of xstrs that failed. Without pulling up the emitter resistors, all of the base-emitter junctions will measure low, though not dead short, unless they all are. The base-collectors....I've already explained that, so you WILL have to remove the xstrs to find the culprits. I haven't looked up to see what Peavey installed. MJ15022's and MJ15023's are what most folks are replacing them with these days, using them in a complete set. They are a bit pricey....$5.56 ea on the MJ15022's is what I paid last month for a pair.
        Logic is an organized way of going wrong with confidence

        Comment


        • #19
          I finally had a chance to pull the power xstrs yesterday evening and measure them. It was easier than expected. I just flipped the board over, soldapullt the leads and went back and forth trying to keep the solder molten and since the heat sink held them above the workbench, they just fell out. :

          Q128 (70473180)
          .488, .455, OL

          Q129 (70473180)
          000, 000, 000

          Q130 (70473180)
          .488, .443, OL

          Q131 (70473180)
          .488, .444, OL

          Q132 (70483180)
          .490, .461, OL

          Q133 (70483180)
          .490, .460 , OL

          Q134 (70483180)
          .127, .127, 000

          Q135 (70483180)
          .491, .463, OL

          All measurements shown in B-E, B-C, C-E order
          OL = open

          So it looks like Q129 and Q134 are toast. It seems odd to me that it wasn't a pair, but maybe I'm still thinking tube amp too much. Here are the pairs per the schematic, unless I'm reading something incorrectly (always a possibility). I've underlined the obviously bad ones.

          Q128 - Q132
          Q129 - Q133
          Q130 - Q134
          Q131 - Q135

          Am I safe just replacing those two or should I replace as pairs?

          My plan now is to retest the other xstrs and diode, just to be sure, then check resistors, etc. I'm not sure about how to go about testing the ICs though. After that, ordering all the parts.

          Any other hints or suggestions?

          Thanks!

          Comment


          • #20
            There are no pairs. There is a row on the plus side, and a row on the minus side. The fact they are drawn one above the other, or that they sit in the same positions in the two rows means nothing.

            Yes, those two look bad, so all that means is one of four from each side has shorted.

            It is typical that when one fails on the plus side, one will then fail on the negative side. Only a coincidence if the two seem "paired" up.
            Education is what you're left with after you have forgotten what you have learned.

            Comment


            • #21
              You've made good progress! I just looked up the Peavey P/N's 70473180 & 70483180 to find they're MJ15025G & MJ15024G TO-3 parts, readily available thru Mouser or Digi-Key Electronics (usually). Similar price to the MJ10522's I had mentioned. Also check R197 & R200 0.33 ohm 10W ceramic bathtub resistors. They being 10W parts they may be ok, since you had found CR140/CR141 measuring as a short until you lifted one end each. Now we're just checking to see if they measure the same value as the other's, since a lot of current was flowing thru them.

              Did you ever find a pair of fuses between the +/- High Voltage power supplies and the amp module where that connects? Shorted power xstrs on one of the two channels would normally pull sufficient current to blow the mains fuse, yet the amp works with just the good amp channel. I didn't see any circuit protection per channel to dis-connect the power amp from the amp's power supply. Not shown in the schematics.
              Logic is an organized way of going wrong with confidence

              Comment


              • #22
                I finally had time to go over this one more time. I checked the resistors suggested (as well as all of the flame proof and big ceramic ones, I have plenty of 1/4W resistors so if I discovery any bad ones later that's not a problem) and didn't find any bad. I rechecked all of the xstsrs and diodes again. In all cases, except below, the "good" board and "bad" board read the same. Previously some looked shorted, but I hadn't compared it to the other board and/or I was measuring from the component side previously but switched to the trace side tonight.

                Q121 (I said Q120 in an earlier post, I think I had Q120 and Q121 backward when writing them down previously)
                I wanted to be sure it was not a nearby resistor causing the weird readings so I pulled Q121 and it is showing a short. The Peavey Cross Reference List says to replace it as a pair. I assume they mean along with Q120. As previously mentioned Peavey lists MPS-U10 and U60 as replacements, but are obsolete. I'm not sure what are safe replacements?

                There was another "weird" reading when comparing, Q119. In this case it actually seems like the "good" board might have a bad xstsr:
                "bad" board: .630/.630 reversed: OL/OL (OL across C-E)
                "good" board: .570/.590 reversed: OL/.920 (OL across C-E)
                Does it look like Q119 on the "good" board is on it's way out?

