Rising DC Offset

[pmchoong]

Hi All,

I have a problem with my Aleph J. One channel is working fine. But for the other channel, the DC offset seems to rise infinitely. When it reaches about 300mV, I turned it off. It started from 65mV and after about 2-3 seconds, it begins to climb and in a few seconds later, it reaches 200-300mV. The rail supply is 48V and before connecting the board, it was 56V.

I am using 2sj109 and bc550 for Q3 and Q4.

Thanks.

 

[biju55]

where did you buy those 2sj109 ?

 

[uli]

what voltage do you measure across R7? leave the amp on until it temperature settles.

 

[pmchoong]

where did you buy those 2sj109 ?

It was brought long ago. The other channel uses the same sj109 from the person I brought. I brought about 10 pieces from him. I have burned about 4 of them.

 

[uli]

the sj109 is likely not the fault. check all voltage Parameters in the ccs

 

[pmchoong]

the sj109 is likely not the fault. check all voltage Parameters in the ccs

the dc rises too fast. i don't know if i have enough time to measure it.

 

[Zen Mod]

reflow all solder joints

replace input CCS components

replace small BJT in Aleph CCS

 

[uli]

the dc rises too fast. i don't know if i have enough time to measure it.

running this without a load is the way to go. even offset near the rail voltage should not harm

 

[pmchoong]

reflow all solder joints

replace input CCS components

replace small BJT in Aleph CCS

I have replaced all Q2, Q3 and Q4. They are new. The only different is thatthay are BC550.

 

[uli]

start tracking down the fault BEFORE swapping parts which are likely ok ;-)

 

[pmchoong]

running this without a load is the way to go. even offset near the rail voltage should not harm

you mean remove the mosfets?

 

[Zen Mod]

he meant nothing attached to amp's output

 

[Zen Mod]

I have replaced all Q2, Q3 and Q4. They are new. The only different is thatthay are BC550.

took care of pinout difference ?

 

[uli]

you mean remove the mosfets?

no, they are integrated part of the circuit. Load would be a Speaker. start measuring the voltages across the resistors like R7, R8, Z1, Q3 E-C, Q4 E-C.

then compare to the working channel. maybe only the value of R7 Needs to be changed a little bit or the value of Z1 is slightly out of spec.

does the offset run to - or + supply?

having too less voltage across R7 means positive offset and the way round

 

[biju55]

Re-solder the 2sj 109 and its related components . Clean well the space between the legs of sj109 .

 

[pmchoong]

no, they are integrated part of the circuit. Load would be a Speaker. start measuring the voltages across the resistors like R7, R8, Z1, Q3 E-C, Q4 E-C.

then compare to the working channel. maybe only the value of R7 Needs to be changed a little bit or the value of Z1 is slightly out of spec.

does the offset run to - or + supply?

having too less voltage across R7 means positive offset and the way round

Voltage across R7 is 5.10V in the 2 seconds that I measured it.

 

[uli]

let the amplifier warm up! it makes no sense to Switch off after 2 seconds. again, does the Offset drift towards - or +?

 

[pmchoong]

let the amplifier warm up! it makes no sense to Switch off after 2 seconds. again, does the Offset drift towards - or +?

It drifts towards the +.

If i let the offset to climb, it will trigger the speaker protection board.

 

[uli]

put away the speaker protection when bugtracking

 

[pmchoong]

put away the speaker protection when bugtracking

wouldn't it burn the mosfets?