Hello !
I've got a broken L-3. The right channel fuse keep blowing.
So far I've checked :
1) power circuit is good at 38.0 V on +B and -B at idle without LS. Also the -46 V is OK
2) as the fuse on the speaker line blows, I've checked that the switching to select A or B speakers has no fault.
3) I've checked the right channel power transistors. They are fine on my little Chinese tester.
There is no mechanical damage nor obvious leaking capacitor. The amp appears like it has never been opened.
All I can do and check has been checked. So I _do_ need your help and advice !
Many thanks in advance for your help !
P.S. : the amp is version E with the 2550 transformer (which was set for 220 V and I switched it to 240 V as I live in Europe.)
I've got a broken L-3. The right channel fuse keep blowing.
So far I've checked :
1) power circuit is good at 38.0 V on +B and -B at idle without LS. Also the -46 V is OK
2) as the fuse on the speaker line blows, I've checked that the switching to select A or B speakers has no fault.
3) I've checked the right channel power transistors. They are fine on my little Chinese tester.
There is no mechanical damage nor obvious leaking capacitor. The amp appears like it has never been opened.
All I can do and check has been checked. So I _do_ need your help and advice !
Many thanks in advance for your help !
P.S. : the amp is version E with the 2550 transformer (which was set for 220 V and I switched it to 240 V as I live in Europe.)
What is the DC offset? This is the voltage at the amplifier output (on the speaker fuse). It should be zero.
Is the 12 volt supply OK (zener diode DZ401) and the components feeding it. Seem to recall the 1k5 (R406) can go high.
Is the 12 volt supply OK (zener diode DZ401) and the components feeding it. Seem to recall the 1k5 (R406) can go high.
In any case, from here to the point the amp is OK, and for security reason, use a limited current supply, which may be easily implemented putting a filament lamp (Nor led not fluorescent) in series to the amp. It is called series lamp, and protects the device and yourself for heavy malfunctioning in the device. The lamp must be of the same expected max load used, say, a 100W lamp for a 100W full output amp. When switching on you will see the lamp flash full bright for few seconds until caps become charged, and then the lamp must almost appears dark, in normal conditions.
Also suggest to use a lamp in the speaker output, whose voltage rating is above the rails of the amp, if something goes wrong it will bright and in any case it is cheaper to replace than a speaker blown.
Also suggest to use a lamp in the speaker output, whose voltage rating is above the rails of the amp, if something goes wrong it will bright and in any case it is cheaper to replace than a speaker blown.
Besides failed transistors, you have a couple pots that may be oxidized.
But you should study the schematic until you understand the purpose of every part of the power amp so that you can compare measured voltages with what it should be. If the fuse is blowing then you should have noticed a DC offset, then trace back through the circuit to determine why there is a DC offset. Trial and error usually results in burning up a lot of new parts.
But you should study the schematic until you understand the purpose of every part of the power amp so that you can compare measured voltages with what it should be. If the fuse is blowing then you should have noticed a DC offset, then trace back through the circuit to determine why there is a DC offset. Trial and error usually results in burning up a lot of new parts.
Hello,
Here I am again.
I had a senior moment because I removed the wrong channel output transistors to check .... The one on the failed channel are fried.
I've nearly 0V at the 19.2 zener on the faulty channel. The 1N4003 feeding it is good as is the zener. There is 19.5 V on the other channel at the zener.
I wonder if I can check voltages with the amp missing the zener, the 1N4003 feeding it and the output transistors ?
Also it is difficult to reach components on the amp with the radiator because the copper face of the printed circuit board is facing the radiator and on the other face, there are a lot of wires and cables and the rear panel.
I would not like to get shocked by discharging a cap on my hands or screwdriver or test probe ...
Testing for now will stop because I need a set of power transistors (I bet I can not find the original 2SB686 and 2SD716 given the amp is from 1979 according some date codes on parts).
Thanks for your help !
Here I am again.
I had a senior moment because I removed the wrong channel output transistors to check .... The one on the failed channel are fried.
I've nearly 0V at the 19.2 zener on the faulty channel. The 1N4003 feeding it is good as is the zener. There is 19.5 V on the other channel at the zener.
