DAN ARMSTRONG: Blue Clipper

I thought I might finish off layouts for the rest of the Dan Armstrong series, as I've already done a few.  Anyways, while I was looking at schematics of the Blue Clipper from all the usual sources online, I noticed that the opamp bias looked a bit odd.  Then I noticed a few layouts used different values, so I checked out some photos to see if I could make sense of it.

I found a clean photo on a Japanese listing for a Blue Clipper, and these are the values that I could see (trace image below).  There was another shot that I found online from a different source that has the same values, so the one below is not an aberration of some kind.

Significant differences are the input bias and the resistor controlling gain.  The schematics online that I've seen have a 20k / 240k pair of bias resistors (8.3v bias), whereas the one I'm looking at is a relatively normal 200k / 240k pair (4.9v bias).  Did someone miss a zero once and the mistake was copied by others?

The other difference is the online schematics usually have a 2.4k resistor for gain, whereas this one has 150 ohms.  That's a pretty big difference on a circuit like this (about 1000 times gain vs 63 times).

So what does it all mean?   Well, I do know that I'm certainly not the first person to work this out, as I've since found a few mentions elsewhere of correct values (from decades ago) - and there are of course layouts out there with correct values right now along with the incorrect ones.  

The opamp outputs are completely different due to the odd bias and to a lesser extent the lower gain.  Blue = traced values.   Green = online schematic values

Output level differences aside, the waves are at least similar by the time it's heavily filtered by the 33n cap across the output.

Drop that 33n on the output down to 3.3n and you can start to see some real difference - probably going to sound a bit fuzzy.  Blue = online schematic.  Green = trace  (both using 3.3n on the output).  33n is often reported as being too dark for most people.  

3.3n might be too big a drop, so maybe start with something around 10n and see where you land.  

Using the 33n cap on both, the frequency response is about the same.  The difference is output level is very clear.  The peak is at around 250hz / 300hz, which is pretty muddy by most standards.  

Here are a few values from the stock 33n to 3.3n.  Left to right, 33n, 22, 15n 6.8n 3.3n

DAN ARMSTRONG BLUE CLIPPER - VERO LAYOUT

DAN ARMSTRONG BLUE CLIPPER - VIDEO DEMO

Now I know people hate on JHS for various reasons - but this is a good demo...  and face it, the guy is living the dream, right?