MixBuzz1 Build/Support thread

[piano]

Briomusic - did you ever get the meter problem sorted out?

Few problems left to deal with:

Meter in IN and OUT modes only shows signal applied to the right channel, but on both bargraphs. What could be causing this? (When calibrating I also noticed that only the right-hand Offset and Range trimmers had any effect on the LED meter.)

I have the same behavior. Signal applied to right channel, but on both bargraphs. Also I get the bargraph lighting up full but signal is overlayed that in a brighter LED. Cool effect but I am sure this is not what was intended.

The fix for this was not posted (well a diode was reversed - but that did not fix the problem.

What is the fix.

 

[Harpo]

edit: sorted!
harpos post made me think about my vca's and it turns out i had R3l/r wrong 4k99 is on the BOM and correct for 4 VCAs but I only have one so 20k is the correct value. back at unity gain now

... and maybe decreasing C1l+C1r from 56pF to at least 15pF up to 22pF.

 

[briomusic]

Briomusic - did you ever get the meter problem sorted out?

Few problems left to deal with:

Meter in IN and OUT modes only shows signal applied to the right channel, but on both bargraphs. What could be causing this? (When calibrating I also noticed that only the right-hand Offset and Range trimmers had any effect on the LED meter.)

I have the same behavior. Signal applied to right channel, but on both bargraphs. Also I get the bargraph lighting up full but signal is overlayed that in a brighter LED. Cool effect but I am sure this is not what was intended.

The fix for this was not posted (well a diode was reversed - but that did not fix the problem.

What is the fix.

@piano,
tbh i gave up on the meter for a while, because I am using it in G.R. mode most of the time. As my mixes are pretty much mono for all the loud stuff (kick, bass, vocals) I can mostly see what's happening from just looking at the right channels levels. For my next builds I decided to have an analog GR meter and no level meter, as I like to see the needle dance just around 2dB Gain Reduction. I have just borrowed an oscilloscope so might poke around the LED meter model for a bit. If I understand correctly, the 4011N chip is an oscillator that switches the 4016N so that each channel's level is displayed in turn. this is clearly not happening in our builds, so that's  where to look for the problem.

@harpo,
C1x is to keep the 5534 from oscillating, right? Why would I need to change it if I change the amount of negative feedback by changing R3x?
IIRC a lower value would mean a higher cut off point for the frequencies that get fed back = attenuated. Am I losing bass respose by keeping the 56pF in there? These are extremely small caps anyway, please explain - thanks!

 

[Harpo]

C1x is to keep the 5534 from oscillating, right? Why would I need to change it if I change the amount of negative feedback by changing R3x?
IIRC a lower value would mean a higher cut off point for the frequencies that get fed back = attenuated. Am I losing bass respose by keeping the 56pF in there? These are extremely small caps anyway, please explain - thanks!

The C1x R3x combination in the inverting opamps feedback loop form a 1st order (-6dB slope) lowpassfilter with -3dB cutoff set @ 1/(2*PI()*R*C). If you want to keep the same/similar frequency response and you change one parts value (R3x from 4K99 to 20K), you'd have to change the other as well.

 

[briomusic]

C1x is to keep the 5534 from oscillating, right? Why would I need to change it if I change the amount of negative feedback by changing R3x?
IIRC a lower value would mean a higher cut off point for the frequencies that get fed back = attenuated. Am I losing bass respose by keeping the 56pF in there? These are extremely small caps anyway, please explain - thanks!

The C1x R3x combination in the inverting opamps feedback loop form a 1st order (-6dB slope) lowpassfilter with -3dB cutoff set @ 1/(2*PI()*R*C). If you want to keep the same/similar frequency response and you change one parts value (R3x from 4K99 to 20K), you'd have to change the other as well.

Trying to calculate what the actual effect on the freq response would be but clearly I am going wrong somewhere. Would someone like to help me:

1/(2*PI()*4990Ohm*0.000056uF)=0.945857513438784 Hz  ...a LPF at just under 1 Hz??!
1/(2*PI()*20000Ohm*0.000056uF)=0.004724555912615 Hz    ....errrm?

