f5 power up issue

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I had previously built cviller boards, this one is using Daniels boards.

At first power up, R11 and R12 went way over voltage and started to smoke... meters showed 4 or mover volts each.

Looking at the board, I think I installed Q5 and Q6 backwards, would that cause this issue? I managed to flip them over but I have the same results so maybe they are a dead short inside?

Any thoughts? Here's a picture, you can see the orientation of q6 on the board, and the print on the device is facing us.

q6.jpg
 
I just double-checked my boards.... both seem to be right. I'm using the Toshiba outputs so I though the pin out could be backwards or something like that but they appear to be oriented correctly. Q1 and Q2 are in there right, as were Q5 and Q6.

I only powered up one board, I'll saddle up the other one tomorrow evening and see what happens.

Any other ideas would be appreciated!
 
any substitutions in your build on the transistor side maybe a different pin out on a sub part. your soldering looks great some times a cold solder joint can cause major pain....

Elwood

Q6 and Q5 are as called on the schematic, I'll double check Q1 and Q2. I verified Pin 1 of Q4 goes to R14 and that R14 is 47ohm.

My eyes are old(er) so I double check my soldering with a magnifier. In a former life as a tech I used to build and repair analog to digital devices so I'm pretty wary about poor solder connections.


How did you set the trimpots?

Hello sir, thanks for checking in.

I set the trimpots to minimum, or what I figured were minimum based on which pins are shorted on your boards. After the initial wisp of smoke, I turned them fully the other way and had the exact same result.

I then measured across R4 and R3 in circuit to validate I had them set to minimum again.

For my first build using cviller boards, I did make that error. So initially I thought that I must of had them set wrong.
 
Trough the picture I see that your trimpots are not correct installed! (The pcb has two small circle with a line like a bolt so the bolt direction is marked on the pcb!!!)

This is not a problem (little tricky to set bias afterwards) but maybe this confused you when you wanted to put them to zero so if you wanted to zero them the first should be fully clockwised and the second(this is half in the picture) should be fully anticlockwised, but again is better to measure the resistance behind the pcb with an ohmmeter, there you can see the two legs of the trimpots that are shorted and the free one this two positions should give you zero ohm. (for safe I measure between the two legs of R4 to get for sure zero ohm, and then the same for R3)
 
Last edited:
A photo! Also from the pic I see R11 is very small is this 3W? and be careful 0.47R (looks like a dale resistor and I don't beleive they have that value of resistors 0.47R) also the full output burnt the mosfets for sure! (I have made the same mistake in the begging, the only parts that I have burnt were the R11 & R12 and the two mosfets, all other component were ok!)


PS from your pictures Q5-Q6 are correct instaled!
 

Attachments

  • q6.jpg
    q6.jpg
    110.2 KB · Views: 233
Last edited:
The second mistake I made was to use a bad thermal conductive between the heatsink and the mosfets! So the heatsinks were about 40oC warm and the mosfets about 72oC and the sound was very unbalanced I had a big, huge, abysmal bass and all the other spectrum was lost, then I tried to increase bias so as to get the heatsink hotter and the sound changed a alot but the temperature was 43oC for heatsink and 82oc for the mosfet for (1.3 to 1.5A bias or 0.6V to O.7v), the increased bias sounded more balanced but not correct not involving not what I expected....


So I changed the thermal conductive with just a thin layer of kapton (there is a lot of choises, now I measure 44oC heatsink and 56oC to mosfets)), and the amplifier now was back fully balanced and good sound.... because again I had a little hum problems, so I made a star ground(the higher freq. became cleaner) and since then I enjoy this fabulus amplifier....!:)

One more tip, the mosfets should be fully electrical insulated from the heatsinks!
 
Last edited:
Trough the picture I see that your trimpots are not correct installed! (The pcb has two small circle with a line like a bolt so the bolt direction is marked on the pcb!!!)

Good catch but this is the board that I have not powered up yet. The board I'm asking about has the proper orientation of trim pots.


A photo! Also from the pic I see R11 is very small is this 3W?

Yes, it is a 3W panasonic like the other larger resistors in the photo.

But I'm confused, R11 and R12 are not in this photo.


PS from your pictures Q5-Q6 are correct instaled!

thank you for that confirmation
 
Also, I was not sure I wanted to have the trimpots set all the way in either direction for fear of frying something so I set the pots to the 600ohms on R3 and R4 from the F5 manual in the second to last circuit depicted. I figured this was safe as 600 was the "fixed" setting.
 
Also, I was not sure I wanted to have the trimpots set all the way in either direction for fear of frying something so I set the pots to the 600ohms on R3 and R4 from the F5 manual in the second to last circuit depicted. I figured this was safe as 600 was the "fixed" setting.

600ohm is safe! I measured the trimpots on one channel pcb full assembled and found for bias 1.4A the one is about 587ohm and the other about 710ohm, but for sure you gonna have a big dc offset! The best way is to start at zero and reach the 1.3A=0.60v by keeping the dc offset between zero mv!
 
Last edited:
I'm officially frustrated.

I powered up the 2nd board with exactly the same results.

I set R4 and R3 to 600 ohms prior to install (left the trimpot backwards though) so I'd be "ballpark".

I checked the orientation and placement of Q1 - Q6. Made sure that Q3/Q4 pin2 was not shorted to the heat sink.

Carefully connected my meter leads so as not to short anything, used some electrical tape to protect the thermistor leads too.

The voltage across R11 and R12 is way high on this one too. It reaches .600vdc when the AC in is only 21vac.

So whatever I've done, I did it the same way for both boards. I'm wondering if the Q1/Q2 buy that I did here on diyaudio are bad?

All ideas are welcome.

Maybe I can take some close up pictures of the build?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.