Diamond Bass Comp

[phi1]

Built this as a commissioned request for a diamond bass comp. I found this pcb from lectric-fx made to fit in 1590B. I referenced the fsb diamond bass comp thread to find the value changes to make it like the bass version. The toggle switch on the side the is tilt EQ frequency (900/250Hz), dpdt with the caps soldered you it.

the fourth knob is a clean blend. The little daughterboard down by the ftsw is an Op-amp based inverting buffet circuit for the clean side of the blend pot. I know there’s some question as to how necessary a blend knob is on an optical comp... just turn the comp knob down and it sounds like the dry signal. But, this was a customer request, and I think gives some opportunity to intuitively dial in the attack and sustain since adding clean will help Preserve attack.

Anyway it turned into a pretty involved build, but I’m super happy with how it turned out and sounds great. Considered getting another pcb to build the simple guitar version for myself, but I also just built the pale green comp and these are in the same ballpark.

 

[debrad]

Hi @phi1 ,

I came across this post of yours while looking for details on a Diamond Compressor with a blend control and just wondered what your thoughts might be on choosing between the "Zirconia Jr." board you used here (https://lectric-fx.com/product/zirc-jr-optical-compressor/) and lectric-fx's bigger brother "Zirconia" PCB (https://lectric-fx.com/product/zirconia-optical-compressor-v-1-1/)?

In addition, I am curious to know if you were able to simply add on the blend board or if the Zirconia Jr. creates any of the "out of phase" issues others mention having to take account for?

 

[phi1]

I chose the Jr pcb for size (1590B!). I don’t recall what all the differences were between them.

The circuit does definitely invert phase, so a typical blend won’t work as the center of the sweep will partially cancel out. The little vero board in the corner is an inverting op amp buffer. Looking at it now I’m kinda surprised I was able to make it fit in that small space! I recently did a similar vero and I feel like it was larger… I could try to confirm the exact schem I used but I’m pretty sure it was based off this. I could also try to help get a vero layout drawn. 1590BB enclosure would definitely be more comfortable!
http://www.muzique.com/lab/buffers.htm
(last one).

I’ve since built a couple diamond type comps on a pcb I designed for myself, but with an op amp simplification instead of all the transistors, and without the blend or EQ. (Just comp and vol knobs). It’s quite nice and easy to build, I think I have some extra pcbs.

 

[debrad]

Thanks @phi1

I just ran across a nice little buffer/inverter circuit on VERO that I think should work nicely (https://www.parasitstudio.se/stripboard-layouts/simple-buffer-and-phase-inverter). It actually has the option to output the inverted and non-inverted signal so I think I'll put those on a switch and combine it with the Zirconia Jr. board with a blend knob.

 

[phi1]

Ah that layout looks great. In my opinion there is no benefit to have the inverted/non-inverted outs on a switch, as the non-inverted will for sure have phase cancellation with this comp circuit.

 

[Feral Feline]

I think the advantage to having it on a switch is if you have something else in your chain (upstream or downstream of the comp) that inverts phase, and it doesn't have a switch to flip the phase.

Some people run a clean-blend for their whole board, for example; live, switching in/out various effects depending on the song — it'd be handy to have something that can toggle the phase, perhaps even standlone.

 

[debrad]

Apologies for bothering everyone again...and for taking this thread off on another tangent but I am in the final stages of completing my own Zirc Jr. compressor with a spliced in blend and something just occurred to me:

Like @phi1 my plan is to run the input into both the compressor and the buffer/inverter at the same time then join their two outputs with a B100k potentiometer.

Since the compressor circuit has its own output volume/level, should I be mixing the raw and compressor signals BEFORE that level control? If not, will the raw signal be at the same level as the compressed signal? If not, will that prevent the blend control from properly mixing the two signals?

- brad -

 

[phi1]

I put the blend after the Comp volume control, that way the signals can be more or less unity level before blending. (Well compressor is affecting dynamics so hard to say what’s truly ‘unity’.

If you put volume control after, then the comp signal will be loud coming into the blend, so you will get less dry signal with the blend knob.

 

[debrad]

Thanks @phi1. Right after posting it occurred to me that the level control is probably more like the "make up gain" of the compressor...

 

[phi1]

It’s more like attenuation. The comp circuit amplifies the signal a fair amount and then the volume knob is just a typical voltage divider to bring it back down around unity (or however much boost you want)

 

[debrad]

Got everything wired up and it tests OK in terms of FUNCTIONALITY but what caught my ear was the fact that PEDAL's output with the Blend knob set full CCW (ie. all clean signal and no compressed signal) is lower than the bypassed signal.

My assumption is that when the Blend knob is set to "full clean", the output is just the dry signal passing through the buffer...so, shouldn't it be equal in volume to the bypass signal?

