|
|
Post by leeboy22001 on Jul 30, 2017 17:51:30 GMT -5
Having removed output fets, and half bridge module the voltage is across zener is now 17.6v. Now need to find out why the module is clamping the voltage !
|
|
|
|
Post by leeboy22001 on Aug 1, 2017 14:12:32 GMT -5
I've seen this before.I've purchased this sub off ebay and I think, the light on the back would turn from red to green but could not hear speaker relay.Was able to fix it by replacing zener diode and IR2153 IC. Hi, Thanks, very odd, I've removed the ir2153, and rechecked voltages before soldering new one, I had 17.6v across zd3, and 14.6v at what would be the input to the ir2153, so that all looks ok. I have no output fets soldered in, I soldered in ir2153, and the voltage across zd3 is 10v again !??.. could it be the 15v sm zener is failing somehow? Or the 2n3904 used as voltage followwr is failing under load ? |
|
|
|
Post by dobrykamil on Aug 1, 2017 14:36:17 GMT -5
|
|
|
|
Post by leeboy22001 on Aug 1, 2017 15:42:42 GMT -5
I changed c13 with c26, and I now get around 16v on zd3 !, when I removed c13 it did seem to be a little dry around one pin, so will resolder, if not I'll replace it  |
|
|
|
Post by leeboy22001 on Aug 1, 2017 15:47:03 GMT -5
I've now resoldered c13 and the voltage is around 16v across zd3 !.. Looking promising |
|
|
|
Post by leeboy22001 on Aug 1, 2017 15:55:14 GMT -5
The plot thickens, I resolderrd all the pins of the half bridge module, and now back to 10v, so the extra load of the output pins connected is causing supply to drop. Or the c13 really is duff, will put c26 back into c13 and retry!
|
|
|
|
Post by leeboy22001 on Aug 2, 2017 5:26:00 GMT -5
I'm at a loss, without the output fets fitted, reducing r7 (300k) to 120k gives me around 17v across zd3, and 13.8v at ir23531 input. Low side output is drive constantly, as expected. Nothing on high side, but assume this is sure to no volts on the vb pin as fets aren't fitted as this supplies the floating high side voltage source. When fets are soldered In, volts across zd3 drops to around 9-10v again !.. the low side then drives (10v pkpk)for around 30ms, stops for 5ms, then starts again etc.. the high side output then mirrors the low side, except it's around 250v pkpk?
|
|
|
|
Post by leeboy22001 on Aug 2, 2017 7:09:00 GMT -5
My mistake, the gate voltage seen by the high fet is between vs and the gate input, not gnd and the gate. The gate voltage is as expected, except the on/off nature of the pwm gate signal due to under voltage of vcc. I have replaced the 15v zener, and still only 10v across zd3 !, I may change c23, the 22uf cap
|
|
|
|
Post by leeboy22001 on Aug 2, 2017 16:21:18 GMT -5
I'm at a total loss, have replaced both c13 and c26, the 15v zener, and the 1n4148 diode d3. And I'm no closer finding the fault. Can you send me voltage measurements without cn2 and cn3 connected? Thanks
|
|
|
|
Post by mastertech on Aug 2, 2017 18:21:00 GMT -5
Remove Th1 and put in a 10 ohm resistor and test your voltages again.
|
|
|
|
Post by leeboy22001 on Aug 3, 2017 4:21:03 GMT -5
Remove Th1 and put in a 10 ohm resistor and test your voltages again. Thanks, tried that but no luck. I believe the driver uses a regeneration design, the current via the 300k is only enough to kickstart the ir21531s, once the transformer is energised, the ac voltage on T1 pins 5 and 6, is half rectified and fed back to the ir21531s supply, this will will increase the current and voltage available, and the ir21531 in theory will then go into normal mode. This isn't happening. I removed R9, and added a 100uf cap on the output of d3, to average the supply volts, this was 10.6v, so enough to get it going in normal mode, once connected back into the closed loop it drops back to 9.8v or so. I've replaced R9, and this didn't help. I've also reduced R7 to 100k, to give it more current, but this enabled the pwm from the ir21531 to be active for longer, but still not full time. |
|
|
|
Post by leeboy22001 on Aug 3, 2017 11:55:03 GMT -5
Looks like this design by Klipsch (Promedia 2.1 new version), uses an almost identical smps design, even uses what looks like the same smps module!
|
|
|
|
Post by leeboy22001 on Aug 3, 2017 14:25:55 GMT -5
Perseverance has paid off, replaced both C18 and C27 0.1uf 250v PET caps, and it's now working !.. cost 38p, if only I tried them first ! Lol
|
|
|
|
Post by dobrykamil on Aug 3, 2017 14:37:18 GMT -5
Happy days and good for you mate.What was the actual value of those caps?
|
|
|
|
Post by leeboy22001 on Aug 3, 2017 15:12:09 GMT -5
The capacitor meter can't measure any capacitance on either of them, yet I know the dvm works fine . This would suggest both had gone open circuit! Sub is all back together, and fully working |
|
