24 channel mixer the "Stereotype"

[hakanai]

sorry for the lack of updates. I was out of town for the weekend and had sessions yesterday and today.

So I am going to do a run of boards, both pan and main boards. The latest version of the pan will accommodate both the far more readily available 12 position grayhill, and the original 10 position. it will also accommodate the 2 pole 6 position. So that's just about all the common switches including both of these:

Hi,
http://www.audiomaintenance.com/acatalog/GR-03-002_extended_info.html is this the correct partner switch to the PCBs?

Cheers,
B.

The give a final update on the main boards in the next day or so.

oh and baadc0de, that thing looks like a beast. Good work. Let me know if I can help.

don

 

[galapagoose]

fantastic work!!

this project is inspiring and exactly what i've been looking for..

i know i'm new 'round these parts, but i would love to pick up 2 of the 8 channel boards... i'll be watching eagerly!

 

[shabtek]

@baadcode

i think noise is an issue with passive summing especially when input configuration/termination can be set up so many ways.
Would virtual ground summing be more forgiving?
Good ideas and reasoning for the features.
If I ever do something like this I would want a few more bussing(aux) options and solo/mutes...

 

[baadc0de]

Shabtek and hakanai, thank you for your replies

I'm thinking hard on how noise could be evaluated now in the design stage. Going for an active design would be fine by me, but while I can design a passive system like this (if not easily), an active one is way over my head. I think maybe the summing part could be done by hakanai's boards anyway.. so this is a basic block diagram on how that would be laid out:

But I have no idea how to do the buss select switch? can it be like in a passive design, just a 3p4t rotary that sends LCR to select an appropriate 4buss? Maybe if someone would enlighten me how this could work with and if it could work with the boards presented here?

Thanks,
B.

 

[funkmuffin]

The give a final update on the main boards in the next day or so.

oh and baadc0de, that thing looks like a beast. Good work. Let me know if I can help.

don

Don -- any updates to offer on the mainboards?  I've been hard at work assembling GAR2520s and am anxious to start on the rest!

No hurry, though -- 28 DOAs will consume some serious time while you sort things out.

Thanks!

 

[shabtek]

@ baadcode (are we getting OT?)
do you mean to have pans twice?
do some research on summing, there is a good paper around by Forssell and/ or Hardy.
classic API has some pcbs that can be good for this

 

[baadc0de]

@ baadcode (are we getting OT?)
do you mean to have pans twice?
do some research on summing, there is a good paper around by Forssell and/ or Hardy.
classic API has some pcbs that can be good for this

Hi Shabtek. Yes, there are pans before busses and pans for individual busses as related to the master buss. That's how you could have a mono buss panned centre. About being OT... yes, maybe. Just wanted to get feedback if the boards being produced here could be adopted to my use case. I know the pans can be, but it would be nice to know if the individual channel buffers and the summing network could be. Though the way this is currently laid out, I assume I'd have to have busses x channels opamps inside - yikes!

 

[hakanai]

Sorry for the absence, I have been a little swamped for the last week or so.

@ baadcode (are we getting OT?)
do you mean to have pans twice?
do some research on summing, there is a good paper around by Forssell and/ or Hardy.
classic API has some pcbs that can be good for this

I think this is pretty much right on topic. More mixers the better.
I think he is trying to implement the pan control you would normally find on bus masters.

But I have no idea how to do the buss select switch? can it be like in a passive design, just a 3p4t rotary that sends LCR to select an appropriate 4buss? Maybe if someone would enlighten me how this could work with and if it could work with the boards presented here?

As long as you want a bus select and not a bus assign it should be pretty straight forward. That is, a selector, not a mult. Any switch that has 2 poles and at least 4 throw will do the job. so that 2p6t switch that fits my pan boards would work, or any of the many 1 pole 2 deck switches out there would do it. I found a dozen or so on digikey pretty easily. You had mentioned a 3 pole switch, I assume this is because you are thinking of the LCR as three separate signal paths. I suppose you could do a quasi 3 buss system where the mono signals only join the stereo signals at the last stage. It wouldn't get you much other than the opportunity to implement a funny kind of faux M/S insert, but you'll need 3 channels of compression and a real M/S circuit would probably be cheaper to implement. Ha! that would be kinda interesting to fool around with though.

anyway, yeah it's pretty simple to do the bus select.

I have a bit of time tomorrow to finish labeling the revised mixer boards and order them. I'll get a quote as well and let you guys know what these are going to run.

EDIT: I was posting at the same time as baadcode, so here is more crap

So you don't need more opamps for your design as long as I am understanding you correctly about the selector switch thing. As for my mix boards, you could use everything up to the pan. That is pretty much everything except the bus resistors. So you would go: my board and pan>a 2p4t switch>bus rails. You could hang the buss resistors off the switch if it has solder lugs or make yourself a nice little bus board like the ones you see in the passive boxes all over this board.

 

[baadc0de]

@hakanai
you're right, I just realized this moments ago.. LCR can be easily changed to LR and then 2p6t bussed around. Thanks for clearing this up I'll be interested in the boards too then.

