Yellow humper correct schematic

[Bassbob]

I wanted to build a clone of the Dan Armstrong's Yellow Humper but the only schematic that I found on the net was wrong.
With a bit of reading, thanks to runoffgroove and EddieB's comment on tagboardeffect, some breadboarding and my ears here's what I came up with.
Since I learned a lot from this forum I thought I should share my work

[PRR]

Thanks, but.... I don't see the DC bias.

Compare to Purple Peaker: http://home-wrecker.com/peaker.html

Looks a lot alike except the AC/audio filter values, and DC continuity from Q2 E to Q1 B.

With a string of four 82k E to B, and 68k B to ground, we expect E2 to bias near 4V or 5V, a good working point.

[pacealot]

Good news! At "that other" forum, the 1975 DA handwritten schematics for all their main plug-in gadgets were recently reposted, so now the (at least theoretical) original values of the PP and the YH can now be re-revealed:





Edit: I realise that part of the appeal of this circuit is the ease of tweaking cap (and occasionally resistor) values to target whatever frequency bands one desires, but it is unfortunate that the modded values seem to have gone halfway around the world while the original values have still been tying their bootlaces, as it were. Also it's worth noting that the original schemos for both employ different resistor values from each filter to ground for differing levels of boost at each frequency — something that hasn't appeared on any of the reinterpreted schematics since to my knowledge. At any rate, more (presumably correct?) information has got to be a good thing, yes? If anyone has an original of one or the other (or even both) to confirm or deny these values, chance'd be a fine thing...

[Rob Strand]

Interesting they swapped the order of the Low and High EQ sections on the YH and the PP.  Presumably so the same PCB can be used to get treble boost on the PP and bass boost on the YH in the lone EQ setting.

Probably good idea to do a NPN version.  Seems like a good idea to keep the gains >= 200.
[FYI: The generic 2SA666 has a typical gain of 250.   However these devices come in
gain sorted -P, -Q, -R, -S, -T versions.  For those -R is hFE 180 to 360, and -S is hFE 260 to 520.
I don't know what the original used.  Need PCB pics.
I found a Red Ranger pic (Musitronics era) which had a 2SA666-R.]

Here's some more info:



Oh, it seems flat isn't flat.  I checked the sim and it seems to be the case.   It's the 470k resistors.  They are not large compared to the 2x82k =164k.   If you want flat in the flat position use a bypass switch or change the 470k's to something much larger.

[pacealot]

I found a photo of a WD-era Yellow Humper which I strongly suspect (based on previously trying to work out the PCB layout of a Purple Peaker) has two 1M resistors in the locations where the 470Ks would go according to the schematic (the two that are nearest to the viewer):



Otherwise, visible resistor values seem to be as per the schematic (15K, 33K, a couple of 39Ks, a 150K on the right side where I suspect the input to be). Not a whole lot to go on, but a bit more grist for the mill.

(Oh and that Purple Peaker PCB appeared to have BC183LCs rather than 2SA666s, for what it's worth...)

[Rob Strand]

I found a photo of a WD-era Yellow Humper which I strongly suspect (based on previously trying to work out the PCB layout of a Purple Peaker) has two 1M resistors in the locations where the 470Ks would go according to the schematic (the two that are nearest to the viewer):



Otherwise, visible resistor values seem to be as per the schematic (15K, 33K, a couple of 39Ks, a 150K on the right side where I suspect the input to be). Not a whole lot to go on, but a bit more grist for the mill.

(Oh and that Purple Peaker PCB appeared to have BC183LCs rather than 2SA666s, for what it's worth...)

Good digging.

So maybe the WD version is different and they put the 1M's in there to reduce the flatness issue.
Apparently there are differences between the Musitronics, Re-Issues and WD's;  Re-Issue maybe closer to WD's.
The electrolytic capacitor dates are 1994 which would be OK for WD's.   

I saw a unit with BC183's and small 1/8W resistors.  At first I thought it was a DIY clone but maybe it was a WD era unit.

The typical gains on the BC183LC are quite high, so perhaps another tweak.

[aron]

The purple peaker was my secret weapon with a Pignose 150R. I put it in the effects loop and it sounded great!

