Rockford T10001BD blown output fet

[Perry Babin]

Solder a bridge between pins 1 and 2 of U200. That should bring yours to 50%.

 

[ComputerJLT]

you did that or thats how it came?

should this bridge be permenant?

 

[Perry Babin]

I did that. It's needed on many class D amps when the feedback loop is broken due to missing outputs. This only works on 'clocked' class D designs. It won't do anything for self oscillating class D amps.

It's only used while the outputs are out of the circuit.

 

[ComputerJLT]

ok;

oh and as far as my meter; it reads infinity with the leads open in diode check mode; i think it just measures resistance but i'm not sure cus the value of resistance was different on the diodes than when the meter was in diode mode.

 

[ComputerJLT]

ok shorting those pins made the balance 50/50

now i have no drive signal on q208/q209 again lol wtf

 

[Perry Babin]

Are all of the diodes back in the circuit and with the correct orientation?

 

[ComputerJLT]

yea it was a solder joint on one of the resistors; I redid all of my solder joints and signal is back but its still only 5v.


think the value of one of the resistors I added in could be making the drive voltage low?

R231 234 and 235 have been replaced with correct value 1/4 watt 5% resistors at 2kohm (2x 1k in series), 2.7ohm, 100ohms respectively. All values match what the schematics have on them and they also match what their sister resistor values are. (all resistors were tested before i soldered them in too)

 

[Perry Babin]

Confirm that you have the same signal levels on the cathodes of D207 and D207. The first two are D207/208 with the differential probe setup and the ch2 probe on the collector of Q203. The second ones are with a single probe but with the collector of Q203 shorted to the negative speaker terminal.

 

[ComputerJLT]

my waveforms look the same at the cathode and almost no voltage is dropped to the anode....

 

[Perry Babin]

Go through the circuit to see where you're losing the other 4v of drive. It's about ±9v here.

 

[ComputerJLT]

will voltage be 1:1 from the base to the emitter on Q203/Q206?

 

[ComputerJLT]

q203 and q206 both show about 7.5v on the gate but only about 4.7v on the emmitter....

think thats the issue?

 

[Perry Babin]

These are the base and emitter waveforms for Q203 (first two) and Q206. Hold the cursor over the thumbnail to see the filenames. It appears that Q203 may be weak.

This was taken with the differential probe setup (collector of Q203 as the reference). The bottom of each waveform was aligned with the reference to make it easier to see the true amplitude.

 

[ComputerJLT]

i re-replaced q203 and q206 and all is well and right with the world!!

i guess one was damaged when the set of fets i put in there blew out

 

[ComputerJLT]

got all the new fets and such in the mail today going to try and put them in tonight or tomorrow and hopefully not blow them up.... if the first set dies i'll buy the ones that rf used in the later amps but these were cheap and i have an idea how to use the rest if the first set dies...

anyway the caps i ordered were the same on paper, 4.7uf 250v mylar/nylon wahtever caps like the one that came out but the ones i got are half the physical size of the original. will this be a problem? both weigh about the same; the smaller one is definately mroe dense and less blown up than the original lol

 

[ComputerJLT]

Well I replaced all the output fets and C242 powered it up and all seemed well but as soon as the amp unmuted something brought down the power supply (like a high load) and would cycle doing that about 2/sec.

Q208/Q209 got warm after a few seconds of doing that; much warmer than the rest of the fets

this waas with no load and no input and all 8 output fets installed

 

[Perry Babin]

It could be a number of things (FETs not suitable, shorted inductor, inductor shorted to the board...). If the drivers got hot, the FETs likely failed.

Did you tin the backs of the FETs? If so, did you check for leakage after you tinned them?

 

[ComputerJLT]

did not tin them and did not mount them to the insulator; is that part of the curciut?

i think its the fets not suited; just removed them all they all check out ok and all the waveforms are still good on the gate pins with no output fets...

i looked at the turn on and turn off times of this new fet and its TERRIBLE compared to the old ones; i'll have to post up exactly but i think it was just too slow...

i also had some nasty RFI when they were in; enough to make noise out of my speaker and make my crt monitors flicker slightly

 

[Perry Babin]

If you haven't replaced the TO-92 case transistors with the proper drivers, they may be getting hot because they can't sink the heat to the board like the originals.

 

[ComputerJLT]

those dont get hot at all, the amp was on maybe 5sec while popping, i removed 208/209 to look at the wave forms and the rfi got worse and the wave forms looked all kinds of messed up...

all output fets are out now; i'm going to order the correct ones monday and try agian.