I also think it sounds a LITTLE distorted so, after double checking the buffer/inverter circuit on VERO (https://www.parasitstudio.se/stripboard-layouts/simple-buffer-and-phase-inverter), I'm going to try a few other Op-Amps to see if anything is cleaner than the TL072 I used in my first test.

 

[phi1]

Did you happen to add an output capacitor to the buffer circuit (design notes below the vero layout).

 

[debrad]

Hmmm, I must have missed that. DRAT! What does that output cap do if I add it as suggested?

The other thing that just dawned on me is that the buffer circuit will be powered with the same 18v as the compressor circuit since the recipient asked me to omit the Zirc Jr's charge pump so he can use his 18v tap instead. I thought making the caps rating >18v would be good enough but now I'm wondering if the higher voltage will impact the buffer's function too...

 

[phi1]

Running the buffer on 18 should be fine (cap ratings >18 like you said).

The output capacitor is a coupling capacitor that lets AC signal pass but blocks DC offset if the active parts of the circuit (op amps, transistor stages). They’re always found at the beginning and end of pedal circuits.

I’m not positive the coupling cap will fix your issue, but it should definitely be there anyway. (Between buffer output and blend pot). 100nf or more. Film or tantalum or MLCC. I prefer not to use electrolytic for coupling cap at the input or output since they can sometimes leak DC, which gives a pop sound when stomping.

 

[debrad]

Glad to hear I should be OK with the 18v input but can't believe that I missed reading that comment about the caps...thanks for the quick catch! Hopefully that simple addition will make the levels more consistent...

For my own clarification though, in the layout I'm using, does the 100nf cap on the left side of the vero serve as the input coupling cap? And, with the schematic you originally suggested (or ANY of the buffers shown on the http://www.muzique.com/lab/buffers.htm page), does the 10uf in series with the output serve as the output coupling cap?

 

[Feral Feline]

Yes, in the AMZ link, the 10µ serves as the output coupler, but as phi1 noted electrolytics can sometimes leak DC.

I've read that Tantalum caps sound better than electrolytics and read quite the opposite, and that type doesn't matter much at all, but since there aren't any box-film caps in a 10µ size, I like to swap out large electrolytics for tantalum caps. I've not compared elecs vs tants, I'm just going on what I read; and I've no idea but I suspect tants offer better leakage control than elecs (now that I think of it, I may have read Chuck D Bones stating somewhere on the forum that's the case — tants being less prone to leakage.

Just make sure you get the orientation correct, I've heard applying polarity the wrong way to a tant really stinks, literally stinks.

 

[phi1]

Yeah the C1 on the parasit schem is input coupling cap. For the Amz ones, 10uF is pretty overkill in my opinion, 100nf is likely adequate, and 1uF is plenty. The larger cap lets more bass through, but it’s dependent on the input impedance of the following pedal/amp. (RC filter frequency calculation). The output cap of the comp circuit is 100nF.

 

[debrad]

Thanks @phi1 and @Feral Feline. I have tantalum caps in 100nF, 220nF, 1uF, 2.2uF, and 10uF so I'll choose one of those (likely 1uF based on the comments above). Would you agree with the last comment under the parasit schematic that I should also include a 100k resistor to ground following the output coupling cap?

Finally, going back to my 9v versus 18v fears: is it safe for me to run my on/off LED indicator on the same 18v by simply bumping up my current limiting resistor (I've tried a few searches but the only thing I've come across are others GUESSING that their LED should be fine at 18v if they roughly double the CLR...)? If not, my feeble skills in reading schematics leads me to believe there MAY be 9v on the row of the buffer vero connected to the positive side of the 22uF capacitor (that pair of 10k resistors should give me a Vr of half the input voltage, correct?).

 

[phi1]

It wouldn’t hurt to add the 100k to gnd, but I don’t think it’s necessary. The concept is that even good capacitors may slowly leak a tiny bit of DC which builds up at the output, and causes a pop sound when stomping.

The resistor to ground concept provides a path for that dc to drain. In this case, dc on the output has a path to ground through the blend pot then comp vol pot.

 

[debrad]

OK...I added a 1uF tantalum cap on the output of my buffer with a 100k resistor to ground. Not sure if that fixed my output level issue or if it was removing a TINY bridge between two rows on my vero, but my 100% dry blend level on the compressor now matches my bypass level so I'm happy with that! (I only noticed the bridge after re-flowing all of my soldering on the vero so I may have actually CREATED the bridge doing that!)

Huge thanks to @phi1 and @Feral Feline for your all of your input and my sincere apologies for the hijack on this build report.

Now I just have to make sure those 18v don't blow up my LED and then give the whole thing a good test to see if a guy who hasn't used a compressor (re: me!) can determine whether this build is really compressing or not (ha ha ha).