MN3005 Circuit Calibration

[gbwilliams]

Hey all! I've purchased quite a few MN3005 chips over the years (both original Panasonic and XVive clones) for eventual use in a replica of the Rockman Stereo Echo. I've tested them all in an existing (socketed) Stereo Echo, and they all work, but for the most part, they don't provide nearly as long of a "tail" as the stock MN3005's from the unit. I suspect it's because the Stereo Echo has no trimmer pots inside to calibrate for different MN3005's, so the originals were probably hand-selected to work with the Stereo Echo's fixed component values.

To that end, I'd like to build a basic tester for the MN3005's. I'm considering using the schematic from the original MN3005 datasheet, but it only has a trimmer pot for minimizing distortion, which I believe is the input signal bias adjustment. Would I need to make any other aspect of this design adjustable to be able to properly test a random MN3005 chip? Perhaps there are other circuits that you'd recommend that already have such flexibility? I looked at various Memory Man schematics, but I want something simpler that only uses a single MN3005, so as to minimize external variables.

Any help is appreciated!

 

[JohnRoberts]

Hey all! I've purchased quite a few MN3005 chips over the years (both original Panasonic and XVive clones) for eventual use in a replica of the Rockman Stereo Echo. I've tested them all in an existing (socketed) Stereo Echo, and they all work, but for the most part, they don't provide nearly as long of a "tail" as the stock MN3005's from the unit. I suspect it's because the Stereo Echo has no trimmer pots inside to calibrate for different MN3005's, so the originals were probably hand-selected to work with the Stereo Echo's fixed component values.

Those are relatively expensive components so unlikely to be selected and discarded if low output.

To that end, I'd like to build a basic tester for the MN3005's. I'm considering using the schematic from the original MN3005 datasheet, but it only has a trimmer pot for minimizing distortion, which I believe is the input signal bias adjustment.

indeed that DC bias trim is adjusted for symmetrical clipping of an input sine wave while looking at the output with a scope.

Would I need to make any other aspect of this design adjustable to be able to properly test a random MN3005 chip? Perhaps there are other circuits that you'd recommend that already have such flexibility? I looked at various Memory Man schematics, but I want something simpler that only uses a single MN3005, so as to minimize external variables.

Any help is appreciated!
View attachment 131148

I would begin with a batch of BBD ASRs, tweak the DC bias for symmetrical clipping, then measure output for a nominal level perhaps 3dB below clipping. Measure enough to get a sense of the typical input vs output levels, and how much they vary.

If you are testing a lot of BBDs maybe replace the 100k trim pot with a real potentiometer for better life.

Depending upon how much the throughput gain varies, then you can evaluate whether to change values or add a recirculation level trimpot. Too much recirculation gain can run away which is undesirable.

JR

 

[gbwilliams]

Thanks for the advice, I'll give that a try!

 

[JohnRoberts]

The new non Matsushita (Panasonic) BBDs are run across a different fab process so may exhibit different transfer efficiency. Measure twice , cut once.

JR