| phresh |
Hi, i've just finished building one of the 3886 gainclones from madaboutsound and am getting high DC offset values from one of channels.
One channel works fine, and sounds excellent (bar the lack of stereo of course!). In it's current state, with no case, it has a DC offset of around 60mV, which I expect will drop once it's all cased up properly.
The other channel does not work and gives a DC offset of somewhere around 400mV, far higher than it should be.
I've checked the outputs from the power supply and they are fine, supplying in the region of 32V. I suspect something has shorted and blown the chip, as I did hear something crack (the generic shorting sound). The chip also doesn't get warm under load (or no load).
Should I still be getting a DC offset at the output terminals if the chip had blown? Also, what are the usual culprits for causing high DC offset?
Grounding is the same for both so it can't be an external grounding issue, although I could have some how shorted to it on the board.
Any help appreciated, i'll try and get a picture up tomorrow which may help. |
|
|
| Nordic |
| Sometimes you just get a dud chip.... |
|
|
| AndrewT |
Hi,
even 60mV is high.
are you using DC blocking on both the input AND the NFB lower leg?
Are you using inverting topology? |
|
|
| phresh |
Hi, its non-inverting. Got a cap on the input, but not on the feedback resistor. I've re tested the working amp and that has now got a DC offset of 38mV (80's club classics cd, nice!). I'm suspecting that i've just over soldered one of the connections which has shorted something and blown the chip. Like I said in my original post though, should it be possible to get a DC offset at the outputs if the chip has blown?
I'll sit down with it one night this week and go over all the connections. Seeing as one works and I did nothing different between them it has to be something simple. |
|
|
| AndrewT |
Hi,
the DC coupling in the NFB lower leg results in unmatched impedances on the input pins of the chipamp. The DC gain is probably set to about 20times and this guarantees a high output offset.
How are you measuring the output offset?
You should be fitting a shorting plug to the chipamp input so that is sees a source impedance of 0ohms. There should be no load on the output terminals
Can't help with the 400mV. |
|
|
| BrianGT |
I would guess that it is possibly a bad chip. I would check all of the components on the board, comparing the values with the working board, and if you don't find any differences, I can send you a replacement chip.
--
Brian |
|
|
|
