A 10k would work.
Tie the wiper of the pot to the right-most leg (legs down, looking at shaft). The outer legs would go into the board.
The low level oscillation won't generally cause any problems and probably isn't audible but it shouldn't be there.
The instability is likely from too little damping in the circuit. If you get it biased properly, the oscillation should go away.
Tie the wiper of the pot to the right-most leg (legs down, looking at shaft). The outer legs would go into the board.
The low level oscillation won't generally cause any problems and probably isn't audible but it shouldn't be there.
The instability is likely from too little damping in the circuit. If you get it biased properly, the oscillation should go away.
well I know this is forever old but I finally picked this amp up again. I soldered in the 10 k pot and adjusted it until I got .000 volts across the emitter resistor.I ran the amp for about 25 minutes as loud as my power supply would let me go and the amp heated up a bit and I re checked the bias and it was the same. The amp appears to play fine but the newly repaired channel gets hotter than the original channel. it isn't way hotter but it is a good ten to fifteen degrees hotter. Does this indicate the bias is incorrect? I'd also like to note that the original channel bias measures at .001 volts but after about 20 minutes it was sitting at .008 volts. I have never had a problem with that channel. Is it worthwhile to re adjust the bias for the unrepaired channel? Should I try to match the newly repaired channels bias to the original? I have yet to put the o scope on it yet to see if that low level oscillation was still present.
I am beginning to think this amp has some kind of curse. I was putting it on the scope last night and then went to check the bias and i slipped off one of the emitters of one of the outputs and shorted it to the heatsink. The output looks really funny now when i run a sine wave on either channel. Just another thing to fix.
In the owner's manual, the emitter voltage is 0v. Since they are providing voltages accurate to the mv in the same chart, I assume that the bias voltage may be OK at 0.000v. If you can find the reason for the oscillation and can get a clean signal with the bias at no more than 0.001v across the emitter resistors when the amp is hot, that's probably how you should set it.
If both channels are distorted, you may have damaged a ground trace somewhere. Does the RCA ground still read ~0 ohms to the non-bridging speaker terminals?
How does it sound?
Try connecting your scope ground to one of the non-bridging speaker terminals.
If both channels are distorted, you may have damaged a ground trace somewhere. Does the RCA ground still read ~0 ohms to the non-bridging speaker terminals?
How does it sound?
Try connecting your scope ground to one of the non-bridging speaker terminals.
Found it! When I shorted the emitter to ground it blew the shield ground in my test head unit. I repaired that and the output looks normal again and I am not seeing any oscillation after setting the bias. I haven't been able to fully heat it up again yet so I'm not sure if the issue with one channel being warmer than the other is still there but like I said in an earlier post its about a 15 degree difference. It worries me a bit but for all I know it could be normal.
to the touch it doesnt appear that one side gets hotter any faster than the other. when i use my temp probe i am only seeing maybe a 5 to 10 degree difference between the channels and it stays consistant regardless of how long or loud its playing. The bias voltage is holding better in the repaired channel than the original. I ran the amp at approximately 50 percent power for 25 to 30 minutes and heated it up until it was too hot to touch and it appears to be holding up well so far.
Its a volt or two if I remember correctly. I will have to check it in a bit when I do some other testing and I will have the exact difference then. I have a feeling my test deck has different rca voltage from one channel to the other and this may be the reason for the difference in output.I really need to find a new test unit.
Swap the RCAs or measure the voltage directly on the RCA cables to determine if it's the head unit.
If it's not the head unit's fault, disconnect the load, drive the amp to near clipping and measure the voltage there. Post the voltage you read on each channel. Measure directly across the speaker terminals for each channel.
If it's not the head unit's fault, disconnect the load, drive the amp to near clipping and measure the voltage there. Post the voltage you read on each channel. Measure directly across the speaker terminals for each channel.
