No announcement yet.

Discrete SS Guitar Amp Troubleshooting

  • Filter
  • Time
  • Show
Clear All
new posts

  • Discrete SS Guitar Amp Troubleshooting

    Hey all! I'm new to this forum but it seems like home to a great many intelligent helpful people. Hopefully I will be able to get some help with my guitar amp. I have had practical experience with electronics for many years, but not so much the theory.

    I have an old Heath TA-16 from probably the mid-70's. It is fully solid-state, completely discrete and a fabulous amp. I was able to get my hands on the schematic here:

    It is connected to two 10-inch speakers in parallel. I couldn't find the impedance labeled on the speakers, but there is a label near the speaker leads that reads: 25W RMS at 4 ohms and 15W RMS at 8 ohms. Judging by the fact the speakers are wired in parallel, I'm assuming they're 8-ohm speakers?


    1) The amp does not seem nearly loud enough for the power rating; also, sometimes when I turn it up to 8-10 it will be loud and other times it is quiet. I have read a lot of reviews on this amp ( and I don't feel like my amp is giving the kind of performance other people are getting.

    2) There is constant distortion, regardless of whether the volume is at 3 or 10, and regardless of which channel I play through. Turning down the volume on my guitar just makes the volume even quieter, but does not fix the distortion.

    3) The tremolo circuit does not seem to work at all; twisting the knobs does not cause any effect. Note: there is a jack to connect a footswitch, but I don't have one here at the moment to try it. I have been assuming the tremolo should work without the footswitch plugged in, but I could be wrong.

    4) The tone knobs do not seem to do a lot either. I dunno if it's just a circuit that does not provide a very drastic change or if something needs to be fixed/replaced/changed.

    With this entire amp, it seems to be original parts, all discrete. The resistors are the old style large brown cylinders, etc. I have gone through and cleaned all the pots with contact cleaner, but that's the extent of what I've done so far.

    If anyone has any suggestions to fix these problems it would be appreciated.

    Thanks in advance for any help you may be able to offer; I look forward to being able to play this amp and have it sound as great as I know it can.

    ~~JoJo Manee

    P.S. This amp used to work better, but has never really worked right in the 7 years or so I've had it. It has been in storage in a garage the last 3-4 years, which seems to have deteriorated it more. It is now safe inside the house again, and I finally have a schematic, so maybe I'll be able to get it fixed.

  • #2
    It sounds like there are a number of problems with your amp.

    I would suggest starting with checking the power supply and then proceed to the rest of the amp. Print out a copy of the schematic and compare the written voltage values with what you find in the amp.

    Also check the entire amp for bad or broken solder connections.

    The distortion that you describe sounds like a bias problem in the power amp. Check to see if either output transistor has been replaced. Compare the actual part numbers with the printed schematic.

    Once the distortion is cleared up then move on to the other problems.


    • #3
      Also test those filter caps C35, C36, C37. Give the amp a smack on top and listen for any changes that could indicate a broken solder point. The circuit board is going to be fragile so be very careful taking things apart. Repeat cleaning of the jacks and pots. I would also test those transistors, preamps right through to the outputs Q9 and Q10. If you have a DMM be sure all your voltages throughout are within spec. The schematic tells you what to look for.


      • #4
        Thanks for the suggestions. Now at least I have a starting place. I'll post soon and let you know what I find. :-)

        ~~JoJo Manee


        • #5
          I would scope the output, my reaction to your description is half the waveform missing - as in either the positive or negative side is not working.
          Education is what you're left with after you have forgotten what you have learned.


          • #6
            Using the directions at:
            and using the ohms setting on my meter:

            [Note: I did these tests with the power off and I did not disconnect the transistors from the circuit. I hope I did them correctly. Also, the red and black in parentheses refers to the red/black positive/negative terminals on my meter. All measurements are approximate - an analog graph doesn't allow for perfect measurements.]