EDIT: also, yes, an LCR insert could be neat or at least very interesting.. on the master. Would be funny to see if there could be an LCR GSSL

 

[3nity]

How you managed to put 2 x 225L in the same slot as a 500 slot??
Just curious..

 

[hakanai]

@hakanai
you're right, I just realized this moments ago.. LCR can be easily changed to LR and then 2p6t bussed around. Thanks for clearing this up I'll be interested in the boards too then.

EDIT: also, yes, an LCR insert could be neat or at least very interesting.. on the master. Would be funny to see if there could be an LCR GSSL

You can still use LCR with the 2p6t. I have made toggles like this for a little mixer i was fixing. It's very simple and effective.

How you managed to put 2 x 225L in the same slot as a 500 slot??
Just curious..

That's a rack i designed and am currently debating releasing into the world for sale

 

[hakanai]

sorry, spent the last day or so chasing my tail in my pcb program. It wouldn't let me export gerbers because of some phantom pin not being connected. So after going through every pin and redoing some sections to clear out any bugs, I remembered I had shrunk the board size from 13x4.5 to 9.3x4. So of course the program didn't dispose of the pins i left off the board when resizing. so a quick enlarge-delete-reduce and we're out the door. I'll let you know when i get some numbers back.

 

[hakanai]

Ok so my pcb guy has gotten back to me and it looks like I can sell the main boards for $55 a piece and the grayhill boards will be sold in lots of 8 for $30. That being said, let me go through the revisions that were done to both.

first the main or "stereotype" board, 6 main differences from the board in original pictures:
1. All the wire to board connections are on screw terminals
2. included the option for dc blocking on the input
3. included the option for CMRR trim
4. included the option for a fader or attenuator between the 2520 and pan
5. the board is significantly smaller with a total size of 9.3" by 4" (mine were 13" by 4.5" and required a big f'n case from Par)
6. These are 8 channel boards (mine were 12)

but just to be clear, you can build it exactly like mine and it is really simple. I only added the optional stuff to make the board a bit more flexible.

As for the Grayhill board, it's almost an entirely new board from the one in the pics:
1. it can now accommodate a 12 or 10 position series 71 switch
2. it can now accommodate a 1 or 2 pole version of these switches (ONLY 1 DECK though, very important difference)
3. Basically this board went from being a pan to being a resistor ladder for any use. It really is a pretty flexible board now.

I will include a basic BOM for the stereotype boards, but it is pretty unnecessary for the grayhill board, or rather here it is:

1 grayhill series 71 single deck PCB mount switch
3-11 resistors (i used Vishay cmf55's)

there you have it

the order is going in today, I should have the boards in about 2 weeks. You can email me at [email protected] to reserve boards and when they are ready to ship to you, we will sort out payment.

I am only going to do a single short run of these boards, so I don't think it is necessary to move this to the grey market, but any moderator please let me know if you would like me to.  

 

[baadc0de]

Okay, 24 channels worth of main boards and grayhill pan boards for me please. Thanks for doing this!

 

[Brolik]

I would like a set of the Grayhill/Resistor ladder boards please.

 

[keefaz]

One main board for me please, I have sent an email

 

[galapagoose]

i'm definitely interested in 2 main boards for a lovely 16 channel mixer!
>> emailed you

 

[Tubemooley]

I just e-mailed to reserve two 8-channel boards. And I went to your website. Rooking good. Thanks, Dana Walcott.

 

[reddwarf]

Hi,
Just sent you an E-mail for two 8 channel boards. Thanks!

 

[hakanai]

So after a few requests for some more info on the circuit I realized I really hadn't given any details. So here is a quick schematic. The story of this circuit is pretty simple. I had gotten this old D&R mixer with some dead pan pots and wanted to replace them with LCR toggles. while doing that I also messed with a 5 position grayhill for one channel. After having that thing apart, re-capping, and getting it up and going again, I REALLY wanted to build myself a mixer. So I had a look at the original 2520 spec sheet and saw the layout for using it as a diff line receiver and thought that would be cool. Then I saw Jeff's summing amp board with it's preferred 47K bus resistors and it all clicked. The pan pots that I had been messing with in the D&R used 47K resistors in it's pan section. So it's basically the API line receiver schemo feeding the D&R pan section, into Jeffs summing board. Really simple. Now, the original API spec was for R1-R4 to be 100K and C1 was 10pf. I mentioned this project to Jeff and he said his console was 20K and 82pf, so I went with that. I used precision resistors for everything so that pretty much sorted the CMRR, but having the option to trim for best results seemed like a good idea for the mark2.
I asked an EE friend of mine what he would suggest for DC blocking and he said 47-100uf at 35v would do it. I spec'd and designed for some panasonic ECE series bi-polar 47uf's. I also spoke with Jeff about the fader option and he mentioned the general rule was a 10:1 ratio to the pan, hence the 1K fader. There is a screw terminal for this with a signal and ground point for each channel. So please note these are untested options.

don

ps oh and for anyone keeping track, i did a last minute revision and the board is now a whopping 9.5" (that's up from 9.3")
Schematic - http://www.hakanairecording.com/mixerpics/Stereotype%20mark%202.pdf