MixBuzz500 build/support tHread

[rob_gould]

Thanks for replay. I'm an absolute beginner....

1st page of this thread shows me:

"For now, please download docs and BOM!!!!

http://twin-x.com/groupdiy/thumbnails.php?album=405"

When I click on the link it reports me fatal error...

Ah, my mistake.  I thought that the BOMs were actually posted on the font page in one of the early posts of the thread but they are not.

You can get the BOMs and some other useful info from Ramshackle Recording's website :

http://cdn.shopify.com/s/files/1/0256/3541/files/MixBuzz_Docs_V1_4.zip?340

 

[rob_gould]

Did anyone get this project working with the DBX Gold Cans?  I am just trying to calibrate now and I'm seeing a 6 - 7dB drop when engaging the compressor with make up gain at 0.  Just wondering if the resistor values I used (which Igor posted in the Mixbuzz1 support thread) are correct or not.

Using DBX202C gold can VCA's in MixBuzz1:

DBX 202C GOLD CANS
RXX APPLY TO BOTH L AND R (R100L, R100R)
R6 40K2
R2 1MEG...1.5 MEG
R100 710R
R4 JUMPER/FERR BEAD
R3 40K2
C1 5PF


Also struggling to get the DC offsets to set correctly with OPA 2604 in IC 10 and IC11 positions, but I'm more concerned about the drop in level when the compressor is engaged at the moment...

 

[rob_gould]

OK, so maybe I was wrong to blame the resistors around the Gold Can for the drop in level when engaging the compression circuit.

I am injecting a 0.775V sine wave, and I am measuring round the circuit using my multimetter set to AC.  I am measuring using audio hot as my reference voltage.

If I measure at pin 2 or 3 of the THAT 1246 I see my .775V

If I measure at the output out the THAT 1246 I see .390V

If I measure on the left hand side of R6 (on the schematic) I see .390V

If I measure on the right side of R6 on the schematic, the same point as DBX I-IN, I see  .780V

If I measure at the output of the DBX (I_OUT) I see .780V

So it looks like the level is maintained through the Gold can, even with the compressor engaged.  Does this seem right?

If I then measure at pin & of the opamp, OPA2604, I see .580V.

So the audio level seems to be dropping inside the OPA2604.

Can anyone comment on this?  Any ideas what might be going on?

Thanks!

 

[daveee]

Hi,
Does anyone have the MixBuzz1 v2.5 schematics they could upload please?
All links I've found seem to be broken.

 

[rob_gould]

Audio circuit schematic now attached to my post two above.

A question - R3 and R6 were specified as 40K2 by Igor, but in the design note 127 from That Corp, they specifiy 50K in those positions (figure 3 here in this doc : www.thatcorp.com/datashts/dn127.pdf)

Why this difference?  Could this account for the fact I'm not seeing unity gain through the compressor circuit?

Thanks,

Rob

 

[Harpo]

I am injecting a 0.775V sine wave,

between XLR-pins2/3 and this AC voltage measured between these XLR pins 2/3 with your sine generator connected.

and I am measuring round the circuit using my multimetter set to AC.

AC volts at a frequency your multimeter can handle, most often far below 1kHz.

I am measuring using audio hot as my reference voltage.

Dunno of any circuit where an audio hot reference (moving target) would be useful. Use the 0V reference voltage.

If I measure on the right side of R6 on the schematic, the same point as DBX I-IN, I see  .780V
If I measure at the output of the DBX (I_OUT) I see .780V

These VCAs are current-in/current-out devices, these pins are already called I-In and I-Out as a reminder. Current is measured in amperes (in this spot and in your case below 0.4mA), not in volt.

If I then measure at pin & of the opamp, OPA2604, I see .580V.
So the audio level seems to be dropping inside the OPA2604.

Assuming opamp-out pin 6, with same value resistors R3 and R6 in front of the VCA (40K2?) and feedback resistor between opamp pins2/6, AC level will be the same with control voltage 0.000V and compensation cap C1 increased to a useful value (maybe 22...47pF). Your 5pF wouldn't help the opamp from oscillating at a frequency outside the range of bats with capacitive component output of the VCA in front connected.

A question - R3 and R6 were specified as 40K2 by Igor, but in the design note 127 from That Corp, they specifiy 50K in those positions

With THAT1246 in front and +/-18VDC supply, the debalancer will not put out more than |18V-2V| peak. The VCA input will not see more than 16Vpk/40K2=0.000398A=0.4mA. Sounds safe enough, except for the +/-18VDC supply when the datasheet states the upper limit for the DBX202C to be +/-15VDC +/-10% +5%, giving +/-16.5VDC +/-15.75VDCabs.max.