                At this point I think my shopping list is the following:
                qty 1 each of MJ15025G & MJ15024G
                qty 1 each of whatever will replace MPSU10 & MPSU60
                qty 1 47ohm FP 1/4W resistor (the one that burned)

                Should I pick up a replacement for Q119?

                I'm not sure about any capacitors. None are obviously cooked and I'm not sure how to really check them without pulling them.

                Am I missing anything else?

                Thanks as always!

                BTW, I don't see any fuses that would be specific to that channel. The GND, +75 and -75 don't have separate/different feeds to the main boards.
                Last edited by stoneattic; 10-27-2018, 02:11 AM.

                Comment


                • #23
                  Q120 is a SDS7204 transistor.
                  https://www.tubesandmore.com/product...7204-300v-10ma

                  Q121 is a SDS7205.

                  I would give Peavey Parts a call on that last one.

                  Comment


                  • #24
                    Suitable replacement parts for Q120 PNP would be a 2N5415 TO-5 metal can xstr, and for Q121, a 2N3440 NPN, also TO-5 metal can xstr. Same pin-out as the MPSU10 & MPSU60. As only 3 mA is flowing thru this stage, you could probably get away with the MPSW42 and MPSW92, which are 1W tall TO-92 xstrs. Not sure what Peavey has for substitutes. I don't think any semi mfgr is building TO-202 packages anymore. For Q119, a 2N5401 would suffice. The difference in readings on the two Q119's may have been from Q121 being bad, or removed from circuit. Both sets of readings do indicate good junctions.

                    As you don't see any supply fuses, do an ohmmeter check between the two amp modules.....NPN T0-3 case on one to the TO-3 case of the other, as well as the PNP T0-3 cases. If you see near 0 ohm (wire resistance), then there aren't any fuses per module. I'm still puzzled how the supplies stayed up when you had shorted output xstrs. Anyone else curious about that?
                    Logic is an organized way of going wrong with confidence

                    Comment


                    • #25
                      Nitpick: I call a 2N3440 a TO39 can, to me the TO5 is the smaller thing. I know not all agree with me.
                      Education is what you're left with after you have forgotten what you have learned.

                      Comment


                      • #26
                        Originally posted by Enzo View Post
                        Nitpick: I call a 2N3440 a TO39 can, to me the TO5 is the smaller thing. I know not all agree with me.
                        .....removing the egg from my face.....you're right! I had to go back to the data sheet on the 2N5415, and there it was...TO39.
                        Logic is an organized way of going wrong with confidence

                        Comment


                        • #27
                          Thanks again!

                          I added the 2N5415 and 2N3440 to my Mouser order. I'm not sure what else I should check, caps, etc., or am I at the point where I should just place the order and install everything and see where I am?

                          Comment


                          • #28
                            Originally posted by stoneattic View Post
                            Thanks again!

                            I added the 2N5415 and 2N3440 to my Mouser order. I'm not sure what else I should check, caps, etc., or am I at the point where I should just place the order and install everything and see where I am?
                            As long as you have emitter resistors for Q120/Q121, and the 47 ohm series resistors that are used in the power supply lines for the 'front end stages', which do include Q120/Q121. The question about what else to order is always a tough one. Every time you find you need something else, there goes another shipping charge paid out to Mouser. I cringe every time I place an order, knowing the next amps on the bench will tell me I should have waited before ordering.

                            I gather the associated emitter resistors for the two power xstrs WERE intact and not open. I think it's time to restore the circuit and see what else, if anything, is needed.
                            Logic is an organized way of going wrong with confidence

                            Comment


                            • #29
                              I checked all of the resistors that were not 1/4w carbons, as I have plenty of those, and went ahead and placed the order. When I receive the parts I will install and report back.

                              Thanks for all the help. Besides (hopefully) getting the amp back into service, I learned a good bit this time around.

                              Comment


                              • #30
                                It took along time, but I finally had a chance to work on this again. I installed the new MJ15025 & MJ15024 power transistors and 2N5415 & 2N3440 voltage gain stage transistors, as well as reinstalling the other odds and ends I pulled/lifted during the troubleshooting last night and tested. Channel A is back and everything appears to be acting as it should!

                                Thanks again for all the help!

                                Comment

                                Working...
                                X