PSU for 16 channels API mixer

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Rob Flinn said:
I've had to deal with a desk that had a bunch of these PSU's to make a 5 rail supply before & it really didn't work.  Despite bags of current in hand you could not put all the channel strips in without the relays in the desk going crazy switching on and off rapidly, & the desk was very noisy in an HFswitching sort of way.  Maybe it was somethng to do with the inductive load of the relays.

I know that you can differing qualitries of Switch mode PSU.    Cyril Jones was supplying PSU's for his Raindirk Sympony made of Thurlby Thander switching bench PSU's that worked very well & took up a fraction of the space of the 2 really heavy 6u linear supplies they originally had.    So I think a little caution is useful here.
Yes. Definitely caution is required. It would be easy to screw up and make something that behaved poorly.

I'm using two ICs MAX7221 and L9823 which slew rate limit to minimize EMI. I also use separate grounds DGND, PGND, and AGND for "dirty", "power" and "analog" respectively. I run LEDs and relays on DGND and really quiet stuff on AGND. That makes a measurable difference noise (and I believe it would with any type of supply).

But even without these precautions, I think an SMPS can be made very stable and quiet. A 1mH choke (like ALFT-03A-13) with a 470u cap will crush everything over 300 Hz. It has to be a common-mode choke of course so that you're filtering the gound lines as well.

I actually reversed polarity on an op amp once and it survived because of the SMPS. I'm not sure why but it did.

Here's a pic of my PS. Largely ripped off from SoundSkulptor's PS2 (which has schematics on their website BTW):

A5pZdbT.jpg


Things aren't tied down quite like they should be but ehh.
 
iprovlek said:
I bought LM338 positive regulator in electronic shop and guy recommended it cause he gives 5A.....

So I googled and find this sites

http://www.decdun.me.uk/Pedja-Rogic/pedja-gc_decdun/pedja-gc_decdun/supplies.htm


http://alectronicx.blogspot.hr/2010/07/power-supply-20v-and-20v-5a-by-lm338.html


http://alectronicscircuits.blogspot.hr/2010/07/12v-12v-5a-regulated-power-supply-by.html

yes, but first you need to calculate how much power you will need.
You are starting in the end.

Roughly for each 2520 DOA calculate 100mA
Each 5534 IC - 10mA
Each IC for active fader ?

16 x 100mA  (Mic Pres) = 1600mA
32 x 100mA (169 EQ each one uses 2x 2520) = 3200mA
16x4x 10mA (each 169EQ uses 4 5534 IC) = 640mA
16 x Active Fader IC comsumption  (10mA as an example) = 160mA
9x4 x 100mA (each summing amp uses 4x 2520 DOA) = 3600mA

Rough Total power consumption estimate for the +/-18 VDC rails:
9200mA
9.2 Amps

Power needs for 48V rail (16 channels): 1 Amp

Power needs for 12V rails for VU's ????

If I made any mistakes trying to estimate the power consumption let me know.

Read Rob Flinn post, this is not an easy task, and building a mixer of the size and complexity you want it's an advanced project not for a beginner a lot of details have to be thought about. Experience is a must, I would advise you at this point to do something simpler.










 
On active fader is one gar2520 opamp...

And thanks a lot Whoops know you calculate everything for my mixer that is very usefull...


Know I must take one hole channel strip with active fader and calculate exactly how much one strip need off current!!!


But find a few schematics that could be ok for this....


Even LM338 can be used for negative current if you saw on few schematics on link that I post here on this topic!!!!


BUT GUYS THANK YOU VERY MUCH EVERYONE!!!
:)
 
Whoops you advice me very good of complexity of this project but

ILL NOT GIVE UP OF IT....

I have some people here in my town that can help....


But I know that I enter in something that isnt for me!!! But of this reason It attracts me more....
 

Latest posts

Back
Top