            Q6 - [NPN - 2N3393 Mixer Amplifier]
            Between base (red) and collector (black): 40K ohms
            Between base (red) and emitter (black): 40K ohms
            Between base (black) and collector (red): 3K ohms
            Between base (black) and emitter (red): 3K ohms
            Between collector (black) and emitter (red): 3K ohms
            Between collector (red) and emitter (black): 20K ohms

            Q7 - [NPN - 2N3416 Predriver]
            Between base (red) and collector (black): 30K ohms
            Between base (red) and emitter (black): 30K ohms
            Between base (black) and collector (red): 3K ohms
            Between base (black) and emitter (red): 3K ohms
            Between collector (black) and emitter (red): 5.5 K ohms
            Between collector (red) and emitter (black): 2K ohms

            Q8 - [NPN - 2N3053 Driver]
            Between base (red) and collector (black): infinite
            Between base (red) and emitter (black): 0 ohms
            Between base (black) and collector (red): 2.5K ohms
            Between base (black) and emitter (red): 0 ohms
            Between collector (black) and emitter (red): infinite
            Between collector (red) and emitter (black): 2.5K ohms

            Q9 - [NPN - 2N2148 Output Amplifier]
            Between base (red) and collector (black): 0 ohms
            Between base (red) and emitter (black): 2.5 K
            Between base (black) and collector (red): infinite
            Between base (black) and emitter (red): infinite
            Between collector (black) and emitter (red): 400 ohms
            Between collector (red) and emitter (black): 0 ohms

            Q10 - [NPN - TA 2577 Output Amplifier]
            Between base (red) and collector (black): 500 ohms
            Between base (red) and emitter (black): 500 ohms
            Between base (black) and collector (red): 400 ohms
            Between base (black) and emitter (red): 1K ohms
            Between collector (black) and emitter (red): 1K ohms
            Between collector (red) and emitter (black): 0 ohms

            Q12 - [NPN - 2N3692 Tremolo Oscillator]
            Between base (red) and collector (black): 40K ohms
            Between base (red) and emitter (black): 40K ohms
            Between base (black) and collector (red): 3K ohms
            Between base (black) and emitter (red): 3K ohms
            Between collector (black) and emitter (red): 10K ohms
            Between collector (red) and emitter (black): 7K ohms

            Q13 - [NPN S2090 Tremolo Oscillator]
            Between base (red) and collector (black): 12K ohms
            Between base (red) and emitter (black): 12K ohms
            Between base (black) and collector (red): 2.5K ohms
            Between base (black) and emitter (red): 10K ohms
            Between collector (black) and emitter (red): 4K ohms
            Between collector (red) and emitter (black): 6K ohms

            From what I've read, these need to be replaced? Sorry for all the questions; I haven't messed with stuff like this for years.

            ~~JoJo Manee


            • #7
              That was a lot of work, thanks, but that doesn;t tell us much. We would prefer your meter set to diode test and the voltages reported. But even then, a huge list of numbers is confusing. The thing to do is check the DC levels on teh schematic against teh circuit, then apply a steady signal to the amp input, and then trace signal level through the stages.

              If there are any jacks such as preamp out, see what signal is there. If there is a power amp in jack, connect a signal there and see what happens.
              Education is what you're left with after you have forgotten what you have learned.


              • #8
                I tested C104 (4000UFD 50V) with my ohm-meter and it started at 0 ohms, charged til about 2K, and then stayed there.

                I couldn't get to the leads of the other large caps on the chassis; they are hot glued in with an inordinate amount of glue, and I can't figure where they are connected to the circuit board.

                Also, I checked the transistors:

                ->Q3 is correct
                ->Q8 is 2N3053 but the only labeling I can make out says 861B
                ->Q9 is supposed to be a 2N2148 but is labeled 2N301 Clevite 120
                ->Q10 is supposed to be a TA 2577 and I dunno if it is the right one or not. The case is labeled 274/FS/417/101/BA. The 274 is in the middle and the other groups of letters are arranged about the outside edges of the case.
                ->Q11 is correct
                ->Q12 is correct
                ->Q13 I could not determine

                The other transistors either did not have markings, or were positioned in such a way that I could not see them.

                ~~JoJo Manee


                • #9
                  UNless the thing was built wrong and never worked, we won;t likely find the wrong parts in it. Besides, transistors are remarkable flexible. You can use almost any signal transistor, as long as it meets voltage and current specs, in such circuits and they'll work.

                  The way to solve this problem is systematic troubleshooting - follow the signal through the amp.
                  Education is what you're left with after you have forgotten what you have learned.