 

[Harpo]

Trying to calculate what the actual effect on the freq response would be but clearly I am going wrong somewhere. Would someone like to help me:

1/(2*PI()*4990Ohm*0.000056uF)=0.945857513438784 Hz  ...a LPF at just under 1 Hz??!
1/(2*PI()*20000Ohm*0.000056uF)=0.004724555912615 Hz    ....errrm?

watch paranthesis and pF is 10^-12 F.
1/ (2*PI()*4990*0,000000000056)=569550Hz 
( instead of 0,000000000056 , for your 56pF a 56E-12 or 56*10^-12 is a lesser source of digit error ), so
1/(2*PI()*20000*0,000000000056)=142103Hz as well as
1/(2*PI()*20000*56E-12)=142103Hz as well as
1/(2*PI()*20000*56*10^-12)=142103Hz

 

[briomusic]

watch paranthesis and pF is 10^-12 F.
1/ (2*PI()*4990*0,000000000056)=569550Hz 
( instead of 0,000000000056 , for your 56pF a 56E-12 or 56*10^-12 is a lesser source of digit error ), so
1/(2*PI()*20000*0,000000000056)=142103Hz as well as
1/(2*PI()*20000*56E-12)=142103Hz as well as
1/(2*PI()*20000*56*10^-12)=142103Hz

Thanks Harpo, I ignored the micro in Farad  :-[
Still, I think I can live with a frequency response up to 142kHz.

 

[piano]

Concerning the meter problems Briomusic and Piano have:

It is Igor's mistake - as he shipped 4403 by mistake. I (Piano) have 4403 transistor installed in the meter board. The correct one is 2N3904, 2N4401 or PN2222 etc. I replaced the wrong 4403's with 2N3904's which it improved the situation but one of the IC's was likely fried as I still get aberrant behavior as the left channel is not working and the 2nd to the last LED does not light. I'll buy a new 4016N for IC 4 and see what happens. See Igor's statement below:

"Yes, here is a mistake, I used package from 4403, BUT the right transistor is 2N3904, 2N4401 or PN2222 etc. In case 4403 installed, I doubt IC4 will be damaged..anyway.
If you got wrong transistors (2N4403) with kit and cannot source 2N3904 or PN2222
(they available from every electronic shop for approx. $0.1), just let me know, will send them for free."

See below for history and description of problem which may be caused by using an incorrect part Igor sent with kits.

Briomusic - did you ever get the meter problem sorted out?

Few problems left to deal with:

Meter in IN and OUT modes only shows signal applied to the right channel, but on both bargraphs. What could be causing this? (When calibrating I also noticed that only the right-hand Offset and Range trimmers had any effect on the LED meter.)

I have the same behavior. Signal applied to right channel, but on both bargraphs. Also I get the bargraph lighting up full but signal is overlayed that in a brighter LED. Cool effect but I am sure this is not what was intended.

The fix for this was not posted (well a diode was reversed - but that did not fix the problem.

What is the fix.

@piano,
tbh i gave up on the meter for a while, because I am using it in G.R. mode most of the time. As my mixes are pretty much mono for all the loud stuff (kick, bass, vocals) I can mostly see what's happening from just looking at the right channels levels. For my next builds I decided to have an analog GR meter and no level meter, as I like to see the needle dance just around 2dB Gain Reduction. I have just borrowed an oscilloscope so might poke around the LED meter model for a bit. If I understand correctly, the 4011N chip is an oscillator that switches the 4016N so that each channel's level is displayed in turn. this is clearly not happening in our builds, so that's  where to look for the problem.

@harpo,
C1x is to keep the 5534 from oscillating, right? Why would I need to change it if I change the amount of negative feedback by changing R3x?
IIRC a lower value would mean a higher cut off point for the frequencies that get fed back = attenuated. Am I losing bass respose by keeping the 56pF in there? These are extremely small caps anyway, please explain - thanks!

 

[baadc0de]

In case anyone has been wondering, the EAO lenses provided by Igor work with audiomaintenance switch EA-03-001.

 

[la-2a]

@evilcat....... did you solved the pop problem switching compression  in / out.?...

Hi everbody, 

I have the same issue. It tested the compressor on 2 locations and the problem seems to be
inside the comp. I have it only  on the left side. I checked all caps values . I hope somebody can help and push me in the right direction to solve the problem .This would be great!!!, want to use mixbuzz soon on FOH  ...pops are not very good on 10000 watts  )

2. how many signal reduction  do  you get out of the mixbuzz .? input= 1v 1Khz  threshold fully CCW  -15.  ratio 10:1.?.

thank you in advance
best regards

 

[evilcat]

Hi la-2a,

No, unfortunately, I didn't solved it yet. In fact, I don't know what to do  ! Like you say, it seems to be IN the compressor.
I tested it in 2 locations too and I have the same problem, it even extend to other compressors when they are plugged in series : for mastering stuff I put a Neve 33609, then a TubeTech LCA2-b, then the MixBuzz and they ALL did a pop when I engaged and disengaged them ! Any other idea than a DC voltage problem, anyone ??

