Announcement

Collapse
No announcement yet.

Output Transistors Hot

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

  • Output Transistors Hot

    I replaced the transistors in the power amps in a Fender M-80 Chorus, along with several diodes and a few resistors.(a couple of you fine folks helped me with the schematic I needed) After I got it together 1 output transistor shorted and let some of the smoke out of the speaker. I replaced that transistor and after running the amp for about 20 seconds one blew in the other channel, both on the positive side. I thought the power supply could be the cause since they were both on the positive side. So I replaced the rectifier diodes and input filters. now one channel seems ok but in the other the emiter current starts low but continues to increase until the transistors get so hot the amp shuts down. It appears to be bad transistors but it has happened with different sets of transistors. The voltage at the base stays the same, the voltage at the speaker output is zero. but the current is out of control. I can't afford to pull any more hair out at my age. any ideas?

  • #2
    The bias circuit is damaged, I'd bet.

    You might scope the output to see if it turned into an RF generator.

    the two amps are the same but for the part numbers, so looking at the top amp:
    The four resistors R127, 128, 129, 130 and the four diodes CR20,21,22,23 form the bias string. Check each diode for drop with a meter on diode test and power off. And check the actual drop with power on. Verify the value of the resistors.


    The power supply just sits there and supplies power. it is not likely the CAUSE of the problems, even though both failures happen to be on the positive side.

    Exactly WHAT power transistors did you install? Actual TIP142 and TIP 147 or something else like NTE parts?
    Education is what you're left with after you have forgotten what you have learned.

    Comment


    • #3
      Thanks for the reply Enzo, I put the scope on the output and found 160mv of 60HZ hum but no sign of RF. Measuring the resistors in circuit I got 2K on R154 & 156 and 4K on 155 & 157. If I remember right the hum popped up about 10-15 seconds after I turned it on after putting the last set of transistors in. I removed R155 & 157 and temporarily replaced them with 5.1K resistors to see if it would make a difference and it didn't. They measured 4.69K out of circuit. R154 & 156 I had replaced with 2.2K and I checked them before installing. I replaced CR36-39 thinking they could be the problem and they each measure about .61V. I did all this back when I was determined to figure this out my self. It seems to me to be a transistor problem but it has happened with different sets of transistors. I'm using TIP142 &147 made by ST Microelectonics, from Mouser. I think the originals were Motorola. Should I get a better quality transistor?

      Comment


      • #4
        The STs should be fine, I was concerned you had made a sub that was inappropriate. I'll think about it at dinner and come back...
        Education is what you're left with after you have forgotten what you have learned.

        Comment


        • #5
          hey,
          I believe Fender has a bulletin out on all the Chorus type bias ckts. THEy basically increase the value of the resistors that feed the + $ - voltage to the bias network...unfortunately, I'm not where I can readily get that info....the fix works like a charm & the tip's cool right down.
          I'll try to get that on in the next couple days...glen

          Comment


          • #6
            The bulletin tells you to change four resistors.

            At the top and bottom of the bias strings are four resistors - R127, 129, 154, 156 - that are currently 1k 2W parts. Change them to 2.2k 2W parts.
            Education is what you're left with after you have forgotten what you have learned.

            Comment


            • #7
              Originally posted by Enzo View Post
              The bulletin tells you to change four resistors.

              At the top and bottom of the bias strings are four resistors - R127, 129, 154, 156 - that are currently 1k 2W parts. Change them to 2.2k 2W parts.
              Holy Moley! They use 2W resistors in a bias string??????? For pity's sake why?

              And why hasn't the guy who came up with that been fired?
              Amazing!! Who would ever have guessed that someone who villified the evil rich people would begin happily accepting their millions in speaking fees!

              Oh, wait! That sounds familiar, somehow.

              Comment


              • #8
                Who knows? They drop about 6-7 volts. Maybe they wanted strong flame proofs for when the inevitable output xstr destruction occurs.

                It was drawn up in 1990, so maybe the guy HAS been fired by now...
                Education is what you're left with after you have forgotten what you have learned.

                Comment


                • #9
                  Does someone have a copy of this schematic they could share with me? I don't have the M-80 series in my files. Thanks, RE

                  Comment


                  • #10
                    Send me an email with Fender M80 CHorus SChematic as the subject line.

                    Address is: tmenzo at msn dot com.
                    Education is what you're left with after you have forgotten what you have learned.

                    Comment


                    • #11
                      What Next?

                      Thanks for your replies, I changed those 4 resistors already. I just don't know what to check next. It seems to me if it were a bias problem the base voltage would be increasing at the same rate as the emitter voltage when the current increased. Should I unsolder the base of one of the transistors and insert a mA meter to see if the base current is rising? Also the hum I measured was with a 10X probe so that is 1.6v and not 160mv.

                      Comment


                      • #12
                        Yikes,
                        I wouldn't insert an ammeter into any part of a transistor power stage. Everything is direct coupled to keep the currents in check....also closed loop for same reason. One slip of the probe or lack of connection could result in a total melt down.
                        There's really no need for an ammeter in this kind of ckt. You really just need to measure the VOLTAGE across the emitter resistors & then extrapolate the current via ohms law.
                        Typically the emitter resistors are of such low resistance that you can consider your voltage reading (ususally in the 5-20mvolt region) to be a direct reading of current...ie; if your reading is 20miliVolts...then the current through the 1 ohm or less emitter resistor is very close to 20miliamps.

                        I'm still thinking as this is such a common issue with out transistors getting hot, that you're still dealing with the standard bias supply resistor issue.
                        You said you already changed those bias supply resistors..what did you change them too?
                        You could probably safely experiment with those bias supply resistors (perhaps + 10-20%) in raising them until the bias current is in a safe range.

                        Also I know that you have to remove the output transistors from the heatsink in order to work on the clad side of the board...when you test, are you completely reassembling the heatsink block to the transistors & the chassis? Are you using new silicone heat sink compound?
                        You probably know that you cannot run output transistors without the heatsink, even if you're just idling the amp...they will get superhot & eventually fry.
                        What is the idle current across the emitter resistors?
                        Is anyone familiar with what the idle current should be on this amp? I don't recall right off.
                        Also, is the sound out clear & not distorted?

                        glen

                        Comment


                        • #13
                          Thanks for the info Glen, I changed R155 & 157 from 4.7k to 5.1k and still have the problem. the voltage across the emitter resistors (.47ohm) starts at about 35mv, after the amp runs for 30 seconds or so it increases to 200mv. At that point the transistors are getting hot and I turn it off. I always bolt it back in the chassis before turning it on but I have not used heat sink compound. The sound was clear when I tried it, lately I've been running it with 20w resistors and no signal until I get the bias problem solved. I got tired of trying to balance it in the cabinet to connect the speakers. I have some 5.6k resistors I'll try them.

                          Comment


                          • #14
                            K Dave,
                            I have to say that I still would suggest that you use silicone heatsink grease even when testing...I know it's gloppy & messy, but is the only way to ensure proper heat transfer (unless of course you are using the rubberized insulators.

                            also wonder if the heat sensing diodes that go in holes in the heatsink are ok. THey have to send back feedback about the heat to the bias ckt.

                            Have you run the amp with no loads & seen if you're getting appreciable dc offset voltage? Perhaps your issues is not so much bias as it is a dc offset.

                            glen

                            Comment


                            • #15
                              My thought also, run it without loads and see what it does.

                              Are both amps running hot similarly?
                              Education is what you're left with after you have forgotten what you have learned.

                              Comment

                              Working...
                              X