                  • #10
                    Wow, that was a lot of work you did there.

                    From what I can decipher from all of the resistance readings, the only transistors that may seem suspect are Q8 and Q9, but these are in circuit readings and all of the results could be off. As Enzo pointed out, if you meter has a diode test these readings would be more useful.

                    This is an unusual output amp circuit in that the two output transistors are not the usual matched pair; the TA 2577 NPN is Silicon while the 2N2148 PNP is Germanium. The Heathkit part number for the TA 2577 is 417-101, so the transistor you have is an original. My guess is that the original PNP has been replaced with the 2N301, which should work as a replacement.

                    I have seen the PNP replaced with a Silicon, which will not turn on early enough and will cause huge amounts of crossover distortion.

                    Testing the caps with an ohmmeter may tell you if they are shorted, but is not really useful for judging the quality of the cap itself.

                    Do as Enzo suggests, and carefully start taking voltage readings and compare them to what is marked on the schematic. This would be the best next step in the troubleshooting process.


                    • #11
                      I built this amp many years ago and still own it. Bill is correct with respect to the unusual output circuit. The transistors should be left original unless you are up to changing the bias circuit.

                      The orther advice given is pretty spot on - I have a few things to add.

                      While servicing this amp it is important to remember that this is about a 40 year old amp. Unless previously changed, most of the electrolytics are dried up or are in the process of drying up. It does not hurt to consider replacing all of the electrolytics as part of the renewal process. The cap C21, if open, will cause significant distortion. Replacing the rest will cure a variety of problems that will be difficult to diagnose or be intermittent.

                      R38 and R39, the emitter resistors, should be checked for opens.

                      Loss of the tremelo function is unfortunately the usual condition of these amps. The light dependent resistor R56 will have the light bulb open up and there is no replacement of which I am aware. With some creativity a substitute circuit could be cobbled up with either a FET or some light source and an LDR but I would get the amp working first.

                      Most important, as stated below, an organized approach with some clean signal source and an O'scope will tell the story pretty quickly as to the current state of the amp. I just wanted to add the other suggestions as fixing the current problem will probably just lead to new problems without some preventive maintenance.