 

[rob_gould]

Thank you for your reply Harpo.  I think I'm misunderstanding some stuff, so I have some follow up comments / questions.

Yes - I'm measuring the sine wave between pins 2 and 3.  The reason I was trying to measure around the circuit using pin 2 as a reference was because I don't see anything on my meter or scope when using ground as a reference.

My audio interface is not truly balanced; it's impedance balanced.  I wonder if this is making things harder to take a reading.

I see from what you've written that measuring voltage on each side of the VCA doesn't really mean anything, but this leaves me stuck about how to measure the level at different places in the circuit.

Also, when you say

Assuming opamp-out pin 6, with same value resistors R3 and R6 in front of the VCA (40K2?) and feedback resistor between opamp pins2/6

Do you mean pin 6?  Pin 7 is the output pin of NE5532 and OPA2604.

Am I correct in thinking that resistors R3 and R6 effectively control the unity gain value?

And am I right to judge from your answer that yes, with the resistor values I've used, I should get unity gain?

And does C1 have any effect on setting unity gain?  Or is its job just to cancel oscillation?

Thanks

 

[Harpo]

Yes - I'm measuring the sine wave between pins 2 and 3.  The reason I was trying to measure around the circuit using pin 2 as a reference was because I don't see anything on my meter or scope when using ground as a reference.

The input of the balanced line receiver THAT1246 exclusively operates the differential between pins 2/3 without reference to and whatever you might call "ground" (at least misleading without qualifier like chassis-ground, safety-ground, psu-ground, ..., they are all different potential with different meaning). After debalancing, the Vac signal level can be measured in respect to 0V reference voltage.

My audio interface is not truly balanced; it's impedance balanced.  I wonder if this is making things harder to take a reading.

No difference for 99.9% of the usual suspects when not running loooong wires in rf-contaminated locations.

Also, when you say

Assuming opamp-out pin 6, with same value resistors R3 and R6 in front of the VCA (40K2?) and feedback resistor between opamp pins2/6

Do you mean pin 6?  Pin 7 is the output pin of NE5532 and OPA2604.

Was the obviously wrong assumption, your '...pin & of the opamp...' might be referencing a pin 6.

Am I correct in thinking that resistors R3 and R6 effectively control the unity gain value?

yes, as long as you don't miss the rest of the sentence 'with control voltage 0.000V' ...

And am I right to judge from your answer that yes, with the resistor values I've used, I should get unity gain?

yes, with no faulty parts. (you maybe operated -different from schematic DBX202X that allows a +/-18Vdc supply- a DBX202C VCA without circuit modification outside its supply limits. Unfortunately exceeding parts abs.max values most often is a single try experience. Maybe you got lucky and your VCA survived...)

And does C1 have any effect on setting unity gain?  Or is its job just to cancel oscillation?

Yes and yes, depending on frequency. The cap in conjunction with R3 set a LPF with its -3dB point in Hz set at 1/(2*PI()*R*C) with R in ohm and C in farad. (from some posts above your IMHO too low value C1 would be 5pF = 0.000000000005F = 5E-12F, setting a HPF at 792kHz)

 

[rob_gould]

Yes - I'm measuring the sine wave between pins 2 and 3.  The reason I was trying to measure around the circuit using pin 2 as a reference was because I don't see anything on my meter or scope when using ground as a reference.

The input of the balanced line receiver THAT1246 exclusively operates the differential between pins 2/3 without reference to and whatever you might call "ground" (at least misleading without qualifier like chassis-ground, safety-ground, psu-ground, ..., they are all different potential with different meaning). After debalancing, the Vac signal level can be measured in respect to 0V reference voltage.

OK, clear.  I get .338V when measuring at pin 5 / 6 of the THAT 1246 with respect to audio ground.  That looks OK to me (1/2 of .775V)

My audio interface is not truly balanced; it's impedance balanced.  I wonder if this is making things harder to take a reading.

No difference for 99.9% of the usual suspects when not running loooong wires in rf-contaminated locations.

Thanks for the confirmation

Also, when you say

Assuming opamp-out pin 6, with same value resistors R3 and R6 in front of the VCA (40K2?) and feedback resistor between opamp pins2/6

Do you mean pin 6?  Pin 7 is the output pin of NE5532 and OPA2604.

