Summing Mixer (Majestic Buzz)

[Gustav]

I made this Summing Mixer from a schematic by Tim Farrant (with his permission).

And heres a package of some info and a self etch file, in case anyone wants to use it.

https://pcbgrinder.com/download/MB_doc.pdf

Just wanted to share.

Gustav

 

[Gustav]

And a question.

The unit works and measures perfectly, but it lags on power up.

- I don't experience the lag if the audio trafos are out-of-circuit.
- The output pin on my 7815 regulator measured 0.6V while the unit is hanging, but it eventually comes on every time.
- It consistently takes 15-16 seconds for the power to come on.
- Taking the LED out of circuit makes no difference.
- If I don't discharge the cap, I can toggle it on/off smoothly after the initial start-up.

I thought about swapping the regulator to see if it would help, but rather than doing that semi-randomly, and then proceed to trying another cap across the 1000uF replacing it, or replacing it with something larger, I thought it would be more fun to do some more speculative trouble-shooting.

I don't know if resistance in the power rails after the regulator can affect the RC time constant on the PSU cap (1000uF), so I tried to calculate if the 1K5 resistor (only on the positive rail going to an LED), and that could not account for a time delay- unless I botched the decimal point (1.5 seconds).

I am guessing the Voltage Regulator is not the main suspect, since it does come on, so am I correct thinking that the cap is? And can I measure the cap to see if its charging correctly? (Voltage is as expected across the pins).

I know its kinda silly to ask, and I will just try the cap, then the regulator. I might get a working unit without being completely sure what was actually wrong, though.

So I thought I'd ask you smart people!

Gustav

 

[Gustav]

Put a 1000uF in parralel, and it fires up instantly now, so any considerations on wether Time Constants could be affected by resistance in the powerrails/possible even including the 600Ohm on the trafos must be out (Should have raised the time if that were the case afaik).

I will try to desolder the original 1000uF cap in there and see if it still fired up with just the new one, but...

Does the design need a larger cap (Is the tolerance too tight, and if so, why?), or did I have a bad cap? (Don't have a meter that can measure the original cap when I remove it unfortunately).

And if I change the cap on the regulated + rail, should I change the - as well? Not sure wether symmetry is needed.

I am curious about the "Whys".

Again, it only happened with the audio trafos in circuit.

Edit:Just tried it a few more times - I must have had a few lucky hits, cuz its still doing it. Will try the regulator next

Gustav

 

[ruffrecords]

Try disconnecting R49 and R50 and keep the transformer connected. Does it power up OK?

Cheers

Ian

 

[Gustav]

Try disconnecting R49 and R50 and keep the transformer connected. Does it power up OK?

Cheers

Ian

I swapped the regulator (Which was my first thought before I went into over thinking mode). Drained the caps and tried three times now, and it fires up every time. I haven't seen this from a broken regulator before - only dead or alive ones, and I'll check again tomorrow.

What is the thinking behind lifting the 100R caps parallel to the bipolar caps in this scenario? The audio transformers most definitely had an effect, and I am not seeing why, so maybe that will make me understand.

Gustav

 

[ruffrecords]

Try disconnecting R49 and R50 and keep the transformer connected. Does it power up OK?

Cheers

Ian

I was thinking maybe the dc path from the output of one op amp to the output of the other was causing one of them to go into latch up, draw a high current and send the regulator into current limit.

Cheers

Ian

 

[Harpo]

The unit works and measures perfectly, but it lags on power up.
- The output pin on my 7815 regulator measured 0.6V while the unit is hanging, but it eventually comes on every time.
- It consistently takes 15-16 seconds for the power to come on.
- Taking the LED out of circuit makes no difference.
- If I don't discharge the cap, I can toggle it on/off smoothly after the initial start-up.

I am guessing the Voltage Regulator is not the main suspect, since it does come on, so am I correct thinking that the cap is?

Regulators hanging on startup occur in bipolar supplies when there is only negliable/no load connected to 0V reference voltage. (The opamps are only connected across both +/-rails, the LED is only drawing 10mA from the pos.rail. If the neg.rail comes on first, it will shut down the pos.vreg.)
Two 1N400x diodes in parallel to C10 and C11 should fix this latching.

 

[Gustav]

Thanks Ian and Harpo!

Gustav

 

[Spencerleehorton]

hi gustav,

would you be able to put up some audio of something though the summing mixer and a comparison of the same thing not through it please?

Looks great btw.

regards

Spence.

 

[Gustav]

hi gustav,

would you be able to put up some audio of something though the summing mixer and a comparison of the same thing not through it please?

Looks great btw.

regards

Spence.

I am not a good mixer, but I might have a soundclip at "some point" (purposely vague).

My thought is, that any mix made into the mixer will favor the summing on an A/B test, and a mix in-the-box will favor in-the-box!? Get the same person to make the same mix twice, once into both? He will be more familiar with the mix on second pass...

But... I will see what I can do - at "some point"

Gustav

 

[Spencerleehorton]

HI Gustav,

I was thinking just to pass a good track through it and without it for comparison.

regards

Spence.

 

[Spencerleehorton]

Hi Guys,

also my thought was to expand the 16 channels up to 32, could this be easily done using 2 x pcbs?
if that was ok to i would need to add a monitoring section, is this easily implemented?

regards

Spence.

 

[Majestic12]

also my thought was to expand the 16 channels up to 32, could this be easily done using 2 x pcbs?

You don't need two pcbs for that, just add some more inputs with 22k resistors to the summing rails.

 

[iampoor1]

Whats the function of C4/C1/R23?

C1/C4 are used for dc blocking (since it is not a gapped transformer), but why such a large value for C1? Totally lost on the point of R23, I thought the previous resistor set the output load...
Thanks!

 

[Deepdark]

Are left/right channel stereo linked? If I put, for example my bus drum throught it, will it conserve the original stereo panning?

 

[Majestic12]

Are left/right channel stereo linked? If I put, for example my bus drum throught it, will it conserve the original stereo panning?

Both channels aren't linked in any way. Everything you'll send on the right summing bus will be on the right stereo out , same with the left channels.
So your panning won't be changed in any way. It comes out of the summing unit exactly as you send it in.

 

[Deepdark]

Allright thank you

 

[Deepdark]

Hey guys, quick question.  8)

Is it thinkable to add an input gain pot and an output gain pot in that design? The goal is that I would like to control the signal before the tranny, to drive it as desired (adding some saturation??), and an output pot to control the gain to the daw. Is it thinkable, or a total waste of time? Some Log Taper would do the tricks? Which values? 50K?

Thanks a lot guys  ;D

 

[Deepdark]

Oh and another one.  ;D

My system is based around an Apogee Ensemble. So the goal of the summing, in my case, would be to send all my stereo buss into the unit's and send back into the stereo output channel.  So, my output are rated 100ohms on the specs of my apogee, so I guess that 600ohms into the input is ok (looking at the primary of the transformer). But, the input of my apogee is rated 10Kohms, and the secondary of the transformer is 600ohm, too. So, do I have to match the output transformer to get around 10-15K at the output??