Hey guys!
I could really use some help in this. Tonight I lifted the base resistors on the output stage just to be sure it wasn't involved in any way. Also the power is not connected to the output stage, and either is the GND.
As I powered the card up, both driver transistors blew, as well as the fuses on the supply to voltage amp. I have soldered the transistors off and they have a short between all three legs.
How can I debug this? Please help!
Best regards,
/Bo
EDIT: I measured the voltage on the base of the driver transistors, which was 1.3V on the 15032 and -1.3V on the 15033. Is this a normal base voltage on the drivers or would this make them blow?
I could really use some help in this. Tonight I lifted the base resistors on the output stage just to be sure it wasn't involved in any way. Also the power is not connected to the output stage, and either is the GND.
As I powered the card up, both driver transistors blew, as well as the fuses on the supply to voltage amp. I have soldered the transistors off and they have a short between all three legs.
How can I debug this? Please help!
Best regards,
/Bo
EDIT: I measured the voltage on the base of the driver transistors, which was 1.3V on the 15032 and -1.3V on the 15033. Is this a normal base voltage on the drivers or would this make them blow?
I'm an idiot.
Looking at the datasheet for the MJE15032/3 it's obvious that nothing good will come with 1.3V base-emitter voltage. I measured the other card and it was 0.61V. It still puzzles me that they blew though. The datasheet says they are good for 5A, and with R118 and R119 at 27R the current should have been approx 2A, which is well below.
Anyway, I trimmed the bias until I had 0.4V on the bases of the drivers and then soldered them on. All is well and I slowly went upwards, continually observing the base voltage as well as tpA-tpC until I got approx 1.35V tpA-tpC.
Base resistors going back now, and then final test once again!
Best regards,
/Bo
Looking at the datasheet for the MJE15032/3 it's obvious that nothing good will come with 1.3V base-emitter voltage. I measured the other card and it was 0.61V. It still puzzles me that they blew though. The datasheet says they are good for 5A, and with R118 and R119 at 27R the current should have been approx 2A, which is well below.
Anyway, I trimmed the bias until I had 0.4V on the bases of the drivers and then soldered them on. All is well and I slowly went upwards, continually observing the base voltage as well as tpA-tpC until I got approx 1.35V tpA-tpC.
Base resistors going back now, and then final test once again!
Best regards,
/Bo
This weekend I finished up the amp and finally it worked! 😀
I say worked, because it has broken down again.
I used my laptop as signal generator, playing some music from my music server. As I noticed some humming I was to make sure that the 3.5mm stereo plug into the computer was properly seated. When I wiggled it, both speakers gave out a loud pop and then there was silence.
The left channel is fried, both the driver transistors are gone, again. Jan, how is this possible? I was under the impression that the 15032/3 was quite rugged and could withstand some 5A current. Withe the power resistors at 27R, how can there be over 5A in these components? My 60V would just yield 2A.
I have ordered new ones from RS, as well as some 15034/5 as they seem to be even more rugged.
Best regards,
/Bo
I say worked, because it has broken down again.
I used my laptop as signal generator, playing some music from my music server. As I noticed some humming I was to make sure that the 3.5mm stereo plug into the computer was properly seated. When I wiggled it, both speakers gave out a loud pop and then there was silence.
The left channel is fried, both the driver transistors are gone, again. Jan, how is this possible? I was under the impression that the 15032/3 was quite rugged and could withstand some 5A current. Withe the power resistors at 27R, how can there be over 5A in these components? My 60V would just yield 2A.
I have ordered new ones from RS, as well as some 15034/5 as they seem to be even more rugged.
Best regards,
/Bo
well that strange
in my lynx amp, i use 2SA1837/2SC4793 for driver with lower SOA than MJE15032/15033 . but everything is OK.
in my lynx amp, i use 2SA1837/2SC4793 for driver with lower SOA than MJE15032/15033 . but everything is OK.
The only thing I can imagine is that there is a lot of current in the base-emitter junction of the 15032/3's. But where does that current come from?
Obviously the spike on the input has generated a big signal but surely it should not kill the amp??? If the output fuses had blown, OK, but this?
Best regards,
/Bo
Obviously the spike on the input has generated a big signal but surely it should not kill the amp??? If the output fuses had blown, OK, but this?
Best regards,
/Bo
Bo;
Sorry for this short answer, however I'm sadly very busy these days 🙁
For some reason you still suffer from heavy oscillations 😕
What fuses do you use for the output devices ?
Note that the drivers are connected to the +/- rails supplying the VAS stage. Have you placed any fuses on those rails ?
For testing try R101=10K. This will make a 1/10 voltage divider on the input lowering the input to the opamp.
Sorry for this short answer, however I'm sadly very busy these days 🙁
For some reason you still suffer from heavy oscillations 😕
What fuses do you use for the output devices ?
Note that the drivers are connected to the +/- rails supplying the VAS stage. Have you placed any fuses on those rails ?
For testing try R101=10K. This will make a 1/10 voltage divider on the input lowering the input to the opamp.
ACD said:
You mean that the oscillation was triggered by the spike? Because before that it worked like a charm. Also, I don't see any oscillations on the scope anymore. I now use the OPA627 and I haven't seen any oscillations since the swap.
I still have small F1A fuses on the outputs. I also have F1A on the input rails, both channels.
Best regards,
/Bo
Bo;
Sounds very strange though 😕
Many use the LYNX as guitar and bass amp (and you know how those guys just plug their gear in and out all the time)
Was it running very hot before it blew up ?
Sounds very strange though 😕
Many use the LYNX as guitar and bass amp (and you know how those guys just plug their gear in and out all the time)
Was it running very hot before it blew up ?