Was the obviously wrong assumption, your '...pin & of the opamp...' might be referencing a pin 6.

Apologies, I didn't see that I'd mis-typed the pin number in my initial post.  My fault.

Am I correct in thinking that resistors R3 and R6 effectively control the unity gain value?

yes, as long as you don't miss the rest of the sentence 'with control voltage 0.000V' ...

Is pin E_C on the DBX VCA the the place to measure this?  I get -0.001V there

And am I right to judge from your answer that yes, with the resistor values I've used, I should get unity gain?

yes, with no faulty parts. (you maybe operated -different from schematic DBX202X that allows a +/-18Vdc supply- a DBX202C VCA without circuit modification outside its supply limits. Unfortunately exceeding parts abs.max values most often is a single try experience. Maybe you got lucky and your VCA survived...)

I've only used the VCA in this circuit where the supply voltage is + / - 16V and I haven't messed around with resistor values, so there shouldn't be any issue with having damaged the VCA.

And does C1 have any effect on setting unity gain?  Or is its job just to cancel oscillation?

Yes and yes, depending on frequency. The cap in conjunction with R3 set a LPF with its -3dB point in Hz set at 1/(2*PI()*R*C) with R in ohm and C in farad. (from some posts above your IMHO too low value C1 would be 5pF = 0.000000000005F = 5E-12F, setting a HPF at 792kHz)

I'll put a 33p WIMA in there because I have some of those in stock.


One further observation.  I wasn't testing with the meter connected at first.  When I did connect the meter, with the compression circuit engaged the needle jumps to about 6.5dB of compression, so similar to the 7dB reduction I'm seeing on the meter in my DAW.

To me this would indicate that the problem is actually somewhere in the compression circuit, because with threshold and makeup gain both set to zero, I'd expect to see the meter read zero compression if the problem with the drop in level was due to something in the audio circuit...  Does that sound right or not?

Thanks once again for your help!

 

[Harpo]

Is pin E_C on the DBX VCA the the place to measure this?  I get -0.001V there

Better use the output pin of the driving opamp for an easier rescaling, but it can be used. With your 50mV/dB or 20dB/V control law, your -0.001V at Ec will cause a current gain of 0.02dB with a working following current-to-voltage converter (a working IC11b would try its best to hold its virtual-ground pin5 at same potential as pin6, so VCA or opamp is either broken or oscillating). The 650R control port impedance of a DBX202C is different from a DBX202X, so your R100 or R60 or R55 parts value would need adaption for similar result.

I've only used the VCA in this circuit where the supply voltage is + / - 16V and I haven't messed around with resistor values, so there shouldn't be any issue with having damaged the VCA.

+/-16VDC is still more than the from datasheet allowed +/-15VDC +5%. Maybe easiest done with the -0.6 voltage drop thru a series diode to each DBX202C supply pin.

I'll put a 33p WIMA in there because I have some of those in stock.

Ok. LPF at 120kHz should help oscillation.

One further observation.  I wasn't testing with the meter connected at first.  When I did connect the meter, with the compression circuit engaged the needle jumps to about 6.5dB of compression, so similar to the 7dB reduction I'm seeing on the meter in my DAW.

To me this would indicate that the problem is actually somewhere in the compression circuit, because with threshold and makeup gain both set to zero, I'd expect to see the meter read zero compression if the problem with the drop in level was due to something in the audio circuit...  Does that sound right or not?

Probably coincidence. With 0.001V at Ec, your meter will not deflect to 6.5dB (whatever uA this might be on your scale). Get the audio signal unity gain correct first before focusing on the disco/eyecandy section. Meter response (yours is a 1mA fsd meter for sure?) can be corrected by tweaking (probably increasing) R50 in a later stage.

 

[rob_gould]

Is pin E_C on the DBX VCA the the place to measure this?  I get -0.001V there

Better use the output pin of the driving opamp for an easier rescaling, but it can be used. With your 50mV/dB or 20dB/V control law, your -0.001V at Ec will cause a current gain of 0.02dB with a working following current-to-voltage converter (a working IC11b would try its best to hold its virtual-ground pin5 at same potential as pin6, so VCA or opamp is either broken or oscillating). The 650R control port impedance of a DBX202C is different from a DBX202X, so your R100 or R60 or R55 parts value would need adaption for similar result.

I tried NE5532 for IC11 and the measurements were the same.  Seems IC11 is not the problem here.  R100 was adjusted as per this post: http://groupdiy.com/index.php?topic=41654.msg587347;topicseen#msg587347 but I did use a 715r resistor as I couldn't get 710r.  I have not examined whether R60 / R55 need to be changed because Igor did not mention these resistors when getting the gold cans working in Mixbuzz1.  Happy to hear any opinion or advice on these though.

