Hi all,
Just finished building Jan Dupont's Lynx amp found at http://www.audio-circuit.dk. i used a +/-35V test power supply and measured 3V at the output. Idle starts at 75mA and cannot be lowered anymore. If I connect the Zobel network, the DC at the output becomes 5-6V and the current goes 300mA+.
I've checked everything, even the resistors. They all passed. All transistors are ok, OPS have no short to heatsink.
Am I missing something? Please help me.
Regards,
JojoD
Just finished building Jan Dupont's Lynx amp found at http://www.audio-circuit.dk. i used a +/-35V test power supply and measured 3V at the output. Idle starts at 75mA and cannot be lowered anymore. If I connect the Zobel network, the DC at the output becomes 5-6V and the current goes 300mA+.
I've checked everything, even the resistors. They all passed. All transistors are ok, OPS have no short to heatsink.
Am I missing something? Please help me.
Regards,
JojoD
Jan,
I've tried disconecting the power transistors already, the dc at the output only decreases to 1.4-1.8V dc.
When the power transistors are connected, I cannot connect the zobel network or even a speaker because the dc only gets higher!
Surprisingly, at +/-12V i can connect a speaker and hear a hum when I touch the input which is normal right.
I'm out of ideas and any suggestion would be helpful...
JojoD
I've tried disconecting the power transistors already, the dc at the output only decreases to 1.4-1.8V dc.
When the power transistors are connected, I cannot connect the zobel network or even a speaker because the dc only gets higher!
Surprisingly, at +/-12V i can connect a speaker and hear a hum when I touch the input which is normal right.
I'm out of ideas and any suggestion would be helpful...
JojoD
Please examine your pin connections to Bias transistor Q105.
Try placing the decoupling cap C104 directly on transistor pins instead of on the PCB.... If those transistors/resistors (q103/104 and R111/112) get really hot, it can be caused by that Q105 almost acts like an "short circuit"... What are the Voltage readings across R111 and R112??
R118/119 will normally not be very hot, however if some reason the OpAmp oscillating they will get hot...............
Try putting a 100-500 pF cap from pin 2 of opamp to GND.
Try placing the decoupling cap C104 directly on transistor pins instead of on the PCB.... If those transistors/resistors (q103/104 and R111/112) get really hot, it can be caused by that Q105 almost acts like an "short circuit"... What are the Voltage readings across R111 and R112??
R118/119 will normally not be very hot, however if some reason the OpAmp oscillating they will get hot...............
Try putting a 100-500 pF cap from pin 2 of opamp to GND.
It's working!
Jan,
You were right, the pins for the MPSA42/92 were indeed reversed.
I followed datasheet from ON Semi, don't know what went wrong though. The amp is now working, with 0.03mV dc offset and idle at 100mA. The MJE340/350 and 330Ohms/2W are very HOT. 
Maybe they got damaged or something during the time I thought I got the MPSA42/92 correctly placed? Just have to replace them tomorrow just to be sure.
Thank you so much, I'll post voltage readings you requested tomorrow.
JojoD
Jan,
You were right, the pins for the MPSA42/92 were indeed reversed.
I followed datasheet from ON Semi, don't know what went wrong though. The amp is now working, with 0.03mV dc offset and idle at 100mA. The MJE340/350 and 330Ohms/2W are very HOT. Maybe they got damaged or something during the time I thought I got the MPSA42/92 correctly placed? Just have to replace them tomorrow just to be sure.Thank you so much, I'll post voltage readings you requested tomorrow.
JojoD
ACD,
I printed the schematics from Your site yesterday. I have few questions about the amp:
1, I use MJE15030/31 in my amp. But I found that the B of this trannies falls rapidly under 100mA. Did You checked it? Because Your predriver stage use them with small current. (MJE15032/33 has same characteristic). Maybe MJE340/350 would be better?
2, Why do You use separated emitter resistors for the drivers (180 for pre, and 5.6 for drivers), instead of single resistor between the emitters?
3, Did You ever try to use output stage as in the ETI amplifiers? (CFP with voltage gain) It can be connected directly to the output of the opamp.