It was not hot at all. Very moderate listening level. You can tell that it's on by touching the cooling fins, but just barely.
Could it be fake 15032/3's? You wrote about it in the QAG.
Another theory: As all four fuses on this channel blew; if the output fuses blew first, because of the large spike, could that result in the death of the 15032/3? I have a feeling that this amp is relying on the output stage being functional.
Best regards,
/Bo
Could it be fake 15032/3's? You wrote about it in the QAG.
Another theory: As all four fuses on this channel blew; if the output fuses blew first, because of the large spike, could that result in the death of the 15032/3? I have a feeling that this amp is relying on the output stage being functional.
Best regards,
/Bo
Last night I soldered the faulty drivers off. Then I tried to find if there was anything wrong with the card, by feeding it power on the VAS-side, but not on the output.
There are some huge voltage differences between the two sides and if I measure between input GND and output, I get +45V, as if the amp has "tipped over" to one side and stays there.
Is it at all possible to run the input stage without the drivers? Any pointers would be helpful, thanks!
Best regards,
/Bo
There are some huge voltage differences between the two sides and if I measure between input GND and output, I get +45V, as if the amp has "tipped over" to one side and stays there.
Is it at all possible to run the input stage without the drivers? Any pointers would be helpful, thanks!
Best regards,
/Bo
Bo,
it might be nice if you measured all the DC voltages in your Lynx, and send it to Jan. This might help Jan a lot.
it might be nice if you measured all the DC voltages in your Lynx, and send it to Jan. This might help Jan a lot.
PMA,
OK. I will do that. Thanks for helping out. The other channel works like a charm, I just have to nail this last bit.
Best regards,
/Bo
OK. I will do that. Thanks for helping out. The other channel works like a charm, I just have to nail this last bit.
Best regards,
/Bo
OK, this is what I have found.
The +45V on the output was because I had forgotten to ground the input GND. It was grounded before but when I took out the card for desoldering the drivers, I forgot to add input GND.
Most of the voltages all seem ok, I have +/-15V on the OPAMP and +/-62V at the rails. The output stage is still not connected.
BUT: There is something wrong with the voltages at the driver transistors. The voltage across B-E was 1.2V, which is way too much. I control measured the other channel, and those were 0.6V.
So, once again, I adjusted the bias pot to lower the B-E on the drivers, powered off, and resoldered the drivers.
I can now start up the input stage. I have 0.6V at B-E on the drivers, but no way near 1.4V at tpA-tpC. I can't risk going up to that level again, since the drivers are clearly over biased at that level.
So, the problem now is that I can't raise the bias without the driver transistors being over biased. Is this any clue to anyone?
Thanks again for all help!
Best regards,
/Bo
The +45V on the output was because I had forgotten to ground the input GND. It was grounded before but when I took out the card for desoldering the drivers, I forgot to add input GND.
Most of the voltages all seem ok, I have +/-15V on the OPAMP and +/-62V at the rails. The output stage is still not connected.
BUT: There is something wrong with the voltages at the driver transistors. The voltage across B-E was 1.2V, which is way too much. I control measured the other channel, and those were 0.6V.
So, once again, I adjusted the bias pot to lower the B-E on the drivers, powered off, and resoldered the drivers.
I can now start up the input stage. I have 0.6V at B-E on the drivers, but no way near 1.4V at tpA-tpC. I can't risk going up to that level again, since the drivers are clearly over biased at that level.
So, the problem now is that I can't raise the bias without the driver transistors being over biased. Is this any clue to anyone?
Thanks again for all help!
Best regards,
/Bo
Bo, i got an idea from Andypairo in the Krill thread.
If you haven´t put the MJEs properly on the heatsink, they get warmer so they pass some more current.
If you haven´t put the MJEs properly on the heatsink, they get warmer so they pass some more current.
Bo;
Strange ?
A Vbe at 1.4V would mean a lot of Current going through the collector and the MJEs would go up in
One could suspect faulty MJEs.......
Do you have the Base resistors and the output devices solered in?
1.4V over tpA and tpC in your case only gives app. 26mA bias Current, which is nothing is the devices are proberly heatsinked....
Strange ?
A Vbe at 1.4V would mean a lot of Current going through the collector and the MJEs would go up in
One could suspect faulty MJEs.......
Do you have the Base resistors and the output devices solered in?
1.4V over tpA and tpC in your case only gives app. 26mA bias Current, which is nothing is the devices are proberly heatsinked....
ACD said:Bo;
Strange ?
A Vbe at 1.4V would mean a lot of Current going through the collector and the MJEs would go up in
One could suspect faulty MJEs.......
Do you have the Base resistors and the output devices solered in?
1.4V over tpA and tpC in your case only gives app. 26mA bias Current, which is nothing is the devices are proberly heatsinked....
Base resistors and output devices are soldered in.
Everything is properly heatsinked. In fact I have yet to feel any heat at all on the drivers.
I want 1.4V tpA-tpC but I'm afraid of going there, since that means too high Vbe (1.2V).
Best regards,
/Bo
I upped the bias some. Now I have tpA-tpC = 1.28V. Vbe on Q110 = 0.62V and Vbe on 0.603V.
Are these sane values of Vbe? Should the difference be that much?
This is exactly like last time I got this card running. All seems fine, and then boom. I'm a bit worried (to say the least) that it's gonna break down again.
Best regards,
/Bo
Are these sane values of Vbe? Should the difference be that much?
This is exactly like last time I got this card running. All seems fine, and then boom. I'm a bit worried (to say the least) that it's gonna break down again.
Best regards,
/Bo
- Status
- Not open for further replies.