"Crush-n-Blend"

[machineintel]

Is there a purpose to having the input debalancing and output balancing drivers, other than only using one pot? Couldn't the circuit be doubled? One for the pos and one for the neg with a stacked pot or a DP 12 pos switcher?

I suppose it's simplifying one thing and complicating another, but wouldn't the input and output amps used now degrade the signal slightly?

 

[dagoose]

BUMP! :green:

I'm in the middle of doing an insertbox (3 x compressor, 3 x EQ) for my mastering and i want to implement this option.
Has anyone build this in practice and how does it work?

I'll try to post the schematic for the relays based box later, i just have a quick drawing here which is hardly understandable even for me :roll:

 

[radiance]

Will those 10uF dc blocking caps be necessary when using a rotary switch instead of a pot?

 

[clintrubber]

[quote author="radiance"]Will those 10uF dc blocking caps be necessary when using a rotary switch instead of a pot?[/quote]
Yes, still. If it's for own use I'd keep them; but in case it's to wrap around a to-be-sold-for-profit-compressor then they could be happily skipped... :twisted:

The clicking-mechanism is different but in one way or another 'a mechanism' is kept: pots with DC are scratchy, but in case of a resistor-string and a rotary switch the taps can & will each have a different DC-voltage on them, because of different offset-voltages of the 2141's.

Most likely not that audible during signal, but it very well may be when no signal is applied.

Bye,

Peter

 

[radiance]

[quote author="clintrubber"][quote author="radiance"]Will those 10uF dc blocking caps be necessary when using a rotary switch instead of a pot?[/quote]
Yes, still. If it's for own use I'd keep them; but in case it's to wrap around a to-be-sold-for-profit-compressor then they could be happily skipped... :twisted:
[/quote]

Well, it's more the never ending quest against caps in the signal path. But than again...this crush/blend thingy will probabely be used with a GSSL comp which does have caps so what gives.....This is probabely a stupid guess..but....is it possible to add some sort of servo circuit to a line receiver like used in this circuit?? Or more specific: What else can be used for DC blocking after a line receiver? I've seen circuits where none was used, no caps, nothing.

 

[clintrubber]

[quote author="radiance"]Well, it's more the never ending quest against caps in the signal path. But than again...this crush/blend thingy will probabely be used with a GSSL comp which does have caps so what gives.....This is probabely a stupid guess..but....is it possible to add some sort of servo circuit to a line receiver like used in this circuit?? Or more specific: What else can be used for DC blocking after a line receiver? I've seen circuits where none was used, no caps, nothing.[/quote]
Of course it also depends on the amount of fiddling you're expecting to do. But a blend-control it won't be rare that an adjustment is to be made.
Up to the user/builder to decide if he/she is willing to put up with eventual little pops. Or just try it: measure the DC across the control, skip the caps, listen.

Using a servo here would be overkill imho, easier alternatives are to use big film-caps in the 'direct' signal path* and/or trim offsets. And/or tweak raise impedance-level a bit to reduce the value of the film-caps.

(*: as opposed to using ehh, again film caps in the 'indirect' signal path of a servo)

Proost,

Peter

 

[dagoose]

Just asking, are we talking about the 'original' crush n blend or the clintrubber version here with MS?
I'll probably go for the original design (just crush n blend) because i think MS is more interesting for EQ in my case and i wamt the crush n blend (ofcourse) just on the compressor side.

Will there be large tracking problems with the potmeter version when i use a decent stereopot or is it better to make it stepped?
Clicking because of DC ofset is not a problem for me, i'll just choose one position and keep it that way during the whole mastering render.

 

[Grooveteer]

[quote author="SSLtech"]Burr-Brown INA134s for the 2141s,
Burr-Brown OPA134s for the 2142s,

Keef[/quote]

Keith,

I'm trying to build this, but ran into one of my noobish problems again; I'm using BB OPA134s for the 2142s, I printed tha datasheet of the OPA134 and I was expecting to see one input and + & - out, right? Datasheet shows + & - input and one output.. What am I missing?

I'm lost here... again

 

[radiance]

I think Keith ment the DRV134 instead of OPA134.
You can also take the That 1240 instead of the ssm2141 and the That 1646 instead of the ssm2142

 

[ChrioN]

Is it possible to use transformers for in and out balancing with good results, instead of the chips?

 

[radiance]

I think the pin numbers in the schematic are wrong.
Looking at the pin out from the ssm2142 pin 3 goes to ground and 4 is input.
8 is output + and 1 is output - but those are swapped to reinvert the signal right?

 

[SSLtech]

Ah -looks like I confused the input and ground pins when I created the SSM2142 component in the drawing software...

Anyhow, the balancing ICs (U1, U2, U4, U5, U6 and U7) are all optional, and can be replaced by transformers, or just unity gain unbalanced buffers if needs be.

The functional part of the circuit is around U3 and VR1... the rest is just there to help it happily 'meet' the outside world.

Keith

 

[Grooveteer]

Just tried to build this on breadboard using INA134, OPA2134 and DRV134, but it distorts horribly. I don't have scope to trace the signal.

I used the datasheets of those chips for pin out identification.

Any idea where to look? Sorry for the noobishness...

 

[clintrubber]

[quote author="Grooveteer"]Just tried to build this on breadboard using INA134, OPA2134 and DRV134, but it distorts horribly. I don't have scope to trace the signal.[/quote]
You hear it distorts, so you might try to let your ears do the scope-job: connect the output-socket at various stages in the circuit to see where it goes wrong.

 

[Grooveteer]

Brilliant! I'll try this as soon as I get home later this week.

By the way Peter; Do you live in :idea: town?

 

[dagoose]

OK, back frm moving houses and stuff so time for some serious diy'ing again :grin:

I'm starting of again with my mastering insert box with crush n blend and i was wondering.
Shouldn't there be some sort of summing resistors between the input and send to comp? :?:
I'm going to build this P2P on veroboard BTW

 

[livingnote]

Hey, I'm working on it too right now. As far as I can tell these SSM chips already have their input resistance in the package. I'll be finishing my PCB design in a few hours, if you want it, give me a holler.

 

[Sleeper]

Holler :grin: :grin:

 

[livingnote]

Oy, ok, it's finished, built to the exact circuit posted, with the exact chips...and it distorts like wacko. I give it a dB to chew on and it goes gremlin on me...hmmm and the distortion seems to not really care about how much you feed it...what's going on here?...

lemme try something here...I'm gonna bypass the OPA and the pot and go straight for line-in line-right back out and see if that works...

 

[livingnote]

funny thing is this...the VU meter is giving me big style indication...even though it isn't getting anything to eat because the TL071 driving the meter is not in place...okay, maybe I screwed up the pins...

ah no. if the OPA is putting out serious stuff then the signal will survive all the way to the meter...boy you oughta hear my system beeps through this thing.


Here is the copper layer

And here is the component side

Just don't etch it unless you wanna go troubleshooting

Ok this gets even funnier: pull the OPA out and you still have the same distortion. Before I forget...

Here's the Eagle schem. If you wanna tinker and do Eagle, send me a message, I can send the .sch and .brd. Excuse the SSMs masquerading as 5534s, I didn't have them in the library on the fly. Pins are right, though.