Sajti
I printed the schematics from Your site yesterday. I have few questions about the amp:
1, I use MJE15030/31 in my amp. But I found that the B of this trannies falls rapidly under 100mA. Did You checked it? Because Your predriver stage use them with small current. (MJE15032/33 has same characteristic). Maybe MJE340/350 would be better?
2, Why do You use separated emitter resistors for the drivers (180 for pre, and 5.6 for drivers), instead of single resistor between the emitters?
3, Did You ever try to use output stage as in the ETI amplifiers? (CFP with voltage gain) It can be connected directly to the output of the opamp.
Sajti
Hello sajti,
I appreciate your interest in my amplifier, however as stated on my website, this amplifier is not entended to be a "State of the art" product, but more a tribute to the classic power amplifiers...
Saving said that, I am aware of the problems with MJE15032/33.
Personally I have only used them in 1 of the 11 amps I have build sofare..... I prefer MJE340/350 however after receiving app. 30 emails a day (from people who didn't get that this amp was ment as a replica) regarding using these "old" transistors as drivers, I've changed the schematic to MJE15032/33
I've have used seperate predriver and driver resistors because I've for the last 25 years have had very good experience with this topology. There are no problems in using a single local feedback resistor (Emitter to Emitter), however I can't see the meaning of that in this type of circuit designed for qverall feedback
I'm aware of the circuit in the ETI amps, and it may be incorporated when I decide to make a real "High End" amp.
For now, I'm building several variants of current design to different peoples and companies.
But please keep the comments coming.............
New ideas are always welcome
I appreciate your interest in my amplifier, however as stated on my website, this amplifier is not entended to be a "State of the art" product, but more a tribute to the classic power amplifiers...
Saving said that, I am aware of the problems with MJE15032/33.
Personally I have only used them in 1 of the 11 amps I have build sofare..... I prefer MJE340/350 however after receiving app. 30 emails a day (from people who didn't get that this amp was ment as a replica) regarding using these "old" transistors as drivers, I've changed the schematic to MJE15032/33
I've have used seperate predriver and driver resistors because I've for the last 25 years have had very good experience with this topology. There are no problems in using a single local feedback resistor (Emitter to Emitter), however I can't see the meaning of that in this type of circuit designed for qverall feedback
I'm aware of the circuit in the ETI amps, and it may be incorporated when I decide to make a real "High End" amp.
For now, I'm building several variants of current design to different peoples and companies.
But please keep the comments coming.............
New ideas are always welcome
Emails
Jan,
I got your emails, thank you.
Before I send you any voltage readings, I would first like to replace ALL transistors in the pcb as well as the bias transistor so that I have a fresh start before measuring anything. I'll send them data to you asap.
This sure sounds very good for a "replica" of an old amp. Sounds like an old SAE my father and I used to listen.
BTW, what idle current are you using?
Regards,
JojoD
Jan,
I got your emails, thank you.
Before I send you any voltage readings, I would first like to replace ALL transistors in the pcb as well as the bias transistor so that I have a fresh start before measuring anything. I'll send them data to you asap.
This sure sounds very good for a "replica" of an old amp. Sounds like an old SAE my father and I used to listen.
BTW, what idle current are you using?
Regards,
JojoD
Hi Jojo,
Glad you got the emails
I'm also happy to hear, that you think I managed to "recreate" this amazing sound from the old amps
The idle we have found most suitable is app. 315mV across each of R122 and R123, making the Drivers running class A while output transistors are running class B.
I know that many will comment this, however it work just fine, and most amps from the old days used this technique....
You can raise the bias if you want to... The design allows you to run class A all the way, however you will need some pretty big heatsink and some fans
I'm looking forward to see your voltage readings.
Glad you got the emails
I'm also happy to hear, that you think I managed to "recreate" this amazing sound from the old amps
The idle we have found most suitable is app. 315mV across each of R122 and R123, making the Drivers running class A while output transistors are running class B.
I know that many will comment this, however it work just fine, and most amps from the old days used this technique....
You can raise the bias if you want to... The design allows you to run class A all the way, however you will need some pretty big heatsink and some fans
I'm looking forward to see your voltage readings.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.