Crush N Blend Rev 6 V3 (CNB / Wet-Dry Circuit) + Support

[innercityman]

After some inspections, I saw a short between my left out connection and one solder pad of the IC just near by (I took my out connections where the 100p cap's legs were lifted)

In fact after referring to the GSSL layout, I realized that it was not a short, the solder pad where the 100p cap leg is lifted has to be connected to one solder pad of the 5532 IC. So this is a wrong way to explain the issue. Just in case, I changed the place where were solded my out connections as you said in your guide that there is no matter which one of the 2 left holes you use for it, but I'm afraid that this doesn't change anything. So I'll wait for your answer before trying a new test.

One more question though (maybe stupid)... On your wire guide, only the left and right signals for wet, dry and out are connected. So what about the 0 point which is the ground right ? following your guide, I connected the ground for the 15 and 12 volts only. The ground for 15v is connected directly to the chassis of the unit, and the ground for 12 v is connected to the Ov pad on the GSSL just beside of the +12v. So I've just cut the O points (brown) wires. Is it ok with that ?

 

[livingnote]

Hmm - that is indeed a lot of stuff all at once.

For 0V reference - you only really need one connection unless you are doing separate relay and separate
signal ground - other than that, one single 0V line to the main PCB is sufficient.

Going over the wiring guide one more time - yes, I made a mistake in my thinking there, sorry.

The 100p and 15K idea was meant to make things simpler, but I totally overlooked that doing so interrupts
the feedback loop in the GSSL output section and of course causes the amplification to go into saturation.

So it's back to the old fashioned way, then: You will need to restore the feedback loops by at least connecting
a wire to the board from the lifted 15K and 100p and instead lifting the components on the other side, i.e.
the legs with the OUT R and L red markings on them (the 100µ and 100n cap and the 10k resistor), solder them
together and use those as your CnB OUT connection. The holes that remain will be your WET L and R to the
CnB.

Terribly sorry about this but I was completely unaware that of course, you can't just reverse the connection order
without breaking the feedback loop.

Let me know if you have any other questions.

For your inconvenience, I will send you a CnB PCB (unpopulated) free of charge, Cap or Direct version, whatever you
prefer

 

[innercityman]

Ok Lukas, so if I understood, I need to solder back the legs of the 15k resistors and the 100p caps where they were soldered originally, and lift the legs of the 100µ, 100n caps and the 10k resistors and solder them together, and then solder the left and right out wires to them right ? So there will be 3 left holes on each side. And I suppose there is no matter which hole I use to solder my wet connections... That would be nice from you if you could update your wiring guide.

And thanks for the offer, I'll take the classic direct pcb version (the same I bought to you).

 

[livingnote]

Alright - wiring guide updated and should work now. Let me know if you run into any other issues.

 

[innercityman]

Hey Lukas, I've been able to test my unit today after wiring it following your wiring guide updated, and it works perfectly. So nice !

Thanks for the support !

 

[innercityman]

Hi Luckas,

I have a question... When the dry/wet pot is in the center position, I should have a signal half dry, half wet right? in fact on my unit when the pot is completely on the wet side, when I turn it to the dry side, the dry signal comes too fast (in my opinion). I saw that on the back of the 10k pot, there is a screw. I guess it is to adjust the impedance of the pot. So can I try to turn this screw to adjust the balance between dry and wet signal ?

Thanks

 

[livingnote]

Congrats for making it work!

Not really, no - as far as I know there isn't a pot with impedance you can adjust - at least not one that you can
pay for. The 50/50 issue has, as far as I know, to do with the question of how you set your makeup gain - if the
makeup for wet is higher/lower than the unit's incoming DRY signal, it will offset the mid point.

 

[Suburban Ninja]

Hello Lukas.
Thanks you for this great project.
If i  understand your wiring guide, i have to connect each wire (blue,red,green) to gssl pcb and  keep the 2x22uf ,2x15k,2x100uf solder like original gssl without Mod?

Regards

 

[Suburban Ninja]

I didn't see the update sorry

 

[livingnote]

Cool no prob. Did you manage to get it working?