I wonder if I can check voltages with the amp missing the zener, the 1N4003 feeding it and the output transistors ?
Also it is difficult to reach components on the amp with the radiator because the copper face of the printed circuit board is facing the radiator and on the other face, there are a lot of wires and cables and the rear panel.
I would not like to get shocked by discharging a cap on my hands or screwdriver or test probe ...
Testing for now will stop because I need a set of power transistors (I bet I can not find the original 2SB686 and 2SD716 given the amp is from 1979 according some date codes on parts).
Thanks for your help !
Last edited:
I have to add :
The 1.5kOhm resistor is cut. The diodes are good. But I've lost the 1N4003 in the garage... So I put two 1.5kOhm resistor in both channels and replaced the to 1N4003 by two 1N4007 I had in stock.
Now, regarding the output transistors, is it possible to use TIP35C/TIP36C instead of the 2SD716/2SB686 the most notale differences being the power, and the transition frequency which is 3mHz for the TIP and 12/10 mHz for the Japanese ones...
Thanks a lot for your help !
The 1.5kOhm resistor is cut. The diodes are good. But I've lost the 1N4003 in the garage... So I put two 1.5kOhm resistor in both channels and replaced the to 1N4003 by two 1N4007 I had in stock.
Now, regarding the output transistors, is it possible to use TIP35C/TIP36C instead of the 2SD716/2SB686 the most notale differences being the power, and the transition frequency which is 3mHz for the TIP and 12/10 mHz for the Japanese ones...
Thanks a lot for your help !
Easily done 🙂
The zener... are you saying the actual diode is OK and it is just the 1k5 that is open circuit. I'm not sure what you mean when you say the resistor 'is cut'.
The TPI35/36C should be OK electrically, however you must first remember to set the bias preset to minimum resistance before powering on.
As has been mentioned, use a bulb tester initially. You also need to check the emitter resistors on the output transistors and I would also replace the driver transistors. BD139/140 could be a possible substitute even though they are a different package.
The zener... are you saying the actual diode is OK and it is just the 1k5 that is open circuit. I'm not sure what you mean when you say the resistor 'is cut'.
The TPI35/36C should be OK electrically, however you must first remember to set the bias preset to minimum resistance before powering on.
As has been mentioned, use a bulb tester initially. You also need to check the emitter resistors on the output transistors and I would also replace the driver transistors. BD139/140 could be a possible substitute even though they are a different package.
Yes, the actual diodes used on both channels are OK. And yes, the 1.5k is open circuit.
Is there a TO92 substitute for the drivers circuits ? I would like to keep the circuit "like" the original ...
You are right that I should check the emitter resistances.
I have a couple of 50 W 8Ohm resistors to use as load so I bet I'll pass on the bulb charge (I do not have any bulb in 60~80 W left at home)
Thank you for your help !
Is there a TO92 substitute for the drivers circuits ? I would like to keep the circuit "like" the original ...
You are right that I should check the emitter resistances.
I have a couple of 50 W 8Ohm resistors to use as load so I bet I'll pass on the bulb charge (I do not have any bulb in 60~80 W left at home)
Thank you for your help !
An 8 ohm speaker load is the worst thing you could use for initial testing because that would just pull lots of current in the event of a problem. Your initial tests have to be with NO speaker load.
The bulb limits the power that can be drawn from the incoming mains supply. Its foolish to just switch on and 'hope' as any small remaining issue could see new parts get zapped instantly.
I think the BC639/640 are like the BD139/140 but in a T092 package.
The bulb limits the power that can be drawn from the incoming mains supply. Its foolish to just switch on and 'hope' as any small remaining issue could see new parts get zapped instantly.
I think the BC639/640 are like the BD139/140 but in a T092 package.
I agree Karl. I always use lamps as load, for amplifiers, inverters, PSU (Linear os PWM), including large frequency inverters. So I minimize risks of explosions or general damages.
Can you imagine a large frequency converter rated for, say, 200Amper driving 3 star wired 100W lamps? Stupid, yes, but safe.
Can you imagine a large frequency converter rated for, say, 200Amper driving 3 star wired 100W lamps? Stupid, yes, but safe.
Thank you Mooly, I initially thought to test the amplifier without load to check for correct voltages, smoke and set proper DC offset and bias. I did not understand that the bulb was to protect mains power coming into the amplifier... I'm sometimes slow....