I've only used the VCA in this circuit where the supply voltage is + / - 16V and I haven't messed around with resistor values, so there shouldn't be any issue with having damaged the VCA.

+/-16VDC is still more than the from datasheet allowed +/-15VDC +5%. Maybe easiest done with the -0.6 voltage drop thru a series diode to each DBX202C supply pin.

OK, I'll look into this.  In fact my 51X PSU is running slightly low at 15.75V / -15.8V so pretty close to the 15V + / - 5% value quoted on the datasheet, but I'd rather get the VCAs running as close to the correct voltage as possible.

I'll put a 33p WIMA in there because I have some of those in stock.

Ok. LPF at 120kHz should help oscillation.

I tried 33pf on one channel and sadly it didn't make any difference to the unity gain issue when compression is engaged.  I'll swap the other channel in due course to minimise the possible oscillation problem though.

One further observation.  I wasn't testing with the meter connected at first.  When I did connect the meter, with the compression circuit engaged the needle jumps to about 6.5dB of compression, so similar to the 7dB reduction I'm seeing on the meter in my DAW.

To me this would indicate that the problem is actually somewhere in the compression circuit, because with threshold and makeup gain both set to zero, I'd expect to see the meter read zero compression if the problem with the drop in level was due to something in the audio circuit...  Does that sound right or not?

Probably coincidence. With 0.001V at Ec, your meter will not deflect to 6.5dB (whatever uA this might be on your scale). Get the audio signal unity gain correct first before focusing on the disco/eyecandy section. Meter response (yours is a 1mA fsd meter for sure?) can be corrected by tweaking (probably increasing) R50 in a later stage.

Yes, it's a Sifam 19W 0 - 1mA meter.  But yes, I'll ignore this for now.

 

[rob_gould]

Hi Harpo

Did you see my last post? Can you advise on anything else to check? I'm really stuck here and don't want to start changing things randomly in case I break a different part of the circuit

 

[Harpo]

Pull out the audio VCA and jumper VCA-In to VCA-Out (pins at center of the smaller VCA module sides) with a piece of wire (diameter and length of a paper clip might come handy if you have the VCA pins in sockets). If you now have undistorted audio with unity gain level, you know the culprit ...

 

[rob_gould]

Pull out the audio VCA and jumper VCA-In to VCA-Out (pins at center of the smaller VCA module sides) with a piece of wire (diameter and length of a paper clip might come handy if you have the VCA pins in sockets). If you now have undistorted audio with unity gain level, you know the culprit ...

Hi Harpo,

This went off the boil for a bit but I want to get back on it now.  I will do perform the test you suggest this weekend. 

Do I need to remove both VCAs or can I just test one channel and leave the Gold Can in the other channel?

Thanks,

Rob

 

[Harpo]

Pull out the audio VCA and jumper VCA-In to VCA-Out (pins at center of the smaller VCA module sides) with a piece of wire (diameter and length of a paper clip might come handy if you have the VCA pins in sockets). If you now have undistorted audio with unity gain level, you know the culprit ...

Hi Harpo,

This went off the boil for a bit but I want to get back on it now.  I will do perform the test you suggest this weekend. 

Do I need to remove both VCAs or can I just test one channel and leave the Gold Can in the other channel?

Thanks,

Rob

Doesn't matter. Your measure for unity gain (complete unit, XLR voltage in=XLR voltage out) at the jumpered/audio-VCA removed channel. If you have more than one multimeter and a spare jumper, you could measure both channels simultaneously for whatever reason ...

 

[Enchilada]

Hey guys. Does anyone know where I can pick up the PCB kits from it if they're still being made. Been checking this site for a while now and it's been out of stock for ages

http://ramshacklerecording.com/products/mixbuzz-pcb-set

Thanks,

Kris

 

[rob_gould]

Put a wanted ad in the Black Market?

 

[jasonallenh]

Unfortunately, Igor has disappeared, so I don't know if James at Ramshackle has the permission to make more... I wish we had some sort of rule/agreement in place that would allow for such projects to continue.

=jh=

 

[rob_gould]

All true, but my suggestion was to put a wanted ad in the Black Market so someone with PCBs could sell you theirs.  That is how I got mine long after Igor had disappeared and Ramshackle had sold out.

 

[jasonallenh]

Understood