                      • #12
                        I hope I'm not too late to help you with your Heath TA-16 amp. Any way I'm 60 and getting ready to retire soon so I wanted a hobby when I quit working and so I played a little guitar in Junior high school and I always wanted to play again I'm going to try relearn how to play. My wife bought me a Fender guitar for fathers day. So now I need an amp, I was going to build one, I just repaired a small Silvertone amp for a friend and kind of suped it up. I pulled out the bad 14 watt output IC and old power supply. Installed a bigger power supply and a 38 watt output IC and a 6 by 9 Rockford Fosgate car speaker. You now can actually use this amp on stage in a small room now. I work as a truck mechanic but took 3 years of electronics in High School when most electronics had tubes in them. Transistors were just starting to get traction, our electronics teachers said transistors won't ever take over tubes. They also said computers will never go main stream because they were so big, expensive, slow and needed to be in a climate controlled room only large companies can afford them. They were wrong on both. I didn't feel like building an amp mostly because I don't like to do wood working much any more and you need some kind of cabinet to put it all in. I wasn't about to plop down 500 to $1,000 on an amp. I looked on Ebay and anything that was any good costed almost that much. I went on Craig's list in less than 1 page I found this Heath amp, the owner said make me an offer. I have built many Heathkit, Eico and Dynaco kits from a Heath bug zapper, stereos, radios, equlizers to color TVs. I jumped on it gave him $40.00 and was off. I knew it didn't work and looked in rough shape. I got it home pulled the amp out of the cabinet hooked up a speaker looked things over plugged it in and flipped the switch. The pilot light came on and not much else. I'm not a seasoned electronics tech I do it as a hobby I also build and repair computers. So I drag out the DVM and read the voltage at the main filter cap, it was about 37 volts. So the power supply was ok, I unplugged it and gave the amp a good eyeballing, I already had a schematic I got on line I found a burnt resistor supplying power to the pre amps. Replaced the resistor plugged it in and touched the input jack and got a nice loud hum from the speaker, which was going to blow out if I kept it up. I got some longer wire and connected the amps 2 speakers. I found that the reverb tank was missing so I had 2 of them on a shelf one I gave a friend last year for his Fender twin Reverb amp I kept the other, which I had maybe 40 years. Hook up the tank, turned up the control, no reverb so I gave the tank a slight shake and got a nice loud crashing sound. I know the reverb is going into the amp, I checked the output from the amp to the tank with a scope and get nothing. Next I tried the Tremolo that didn't work either. I checked all the voltages in the tremolo circuit and only 2 voltages were even close. I have read most of these blogs, some from techs and they sound like they know thier stuff. The first thing I looked at was the caps, I unsoldered one end of all the caps in the circuit. Every one checked within tolerance some were right on the money. Now what, I checked the 2 transistors both checked out good. There is another part in this circuit called a LDR I had to go on line to find out wht that was. What I found is its a "light dependant resistor" or a photo cell, same thing as in lights that come on in the dark. I did some reading and checked this part and it did what was supposed to do. Simply Transistor Q13 turns on a small lamp that shines on a photo cell which when sees light the resistance drops. The photo cell is connected to the input of the power amp on one end and is grounded on the other. So when the lamp lights the volume will drop, blink the lamp on and off will give you tremolo which you can vary wtih the rate and depth controls. I connected a photo cell to the board where this LDR goes turn on the amp, feed in a signal and shine a flashlight at the photo cell, the volume drops wave the light past the cell fast you get a half ass tremolo effect. But the circuit don't work that controls this thing. There was nothing else to check so I go after the resistors, lifting one end on all of them. There was the problem, every resistor in that circuit was off, not by a few ohms some were off by thousands of ohms. A 47K read 55.5k, a 22 ohm read 26 ohms, a 10K read 12,010 ohms a 560K read 644K and a 6.8K read 8.2k. I replaced all the resistors re connected all the caps and transistors. So back to this LDR thingy, I already had a photocell soldered in I wanted to experiment a little I took one of these new super bright LEDs soldered to a 1K resistor in series and soldered it in place of the bulb in the LDR. Powered up the amp and it started doing what it was designed to do, the LED blinked as I played with the rate and depth controls the LED varied in speed and intensity. I put the LED in front of the photocell and wrapped some electrical tape around it to hold them in place and block out the light from the room the tremolo worked perfect. But we had one snag when I turned the rate and depth controls all the way off the LED stayed lit, shining light on the photocell which made the volume drop. So some one wrote in about loss of volume in one of these amps they couldn't track down that would be something to check. If the bulb in the LDR stays lit you'll loose volume, I would suggest remove the LDR and see if the volume goes up. Being LEDs draw much less current than an incandesant bulb I had to mimic a bulb being in the circuit. I started with a 4.7k resistor connected across the connections on the board. 4.7k did nothing, I tried a 2k still no change, in went a 1k, now I see a change. I tried the controls and everything still worked but the LED still stayed on but not as bright. I put in a 750 ohm resistor now the LED just barely glowed but it was enough to set off the photo cell. I tried 380 ohm resistor, too much, now the tremolo didn't work right. I put in a 470 ohm resistor and hit it right on the head. The LED goes out with the controls all the way down and the tremolo starts to work when the controls are turned up a little. Now to refine the package the whole thing is in. I ordered all new caps and have most of the resistors the ones I don't have Radio Shack MIGHT have, maybe thats all whats wrong with the reverb is all the resistors are out of range. Going to recover the cabinet, replace the 1/2" particle board the amp sits on with a piece of 3/4 oak plywood, cut a piece of metal from a computer case and make a bottom for the amp to cut down on hum, replace the bridge rectifier, a new 3 prong cord, a jack to plug in the speakers, new grille cloth. clean up the wiring from the board to the controls maybe replace it all with shielded wire. I have 2 12" Ibanez speakers I bought 4 of them from a kid who plays head banging music and wanted bigger speakers and hopefully have an amp that will outlast me. Thanks for letting me share my experience with a Heathkit amp and hope it can help someone out there with the same amp. By the way I read a blog about a replacement for a Heath S2090 transistor that is no longer available and can't find a replacement. By accident or its just I had one handy I used a 2N2222 in the metal can in place of that S2090 and it worked perfect. The amp was on with the tremolo all the way up for 2 hours and didn't get warm, I checked the voltages and they were all in range.