So I will search for a suitable bulb in the house. Around a hundred watts should be enough, isn't it ?
I'll check the equivalency of the BC639/640 to the 2SC2235/2SA965 and BD139/BD140
Thank you all for your help !
So I will search for a suitable bulb in the house. Around a hundred watts should be enough, isn't it ?
I'll check the equivalency of the BC639/640 to the 2SC2235/2SA965 and BD139/BD140
Thank you all for your help !
Use as series lamp, one that has more or less the same MAXIMUM output of the amplifier (or load). For a 70W amp, use a100W bulb. Under normal conditions, the lamp will flash when powering it, because of cap charging. Then, slowly the lamp will dim to any quasi unviewable level. If something goes wrong, the lamp will start lighting again.
Found the bulb. Have to make the jig to use it with the amp.
Stay tuned, it could take a couple of days (Alas, I can't spend too much time in the shed otherwise the missus will complain ;-) )
Stay tuned, it could take a couple of days (Alas, I can't spend too much time in the shed otherwise the missus will complain ;-) )
Once again, check that it is of the old filamentary type. Led or fluorescent (Compact or tube) do not work.
In any case, wire it in series with the live wire of the power outlet. An easy way to do suck a thing, is to catch a lamp socket, put a couple of wires of say, 20cm each. One end of both to the socket, and the others to a blown fuse of the same type of the amp uses. Then, get rid of the fuse and replace with the lamp and wires. Another way is to put a couple of crocodile clips and hook them to the fuse clips (Without the fuse in the holder). This is a bit dangerous than the former, but still can do the job.
In any case, wire it in series with the live wire of the power outlet. An easy way to do suck a thing, is to catch a lamp socket, put a couple of wires of say, 20cm each. One end of both to the socket, and the others to a blown fuse of the same type of the amp uses. Then, get rid of the fuse and replace with the lamp and wires. Another way is to put a couple of crocodile clips and hook them to the fuse clips (Without the fuse in the holder). This is a bit dangerous than the former, but still can do the job.
OK 🙂
(the bulb works so well because the filament has a positive temperature coefficient. If the amp draws little current, the filament stays cool and the resistance is quite low so the amp gets nearly full mains voltage. If the amp draws to much current, the filament heats and the resistance rises rapidly. The bulb glows or lights and the overall power supplied to the amp is greatly limited saving failure of output transistors)
That's quite ingenious !
I've some blown fuse from this amp 😀
Thank you very much for the tip !
Yes, I know that. I use them to test for old tube radios. But I often power them from a Variac, giving them a slow start to awake them slowly from their long retirment in attics or basements...
Hello,
Here am I again.
I've put 4 TIP35C/36C into the amp.
Two brand new 1k5 1 W resistors. Checked the voltages with the 60W bulb in place of the primary mains fuse.
It was a walk in the park to set the polarizing voltages, and the two 19.2 voltages are perfect.
But if I am able to set the left channel output voltage at 0.0 V the right (which was faulty) cannot go under the 25 mV value.
I wonder if I should be concerned by this low voltage ? What do you think ?
I'll wait for your advice ;-) and I have a set of speakers and a CD player ready to test this fine piece of gear !
Thanks a lot for your help !
Here am I again.
I've put 4 TIP35C/36C into the amp.
Two brand new 1k5 1 W resistors. Checked the voltages with the 60W bulb in place of the primary mains fuse.
It was a walk in the park to set the polarizing voltages, and the two 19.2 voltages are perfect.
But if I am able to set the left channel output voltage at 0.0 V the right (which was faulty) cannot go under the 25 mV value.
I wonder if I should be concerned by this low voltage ? What do you think ?
I'll wait for your advice ;-) and I have a set of speakers and a CD player ready to test this fine piece of gear !
Thanks a lot for your help !
±30mV at the outputs are tolerable, but the 25mv happens at the middle of the adjusting pots or in an extreme? I strongly suggest to replace the driver too, because although they are (or not appearing as) faulty, they may be partially degraded. As the amp is not able to give 0.00V at the outputs, it is not advisable to realize audio tests still. First fix the DC points before.
Last edited: