24ch API SUMMING MIXER DR-2402

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Thank you but the manual don’t show if there are 1 or 2 star grounds (one for the psu, one for the unit).
On my unit, I reduced my noise by making like this :
- solder small pads on the psu board to have the ground passing through the pcb
- star ground on psu, at corner of pcb, connected to the iec inlet ground and to the pcb ac ground.
- psu xlr ground connected to dc ground of pcb

On the unit, I putted my star ground at the corner of the aca board, marked « ground ».
- this star ground is connected to the ground pin of the xlr Input  from psu, and to the channels boards grounds.
-regarding the manual, I removed the ground wire sitting between the +16v and the -16v of aca board because it is connected by the pcb to the star ground (at the corner of the same pcb)

With this I don’t have 2 points of ground connected together on multiple spots...

I don’t hear noise but my fabfilter eq shows -78dbfs in my daw (audio converters settings : +20db)
Is it a good noise level for this project or is there anything to make it even lower ?

Also, what is the 0 point at the output pot ? I have no gain/no loss with output about 4  o’clock..normal ?

Thank you
 
Completed with a nice grounding / vu meter setted in Db VU
Everything works awesome, -90dbfs on my daw, no distortion.
 

Attachments

  • 1602.jpg
    1602.jpg
    395.5 KB · Views: 241
Oh man getting close now. After a long hiatus I`ve spent the last few days wiring this beast up. Powersupply is up and running, and most of the mixer wired up. Missing a couple of resistors for the aca/bo in order for the first power-up to take place...

I work shift ofcourse, so will be another couple of weeks before I have the time to finish it up, but it really feels in reach now. So looking forward to putting this to good use!
 
fragletrollet said:
Oh man getting close now. After a long hiatus I`ve spent the last few days wiring this beast up. Powersupply is up and running, and most of the mixer wired up. Missing a couple of resistors for the aca/bo in order for the first power-up to take place...

I work shift ofcourse, so will be another couple of weeks before I have the time to finish it up, but it really feels in reach now. So looking forward to putting this to good use!

Hi. Did you test one single channel ? Before having to swap one wire on all channels..
 
Not yet. Isn`t the manual updated to be correct by now? Anyways, Ill be sure to try one channel first. Hopefully it`s wired right  8) (afaik the confusion stems from the volume pot having bottom and top terminals confused in the manual. That and the LED wiring is the wrong way around...?)

A little confusion around switches. As far as I understand "on - none - on" means that there is no center position for the toggle? Aka suitable for both the stereo channels mute switches and the main power switch.

I was hoping to use an illuminated switch for the +/- 16V, but am struggling to find one.
I`ve found this:

https://no.mouser.com/ProductDetail/NKK-Switches/M2122TCG01?qs=sGAEpiMZZMvudeGI7i40XMgdnVkjSffTQp%2F%2Fw8MWVxo%3D

but according to the data sheet, under "Contact materials and ratings", it`s marked as "G" aka Gold: "Rated 0.4VA max @ 28V AC/DC max" so probably not suitable? Being that the DR2402 will be drawing about 1.5 amps?
 
Okay so first power-up worked fine. So far I haven`t run more than one board, and with its 8 opamps + 4 in the aca/bo section, my psu allready seems to be working pretty hard, as the heatsinks (the biggest that could physicly fit inside the psu enclosure) are getting quite hot (too hot to hold your finger on for more than a few seconds). Wondering how it`ll keep up when I add the remaining 16 opamps!

It sounds good and clean when running music through it. I am however experiencing some hum when all faders are wide open. Wondering how you are doing your noise measurements? This might be down to my at this point rudimentary cabling, will do some more testing before I "decide" it`s hum I hear, and just not the background noise. I mean, it`s right down levelwise with what I believe would be crosstalk (that I presume is the leakage of the channel I hear when the channel is muted but main fully open, and monitoring turned way up).

Also wondering on what level the LED`s turn red. They work fine and respond to the level by going from unlit to greenly lit, but I presume my Babyface pro doesn`t go loud enough to pass the "red threshold", as I`ve only tested it unbalanced where the max output of the RME is +13dBu (the mixer will be used with SSL converters in the future, which are 22dBu maximum i/o). The main xlr outs on the RME are +19dBu, I`ll give them a shot as soon as I`ve got the right cabling at hand (my whole studio is packed down at the moment, still trying to find a new nice space for it).

 
Hum is not normal, i never experienced any hum in this design, you have a ground loop. Check contact between all panels. Put your star ground at the corner of the aca board, called "gnd", to avoid multiple connections to the chassis. Pots does not influence the noise level in my build; full open or closed, i still get -90dbfs on my daw. With my audio interface setted up +18db, leds become red about -2dbfs on my daw, but it depends on the sound used..more bass take more level... check grounding and output transformers wires place. Keep them far from any psu cable. Its the best summing mixer ever, after testing few of them. Clean wiring is mandatory. Shielded wires for all pots, not needed for rear panel connections.
 
Would anyone be so kind as to post a picture of their completed power supply? I'm not certain I ordered the correct parts based on how they are fitting on the PCB. Specifically need to see "AC", "B1", and "DC". I'm also curious which trimpots to use. Would these be ok? https://www.mouser.com/ProductDetail/Bourns/3296W-1-502LF?qs=sGAEpiMZZMukHu%252BjC5l7YWXS5CHXEyUCH3WN4C%252BSmS0%3D

Thanks in advance!
 
Nevermind. All set now.

For anyone else confused by the 2402 BOM with no part numbers, look at the BOM for the DR-NEQ3D because it actually has the part numbers listed.
 
Finally got my ad/da hooked up to my DR2402 to see what works and not.

Other than a few channels having their pan reversed and some other tiny problems I`ll sort out, the mixer in general seems to work as it should. I`ve hooked up my SSL Alphalinks to the inputs/direct outputs and the master output, and ran test tones and music through it all yesterday. A few notes:

-All direct outputs seem to be between 1.5 to 2.5 db lower than the input signal. Afaik these should be unity gain? Or is this "close enough"? As I understand it, 6dB is lost in the reciever stage when converting from balanced to unbalanced, and these 6dB`s are (re)gained by the 1646 output driver. All the different channels have slightly different gain, within 1.5-2.5dB`s, all under unity).

-Master section does not seem to have gain "in hand". I thought the booster stage on the 2ACA/BO would have 12dB gain available, but leaving channel and master fader open I get around maybe a dB or two, can`t remember. How are your mixers responding?


Noise seems to peak at -92/-87 dBFS (L/R), with my SSL Alphalinks 0dbFS=+22dBu that should be -70/-65dBu? Well, with all faders open on the mixer and my monitors  on full I still can`t hear the faintest trace of noise. So that`s good news!
 
On further investigation, from the ACA build tips on the capi page:

ACA build tips said:
3. Typically, the vintage API style mix bus is fed with a –2dBu signal thru a 47k bus resistor. So that is a
signal that is unbalanced and 6dBu below the channels direct out. With a -2dBu signal thru a 47k bus
resistor, the output level of the ACA will be approximately -0.5dBu. This is important if you are using
different bus resistors than 47k. If they are not 47k (most are not) you will have to modify the ACA’s
feedback resistors to yield the above levels. If you change the feedback R’s (R10 & R13), you should
also recalculate and change the feedback caps (C11 & C15). The stock –3dB knee frequencies are
shown on the schematic. Try to get as close to those as you can.

4. With the ACA setup to deliver the above-specified gain, apply a –2dBu signal thru a 47k bus resistor to
the ACA’s input, fader at wide open, the resulting level at the output of the 325 Booster circuit should
be approximately +4.45dBu.

So it seems the structure is built around unity gain, so in my build a channel fader wide open, panned hard L/R, master fader wide open results in about 1.2dBFS of gain added. Not too far from the 0.45dBU proposed by the CAPI document (but not exact. I guess this comes down to the dr2402`s receiver stage? Everything else is built according to the ACA/BO specs like 47k bus resistors, 10k fader and according resistors, so not many other variables? I see that the DR2402 receiver stage is using different components from the 2S-LA from CAPI. There`s also no CMRR adjustment involved in this build.)
Also, if the proposed 0.45 dB gain is before the 2503, where`s my additional 6dB?

My direct outputs are about 2dBFS lower than input; this seems to be pretty far off for a +6 part? Can anyone shed some light here? I don`t know how close to unity gain I should expect here.
 
Yep I've tried to contact the DIY-racked "tech" but no answer. The circuit seems to be pretty similar to the JLM Dingo receiver stage tough, and many other 2520 balanced receiver stages.

Anybody actually using the direct outputs on this mixer? I have cabling on the way to fully utilize this puppy (puh, 48 channels of DSUBs is no joke, a pretty costly affair!) and will finally be able to do some more testing. I am planning to use the direct outs feeding my AD, so I would really love to get as close as possible to unity gain! Any tips on where to look for the what is causing this gain discrepancy  between the channels is much appreciated. Could intermittent molex connections cause a gain drop? Or would it have something to do with the tolerance of the resistors surrounding the 1646's? It's built with 1% resistors, but I haven't matched them as I thought this was unneccessary.

Here's hoping....
 
fragletrollet said:
Yep I've tried to contact the DIY-racked "tech" but no answer. The circuit seems to be pretty similar to the JLM Dingo receiver stage tough, and many other 2520 balanced receiver stages.

Anybody actually using the direct outputs on this mixer? I have cabling on the way to fully utilize this puppy (puh, 48 channels of DSUBs is no joke, a pretty costly affair!) and will finally be able to do some more testing. I am planning to use the direct outs feeding my AD, so I would really love to get as close as possible to unity gain! Any tips on where to look for the what is causing this gain discrepancy  between the channels is much appreciated. Could intermittent molex connections cause a gain drop? Or would it have something to do with the tolerance of the resistors surrounding the 1646's? It's built with 1% resistors, but I haven't matched them as I thought this was unneccessary.

Here's hoping....

1% is more than needed, do you think they have to match 0.5% each place on a 48 channels ssl for example ?
I don’t have any drop, it all works as expected with molex and unshielded wires. Totally quiet, everything matched as expected
 
Hey, thanks for your reply.

Only difference between the stock build is I've modded the direct outs to be post fader. I did however try to tap the signal prefader and had the same signal level as post fader (fader fully up).
 
So I finally got the 48 channel snake from Redco and was able to do some further testing.
I've measured the AC voltage between pin 2 and 3 from the input of the mixer all the way to the direct output.

-At the input a +4 dBU (I'm sending out a -18dBFS 1khz from a +22dBU interface) I get about  1.215-1.226 which is pretty close to 1.228 (+4dBU). I've not measured the output of the interface, but I presume the drop is either in my multimeters accuracy (?) or because of the 30ft of DSUB cabling between the interface and mixer. Not too worried here.

-Voltage remains about the same as the input (+/-2mV) throughout the circuit until Rx-9 (the 10k resistor between the input of the 1646 to the output of the 470uf cap/pot input), where the voltage drops to about half, or more. I've substituted rx-9 with a 10k trimmer, and am able to get very accurate trimming of the direct output volume. In the one channel I've tested so far, the trimmer value to get unity gain between the mixer input and direct output is about 5.1k - and it seems other channels will need a similar value.

So my question is: how are you getting unity gain with a 10k resistor?

I will further investigate the ACA/BO gainstaging.

In the CAPI ACA/BO, in the ACA section, are you using 28k feedback resistors as shown in the ACA/BO schematic? The separate 4x ACA schems use 47k feedback resistors, and since the DR2402 uses 47k bus resistors, I wonder if I'm supposed to use 47k in the ACA/BO. Will ask over in the appropriate thread.


Now to get a bunch of trimmers, and de-assemble the unit for the n-th time... :p
 
Okay, fumbling in the dark here, but my best guess is that the 10k resistor forms a voltage divider with the paralelled pan resistors instead of with the 1k fader, and therefore my voltages are off compared to other builders, as I've the direct output post fader instead of the stock pre fader. Luckily it doesn't seem that trimming the 10k down to about 5k affects the pan pot  or volume fader action/response (i don't know if it ever would, just glad the mod works  8) )

I think I'm going to install trimmers in the feedback of the booster stage and trim the output channels so I know exactly what the mixer adds in terms of gain.

Anybody knows the specs of the screws used for all the metalwork that came with the pcbs? I'm having trouble finding them, not regular m-screws. I need to change a bunch after working on this beast for a couple of years now  :eek: :p

 

Latest posts

Back
Top