[Bassbob]

compare to Purple Peaker: http://home-wrecker.com/peaker.html

That's the schematic I put on breadboard with values taken from https://tagboardeffects.blogspot.com/search?q=yellow+humper. Only to find out that the boost in the high and low frequency were way to loud. With some reading I came up with the idea that the two 51K (39K in Yellow Humper tagboardeffect's layout) resistors were missed place so I moved them to be in series with the cap to ground in each frequency section. Just like R11 and R14 in the "handwritten schematic" posted by pacealot https://postimg.cc/3WSTVpLL which are missing in the home-wrecker schematic.

different resistor values from each filter to ground for differing levels of boost at each frequency

I will have to experiment

[Rob Strand]

The circuit has a few dBs of baseline gain.  If you are switching between fully bypass and the EQ it will sound like the *EQ* has a lot of boost but it's really EQ-band-boost + gain boost.   If you use the original switching there is baseline gain in the "flat" switch position so you can't detect the baseline gain boost.

(A simple way to decrease the base-line gain is to increase the input resistor and proportionally decrease the input cap.)

[Bassbob]

I had made a mistake so here is revison 2 of the previously posted schematic.

[antonis]

Still no DC bias from Q2 Emitter to Q1 Base..

[Rob Strand]

Still no DC bias from Q2 Emitter to Q1 Base..

[ElectricDruid]

I was curious about what these actually look like, so I did a sim of the two hand-drawn original schematics from further up the thread. Here's the results, Humper first, then Peaker:




Nothing you couldn't do with a two-band parametric EQ, but this circuit is certainly a lot easier to build!

What I really take away from this is that flat frequency responses aren't actually that *interesting*, and that most of the stuff we enjoy has a more complex response. So don't design a flat "hi-fi" style response into your pedals. Throw in a few quirks!

[Bassbob]

Still no DC bias from Q2 Emitter to Q1 Base..

I admit I don't know what DC bias exactly means but I have looked at all schematics posted on this tread and cannot see a difference with the one I posted that could be such a thing.

Please show me.

Here is the final schematic of what I build. Thanks to Rob Strand to point out a misplaced link.

My intention were not to make an exact clone of the Yellow Humper, I heard a demo, liked the idea of colored boost and wanted to build one. To me, if it sounds good it is good.

[ElectricDruid]

I admit I don't know what DC bias exactly means but I have looked at all schematics posted on this tread and cannot see a difference with the one I posted that could be such a thing.

You can see the difference between this image you posted, which has the problem:


...and this one, the latest version, which doesn't:


What Antonis is saying is that there needs to be a path for DC from Q2 E to Q1 B. And a DC path implies that it can't go through any capacitors on route, which is why the first one is wrong. There's a path from Q2 to Q1, but only via the capacitors. On the corrected version, the path is all via resistors with no caps blocking the way.

PS: I completely agree about the idea of a "coloured boost", and if it sounds good, it is!!

[antonis]

I admit I don't know what DC bias exactly means but I have looked at all schematics posted on this tread and cannot see a difference with the one I posted that could be such a thing.

There must be a DC (resistive) continuity (no series caps) between Q2 Emitter and Q1 Base..
(for reasons beyond present scope..)

That continuity is obtained via 4 X 82k resistors..

Below is your previous schematic (revision 2) from DC point of view..
(caps considered open-circuits hence removed out of circuit..)



You can clearly see that 82k resistors string (red line) is interrupted..

edit: Portugese are by far faster than Greeks..

[ElectricDruid]

edit: Portugese are by far faster than Greeks..

LOL! Not often! What is the Greek for "amanhã", anyway?!?

[antonis]

What is the Greek for "amanhã", anyway?!?

"Αύριο"..
(ávrio)

[Bassbob]

I admit I don't know what DC bias exactly means but I have looked at all schematics posted on this tread and cannot see a difference with the one I posted that could be such a thing.

You can see the difference between this image you posted, which has the problem:


...and this one, the latest version, which doesn't:


What Antonis is saying is that there needs to be a path for DC from Q2 E to Q1 B. And a DC path implies that it can't go through any capacitors on route, which is why the first one is wrong. There's a path from Q2 to Q1, but only via the capacitors. On the corrected version, the path is all via resistors with no caps blocking the way.

PS: I completely agree about the idea of a "coloured boost", and if it sounds good, it is!!

Now I get it.

Thank you!

[Rob Strand]

Now I get it.

Thank you!

You are still missing two resistors.

Look at the markup I did where I added two resistors.  Compare it against the hand-drawn drawings and you will see those resistors are present.  Then look at your schematic.