 

[briomusic]

have you actually measured for DC voltage on the output with a scope or meter?
the mixbuzz could  be the culprit emitting the DC offset, for example the THAT 1646 can add some DC offset if not compensated with NP caps (not sure how this is handled in mixbuzz). if the mixbuzz is emitting DC voltage, than this would be filtered out by a device later on in the chain. However, if said device also has a relay bypass, the DC would come back in while in hard-bypass.
I have also had, on one occasion, a broken THAT1646 that had DC on the output, albeit it wasn't kicking out sound anymore. I am not sure if THAT1646 is protected against phantom power, either, so that could harm it.

 

[la-2a]

Hi, the pop problem compressor  in / out  disappears  if  I connect XLR Ground pin1
to the Board Ground (audio ground) . For now this solves the problem.....
But  I am still confused why this happened only on one channel. On the other ch.
there was a little pop but not much noticeable. Now  on both channels the pop is still there but only
at -60dBm.

edit :  the pop appears when i release the switch ( compressor out ) before I made the ground connection
it was the opposite .......weird.....

regards

 

[Igor]

The pop can be an issue of DC offset. If the pop is at -60 db, IMHO, that's it.
Check with DC millivoltmeter the offsets between pin 1 and pins 2/3 at compressor's xlr's and equipment you connect
(converter's out, CRM or mixer input, etc).
BTW, check if your grounds (AGND and case) are tied together.
The star ground of compressor is located at center big hole on PCB and should be tied to chassis,
don't tie to chassis the power supply.

Meter issue: check if the bargraph meter switches the LED's, check control signals at 4016 or 4066, etc.
Of course, ma bad, if you installed 4403, use pn2222 or 2n3904.

@evilcat: OMG!!!!! Your build looks ABSOLUTELY cool! Respect, man!
Drooling without stop! This is what I meant: design your fronts, use options as you want!!!!!!!
And many thanx for posting the mouser links and attack/release wiring.

@Harpo, Evilcat, Tekno808, Piano and others who complete your builds:

Thank you guys for your incredible help!!!!!!

BTW, new updated docs are here:

http://ij-audio.com/downloads/

 

[evilcat]

Thanks a lot for kind words Igor, from you, it means a lot to me .

There's definetly DC offset @XLRs inputs when the comp is switched ON :
-8V between pin 1 and 2.
-2.7V between pin 1 and 3.
And at the output of the comp, only when it's switched ON : 0.8 and 1 mV @ OP 1, and 2.3 and -3.9 mV @ OP 2.

My power supply is not connected to chassis.

The pop reach 0 Vu when I switched ON the comp, and -8 Vu when I switched it OFF .

There is minimal DC voltages on my console :
3.7 and 4.1 mV @ insert send
4.0 and 5.1 mV @ bus out
3.2 mV @ mix insert send

What should I do ?!! Thanks.

 

[Igor]

There's definetly DC offset @XLRs inputs when the comp is switched ON :
-8V between pin 1 and 2.
-2.7V between pin 1 and 3.

Dat's crazy! Is the star ground on motherboard connected to chassis?

And at the output of the comp, only when it's switched ON : 0.8 and 1 mV @ OP 1, and 2.3 and -3.9 mV @ OP 2.

Some DC offset within +/-2 mV at the outs of THAT1646 etc is acceptable,
Please check the DC voltages between AGND (starground at PCB) and pins of input balanced line receivers (INA137 or THAT1246):

INA137/THAT1246
PIN VOLTAGE

1  0v
2  0v
3  0v
4  -15v
5  0v
6  0v 
7  +15v
8 not connected

As well, try to pull out balanced line receiver chip and check the voltages without it, on socket
(maybe a contamination? 1% from 100, but, everything can be)

Maybe, dead chip (BOTH same???? Strange!), but I think it is ground issue.

 

[benlindell]

How are you guys breaking apart the headers for the sub boards? I keep breaking the female ones very uncleanly, tried lots of different ways and nothing seems to work nicely.

 

[Igor]

Cutter?))))

 

[Benny]

I uses pliers to break them apart but I had to trim them with a cutter knife to get a clean cut.
Fortunately Igor sent more than you'll need, so no problem breaking one or two.

Edit: Igors answer was faster than mine

 

[Benny]

Just finished my MixBuzz 1 and I have to say this thing sounds great.
No problems, no smoke and if you've read the docs it is a pretty straight forward build. Thanks to Igor and all the others for all the information in the thread.