Hello there,
I have been a happy user of F5 using v2 board from diyaudio store for a couple years.
One channel recently developed a strange symptom: it produced sound on and off. Bias measured from R10/R11 looked good. Power supply voltage was 23.6 V for both positive and negative. Wiring looked good. I couldn't figure out what was wrong. My experience level must be at least "intermediate" if not "expert".
When it sounds, it sounds good. It just came on and off occasionally.
What could be the reason for this?
Q2
Since I liked F5 so much, I decided to build one more while figuring out the first problem. I used V3 board this time.
Since I was confident with previous one, I didn't use variac but I just connected the power supply directly to the AC outlet.
R10 and R11 of one channel burnt right away with flash of light. P1 P2 were zero for sure. This channel was exactly the same as the other channel, and the other channel is fine.
I cannot figure out what could be the reason for this either.
Help please!
Thanks,
Doug
I have been a happy user of F5 using v2 board from diyaudio store for a couple years.
One channel recently developed a strange symptom: it produced sound on and off. Bias measured from R10/R11 looked good. Power supply voltage was 23.6 V for both positive and negative. Wiring looked good. I couldn't figure out what was wrong. My experience level must be at least "intermediate" if not "expert".
When it sounds, it sounds good. It just came on and off occasionally.
What could be the reason for this?
Q2
Since I liked F5 so much, I decided to build one more while figuring out the first problem. I used V3 board this time.
Since I was confident with previous one, I didn't use variac but I just connected the power supply directly to the AC outlet.
R10 and R11 of one channel burnt right away with flash of light. P1 P2 were zero for sure. This channel was exactly the same as the other channel, and the other channel is fine.
I cannot figure out what could be the reason for this either.
Help please!
Thanks,
Doug
Sounds like a bad connection or bad solder joint somewhere.
If the amp has burned then a mistake has been made somewhere.
Or the bias was wrong on power up.
If the amp has burned then a mistake has been made somewhere.
Or the bias was wrong on power up.
Bad connection or damage somewhere is also what I have been looking for, but I couldn't find anything wrong yet.
Jfets I use are K170 GR 5B and J74 GR 5C. What are those 5B and 5C mean?
Jfets I use are K170 GR 5B and J74 GR 5C. What are those 5B and 5C mean?
I found the problem!
I soldered everything except the thermistors and output hexfets before mounting the PCB, and then soldered them later to have them at the right height. That was the problem. Solder smeared to the back of the pcb surface while I was doing it from the top, and shorted pins of hexfets. I removed the pcb from the heatsink and found the problem.
Now I put it back with new R10 and R11 (R7 and R8 in v3), mounted the pcb again, and then powered up. No smoke nothing. Yay!
But there's another thing strange going on. I had to increase P1 (post in the positive side, or upper side) way more than P2 to lower the output offset voltage at the speaker post. Even after I rotated P1 entirely, DC offset is 300 mV. I think gain difference between N and P Jfets could be the reason. How can I confirm what is causing this problem? I don't think 300 mV at the speaker post is acceptable.
I soldered everything except the thermistors and output hexfets before mounting the PCB, and then soldered them later to have them at the right height. That was the problem. Solder smeared to the back of the pcb surface while I was doing it from the top, and shorted pins of hexfets. I removed the pcb from the heatsink and found the problem.
Now I put it back with new R10 and R11 (R7 and R8 in v3), mounted the pcb again, and then powered up. No smoke nothing. Yay!
But there's another thing strange going on. I had to increase P1 (post in the positive side, or upper side) way more than P2 to lower the output offset voltage at the speaker post. Even after I rotated P1 entirely, DC offset is 300 mV. I think gain difference between N and P Jfets could be the reason. How can I confirm what is causing this problem? I don't think 300 mV at the speaker post is acceptable.
I don't remember what Idss and Vgs are. I used parts that were lying in my stash.
Is using K170 GR and J74 GR be fine rather using BLs?
Is using K170 GR and J74 GR be fine rather using BLs?
If you've maxed P1 then you will need to replace R3(v2 schematic) with a larger value; try 3KΩ, remember to reset P1/P2 to zero.
Replaced R3, R4 (R5, R6 of V3 schematic) with 3.3 kohm. No luck. Even tried 4.7 kohm, no luck. Very similar to 2.2k.
If I maintain ~ 0V at output, reading from R3 and R4 were 0.3V max.
If I get 0.6 V at R3 and R4, output offset was 300 mV.
Rail voltage was +/- 23.5 V.
Do I have to try different Jfet, or hexfet (mosfet)?
If I maintain ~ 0V at output, reading from R3 and R4 were 0.3V max.
If I get 0.6 V at R3 and R4, output offset was 300 mV.
Rail voltage was +/- 23.5 V.
Do I have to try different Jfet, or hexfet (mosfet)?
I think you've damaged a jfet, or something's wrong in the input area.
300mV of change in output offset for close to 600mA of bias sounds wrong.I would pull the SK170 and test whether it is alive.
300mV of change in output offset for close to 600mA of bias sounds wrong.I would pull the SK170 and test whether it is alive.
It is highly unlikely voltage across R3 and R4 (R5/R6) will be the same as Vgs is often different for N and P channel mosfets.
What is important is measuring bias across source resistors and maintaining <50mV dc offset
Last edited:
If it's less than 5mA, I would replace them with higher Idss jfets (Idss close to 10mA)
OK, I got home early because the weather is so nice (63 F) in Maryland, I gotta go out riding my bike when I can.
But I quickly measured source current: 28 mV across 10 ohm, gives 2.8 mA.
Also I found this:
2SK170 GR, BL, V
So it seems I have to replace current 2SK170 and 2SJ74 GR with BL of my old F5 which started not making sound for unknown reason. If those Jfets were the problem of my old F5, then the new one will still have problem.
I will report back after exploring twisty backroads.
But I quickly measured source current: 28 mV across 10 ohm, gives 2.8 mA.
Also I found this:
2SK170 GR, BL, V
So it seems I have to replace current 2SK170 and 2SJ74 GR with BL of my old F5 which started not making sound for unknown reason. If those Jfets were the problem of my old F5, then the new one will still have problem.
I will report back after exploring twisty backroads.
I just replaced 2SK170 GR and 2SJ74 GR with corresponding BLs. Now I can adjust bias properly while maintaining ~0V at speaker terminals. I haven't played music yet, but will update that later, too.
Thanks for all the help.
There should be somewhere mentioning the Idss ratings of jfets.
No GRs, Yes BL/V.
Thanks for all the help.
There should be somewhere mentioning the Idss ratings of jfets.
No GRs, Yes BL/V.
Hi Doug,
Mr. Pass does, if memory serves, point out that ~8mA is the target Idss in the F6 article but that left me with the same question. Will GR versions work? I'll take your experience to mind and use the BL's. That said, I wonder if GR's would work with a lower value source resistor...
Sinc,
d
Mr. Pass does, if memory serves, point out that ~8mA is the target Idss in the F6 article but that left me with the same question. Will GR versions work? I'll take your experience to mind and use the BL's. That said, I wonder if GR's would work with a lower value source resistor...
Sinc,
d
Doug,
To add: He also notes that above 8mA jfets are fine but may require small heatsinks. I've ordered said heatsinks as all of my BL's are high 8's to about 10 mA
Sinc,
d
To add: He also notes that above 8mA jfets are fine but may require small heatsinks. I've ordered said heatsinks as all of my BL's are high 8's to about 10 mA
Sinc,
d
- Status
- Not open for further replies.