Cheers

 

[innercityman]

Hi Lukas,

I realized when working with my GSSL inserted on the mix buss that when muting right side of the mix, left side was not compressing, when muting left side, right side was still compressing, and when left and right side are on, I have a compressed signal on both sides, and the stereo image is ok. I'm pretty sure I didn't have that before installing the C N' B board. Any idea ?

 

[livingnote]

At what point in the signal chain are you muting? Just so I get it right - you've got the GSSL in the signal chain after all your mixing is done and then you mute one of the two input channels to the GSSL, respectively and get that (off) behavior, right?

The only thing that can cause that that I can think of offhand is if, when installing the CnB, there was some kind of interference that happened in the sidechain summing resistors (the 47k's where it says "SC IN" on the schemo), either because of a solder joint gone bad or because the CnB or something else is pulling the thing to ground somehow.

Is this with Turbo in? Because with the standard GSSL signal path, the thing should either compress on both sides, or not - the CV from either side goes to both VCA's, regardless of what's happening at the input (unless you have a Turbo or other fancy sidechain stuff that feeds the VCAs differently).

 

[innercityman]

Hi Lukas, thanks for the fast reply. In fact I work with an analog console, and the GSSL is inserted on the master out as a mix buss compressor. If I mute one side of the mix or if I move the pan knob from left to right I got the same behavior... pan knob on left side, signal not compressed, pan knob on right side, signal compressed, pan in the middle, signal compressed on both sides. And Yes I have a turbo board installed. I thing you're right about the sidechain that is probably involved in this, I was guessing that too.

 

[innercityman]

Hey Lukas,

I've noticed a problem today that I'm pretty sure I didn't have before... When crush n' blend is on, my right signal is louder than the left one, maybe about 6 db. When it's off the signal is balanced and equal on both sides. You need to know that after some inspections about the difference I had between left and right channel when panning my master knob from left to right, I changed some resistor values on turbo board, to get exact the same values on GSSL's side chain and turbo board. And now that I figured out the issue, the GSSL is behaving the same on both sides without the C N' B engaged, but when engaged, I get this issue.

Any idea ?

 

[livingnote]

Hey

That's interesting - I don't have any idea about what is going on in the GSSL there
but - as for the CnB, you can try removing it from the GSSL circuit and measuring
it out with standard input signals in order to determine if there's a gain issue on
it or not.

You have a fixed-resistor version without trimmers, no?

 

[innercityman]

You have a fixed-resistor version without trimmers, no?

Exactly. PCB rev 11 from Gustav.

I changed the 5,1k resistor on GSSL side chain to a 3,9k to fit the resistor's value on the turbo board. I don't know if this would change anything in the C N' B's signal path...

 

[livingnote]

Ah I meant you have a fixed resistor version of the CnB (though in that one also
trimmers are a possibility). Was anything done to the CnB to counter the
previous imbalance in the GSSL? Really at this point all I can advise is to
first determine if the CnB itself is producing the 6dB by measuring it out separately.

 

[innercityman]

Nothing has been done to the C N' B.

Really at this point all I can advise is to
first determine if the CnB itself is producing the 6dB by measuring it out separately

Ok, so what I should do would be to unconnected it from the GSSL excepting the 12v and ground and power it up right? And where do I need to take my measurements ?

 

[livingnote]

Yes, you disconnect everything from the GSSL except power and then feed the CnB
2x the same input - sine wave, white noise or whatever you want to use for measurement.

Then measure what comes out the other side, testing the L and R channels individually,
and if you have your discrepancy with the CnB standing alone, so to speak, then you can
further probe the signal path within the CnB to determine where the unruly boost is
coming from.

It's best of course if you have an extra set of PCB header connectors with wires lying around
so that you can just unplug the CNB from the GSSL and plug it in to your test rig.

 

[innercityman]

It's best of course if you have an extra set of PCB header connectors with wires lying around
so that you can just unplug the CNB from the GSSL and plug it in to your test rig.

Maybe... I'll check that and let you know.