                        • #13
                          Nice post.
                          If the reverb makes a crashing sound when the pan is bumped, then the the return circuit is good.
                          That leaves the send circuit to be checked out.
                          I do not have the schematic so that is as far as I can go.


                          • #14
                            Welcome. Your reverb problem may be related to the tank's input impedance. I'd have to look up the schematic, but I believe that the amp needs a high impedance tank. Do you know what tank you have there?

                            A quick check would be to read the dc resistance across the tank's RCA jacks. I think that the tank that you need will have an input resistance somewhere between 60 and 180 ohms.


                            • #15
                              Hi Jazz
                              I dug into the reverb driver circuit today, hooked up the tone generator to the input and checked for a signal at the jack going to the reverb tank with the volt meter on the lowest setting I read about .01 volts, thats not enough to run the reverb. I hooked up the scope and the signal was so weak it barely showed up on the scope. I even connected an audio amp and speaker to the output jack and I could barely hear any sound. I did the same thing, start checking the resistors. The few I got to were way off, 176k instead of 150k, 762 ohms instead of 680, 13 volts at the collector of the reverb driver instead of 9 volts. I guess I'll have to replace all the resistors in the amp to get it to work right, it's not that bad of a job and have had to do alot worse jobs, at least the parts don't weigh hundreds or thousands of pounds like I'm used to. There's not that many parts and they're all easy to get to, what is nice, I dont think it was meant too be this way but I shine a flashlight under the board you can see all the traces when you look on top of the board so its very easy to follow the circuit when I look at the schematic. The board is not made to well if I 'm not careful the copper trace will come off the board very easy but its easy to repair, there is alot of room between the traces nothing like a computer board. I don't know what they're soldering these computer boards with but I have many boards laying around and tried to take the caps off them because some are the same value as the ones I need and there real good ones. I tried to unsolder them and the cap got so hot I couldn't touch it and finally the lead pulled out. I gave up on that it's easier to just buy new ones. Had a bit of good luck though, this amp has about 10-10 mfd 15volt eletrolytic caps and several .002 film caps, and old VCR board was a gold mine. I got all of the 10mfd and .002 pluse some .01, .1, .22, .47 and the 2-100mfd 15 volt caps off this one board. I checked some of the caps I scabbed off this board and not a one was exactly what it should be but within tolerance. So far all the caps I've checked from the Heathkit's board were right on, if it reads .47 on the cap, the meter reads .47. If the cap reads 100mfd the meter reads 100mfd, thats hard to believe after 46 years. Several years ago a friend gave me a Harmon Karden Citation II amp, pre amp and tuner he built as kits in 1961. I was still a little green playing with electronics, I put the amp on the bench hooked up 2 speakers plugged it in and flipped the switch, then stood there enjoying a mini light show in the 4 English made Mullard KT-88 tubes, then the 6 amp fuse blew, that was the end of the show. I thought the speakers blew too. So I turned over the amp removed the bottom and found a few caps actually leaked. I checked all the caps many were shorted and some were open. Ones that were not were not even close to the proper value. I got to replacing every cap in that amp and at that time it was very hard to find 30mfd 550volt caps but I did track them down, and they were expensive. Now you can buy them from most electronics supply houses because so many people are into restoring old tube equipment now. Many of the resistors were way off too, so they got all replaced. When I finished I was a little scared to plug it in because I don't have an autotransformer but I did and the tubes started thier little light show again and the fuse blew. I checked and rechecked everything the last thing was the tubes, I took them to someone who had a tester and all 4 were shorted. I didn't feel like spending over $1,200 on 4 original tubes so I got 4 Chinese made 6550s for about $80,00. That amp was used as a subwoofer amp for a surround sound system in my den for years. The amp was actually in my garage because it was cooler and it was the only place I had the room for it. I finally switched to solid state a few years ago and those Chinese tubes were giving out. I put that amp, preamp, tuner and a old Lafayette power amp on Craigs list and it was sold in less than an hour. I packaged up the LDR I made up for the Heathkit with some heavy duty shrink tubing and a little black RTV sealant on the bottom of the LED, soldered the thing in and it works perfect. I'll keep you guys posted on how I do with that Heathkit amp. Thanks