Measured the rca output of my deck and they are pretty close. They are within millivolts of each other. The difference in output between the 2 channels on the amp is approximately 1.3 volts ac regardless of which rca is feeding the channel. I noticed the rail votage is slightly higher in the repaired channel. The original channel is +/- 45.3 volts and the repaired channel is +/- 45.8 volts. Is the difference is rail voltage significant at all? Also i should note the gain pot is damaged. It appears to work correctly but it spins all the way around. It does not stop in the max or minimum position
yes the one channel does run hotter bridged. I think I know at least part of the reason why now. I disassembled the amp last night to clean up the heatsink and add some new compound and it appears that at least one of the outputs is not seated properly. Even with it screwed down I can push down on it and see the compound squeeze out from underneath it. I have to resolder the leads so that it will seat properly and/or come up with a better way to clamp it down.
Desolder it completely and remove it to confirm that there is no foreign material between it and the sink or between the sink and the insulator. Reapply heatsink compound (not too much) and tighten the mounting screw. Apply pressure to the face of the transistor near the legs to displace as much compound as possible. When soldering the center leg, apply a bit of pressure to the leg and bend it over on the bottom of the board and solder the connection. This will ensure that there is a tiny bit of spring tension pulling the transistor to the sink. Only do this for the center leg. Solder the others with no tension.
I removed all the outputs in that channel and checked for debris. None to be found. I then reapplied heatsink compound and soldered them back in as you described. They still seemed kind of loose to me so I fashioned a sort of spring clip/clamp down to hold them in place. I had to use a larger screw for one of the outputs as the hole was stripped from trying to tighten it down too much. Turned out this and or my clamp down was a big mistake. I played the amp for a few minutes and then that output failed. When I removed it I could see that the new screw had pulled a nice size metal shaving out and left it under the transistor.
Seeing as all the holes for the output mounting screws were starting to strip I had to come up with a new way to hold them down. I drilled through the original hole with a slightly larger drill bit. Then I countersunk the bottom of the little heatsink plate just big enough to fit the head of a 4-40 flat head machine screw.On top I am using a nut,lockwasher,and flat washer to hold the transistor in place. Now I can clamp them down tighter than ever and not have to worry about them stripping again. Using this method in addition to soldering as you described earlier Perry I am happy to say that the outputs are now seated firmly.
Instead of the original 2sa1265 and 2sc3182 I am trying out fja4210 and fja4310 as outputs just because I happen to have a few on hand and fairchild list them as a direct replacement for the originals. I haven't been able to run it hard yet but at low power it seems to be working well.The output is clean and I am not seeing any of that oscillation that I had before. Also now the output is a lot more closely matched from channel to channel. At the most I am seeing .5 volts difference but over much of the range it is closer than that. In addition to replacing the outputs I also had to replace one of the small driver transistors mounted next to the outputs. The way things have been going with this amp I almost postive once I get this channel squared away the other one will fail shortly after. I refuse to let this thing defeat me though plus it gives me something to play with in my spare time.
Seeing as all the holes for the output mounting screws were starting to strip I had to come up with a new way to hold them down. I drilled through the original hole with a slightly larger drill bit. Then I countersunk the bottom of the little heatsink plate just big enough to fit the head of a 4-40 flat head machine screw.On top I am using a nut,lockwasher,and flat washer to hold the transistor in place. Now I can clamp them down tighter than ever and not have to worry about them stripping again. Using this method in addition to soldering as you described earlier Perry I am happy to say that the outputs are now seated firmly.
Instead of the original 2sa1265 and 2sc3182 I am trying out fja4210 and fja4310 as outputs just because I happen to have a few on hand and fairchild list them as a direct replacement for the originals. I haven't been able to run it hard yet but at low power it seems to be working well.The output is clean and I am not seeing any of that oscillation that I had before. Also now the output is a lot more closely matched from channel to channel. At the most I am seeing .5 volts difference but over much of the range it is closer than that. In addition to replacing the outputs I also had to replace one of the small driver transistors mounted next to the outputs. The way things have been going with this amp I almost postive once I get this channel squared away the other one will fail shortly after. I refuse to let this thing defeat me though plus it gives me something to play with in my spare time.
When doing this type of machining (which is often required), the drilling creates a small raised area around the perimeter of the hole. This needs to be knocked down so that the transistor can lay flat on the sink. Sometimes it's not obvious unless you sand the area lightly with fine sandpaper. Then it's obvious and you can see how much material needs to be removed.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.