Line mixer (active)

[enthalpystudios]

i'm mostly in it for the experience, and will probably be working on a mixer for a pretty long time. in fact i probably will buy a cheapo in the meantime, and use it simply to monitor. route to tape via a patchbay and outboard cool stuff, and probably mix on the cheapo as well. bear ringer or somethin.

i'd like to build a really cool mixer that does just what i think i'd like in a mixer..... dont really mind if it takes me 5 years.

besides, i have lots of time, and certainly not enough cash for anything sweet.

although i certainly see mci's going for cool prices here and there.

i'd like to build up a simple, small, 8x2 guy based on this design, personally, and for this, i think that scum's 3U idea is a great one. and to be honest, while it would be *cool* to try and learn (for me) more about using DOA's in these types of apps, IC's will work just fine for me. Besides, nice ic's are nice ic's....

and regardless of what i do, cheapo, some wierd custom build, for a big 24 channel guy, who knows maybe i'll step on some cash to buy something.... an 8x2 will still be invaluable to me for playing live, as I'd like to be able to give out premixed stereo stems, as XLR, for certain things.

SO...... dont want to get off topic, but Dave has some relevant remarks... ii mean the bottom line is every wants some type of mixer that makes all their wildest dreams come true. and to build it is a ridiculous endeavour. i know its pretty much one of my biggest goals as a hobbiest and student of audio elec.....

a pcb, however, that fits everyone's bill, is totally not going to happen.. i mean, if we went there, i wouldnt mind a buffered insert blend and a bypassable variable all pass phase alignment knob on every channel. i'd take those sooner than i'd ask for more than 2 aux sends, or many many busses. its just different for everyone, so to make something cool, and something thats going to happen and be a learning experience, we should just keep it simple, and hey, DOA's could be awesome for the mix amps I say.

if it could all fit into a 3 or even 4U rack, an 8x2 that is, that would damn exciting, if you ask me.

and yeah, scum, i'm totally into the alpha pots and those ITT's. honestly i *really* like alpha pots lately. the feel is condusive to my tastes.

also, looks like a 15pin edge connector could also work well... maybe we could just make it 500-series *sized*, even if not totally compatible, maybe it could be 'kind of' compatible.'

like, if you're willing to wire up you're lunchbox a certain way. besides, so much cool stuff is hitting the surface around here in that format, and I know that I wouldn't mind making that mechanical shape multiple times, getting a good formula down for making a 500 series lunchbox diy.

also maybe it's from my nightmares with my mackie24/8, but ribbon cable sounds like babysitting a bad kid to me.... then again, that could be the way mackie did it, i dont honestly know....

i'm into it though, its really cool to look at this schem just to start getting a better idea about how things are done.

and regarding solo, seems like it could be an option, the way i'm imagining things. but maybe not... anyway, for what its worth, i'm personally a fan of pfl, and think it will be handy to have a 3 meter setup, mains and pfl, and for setting previous levels and such, its nice to be able to meter directly, i dont know..... pfl always seemed the proper way to do things for me, always works best in my situatuions.


well good stuff.. thanks again dave, and thanks scum for getting down into it!

oh yeah, as far as caps go, i'd like to spend money on nice caps where its useful, and personally wouldnt mind keeping lytics out of the signal path if possible.

billy

 

[falken]

hello. I have one suggestion if this project really might get made into a kit. I propose we get between 1 and 3 aluminum channel strip layouts and make the aluminum strips available. then its every man for himself on the box enclosure, channel design, circuitboards, components, and final price. that would be cool. maybe there could even be some "add on" short strips to make extra goodies. it would be cool to see what different people could come up with.

 

[the scum]

OK, an update on where I am with this.

I've been tinkering with the board layout. I've captired the input channel schematic and made a mockup of the board, though still a little rough. I just wanted to see how board layout was going to fit together with real-world parts. I'm using Eagle, and will post the files once they're more polished!

I'm also waiting to see of Dave has anything cool up his sleeve.

None of the following is in stone yet, and reasonably malleable.

The updated schem provides dual-footprints for both the opamps in the channel. If you wanna go whole-hog with this, the board will provide for it. I also used the ground sensing summing scheme. I'm going to fiddle with the input buffer a bit, to see if I can rig it so that by stuffing a few different components, it can be the simple follower from Dave's schem, or a differential receiver.

When I do the summers, I'll use the dual footprint for the summers and output drivers. I tend to like DOAs on the outputs, because they'll drive 600 ohm transformers. Only the VU driver will be strictly monolithic.

In sourcing parts I hit a couple of snags:

It looks like the MTH series of switches have been discontinued. I've hopped over to the somewhat similar ITT/Cannon F2UEE switch.

And Mouser don't appear to stock the pots I'd proposed. The 25mm Alphas don't come in a PCB mount audio taper, or a dual-gang linear. So I switched to the similar but smaller 16mm Alphas for the auxes, and similar foorprint for the dual-gang panpot. Any pointers for a PCB mount dual-gang linear? I can't find a good drawing of the footprint of ANY of the Alpha pots on the Mouser website...

The 16mm pots will probably work out better, as the panel layout with the 25mm pots was really tight. As it is, we're pushing the panel real estate. I'm figuring that of the 5.25" panel height, 1/4" at the top and bottom would be used for mounting the thing, leaving 4.75". If I cram everything together, I could probably add one more pushbutton in the space provided...a subgroup/mix assign, maybe?

In the 16mm Alphas, Mouser have a pot with a DPDT push/pull switch...but only in linear taper. They'd be cool for the auxes, but marry the whole thing to a more esoteric part.

For the current mock-up, I used a 20-pin ribbon connector...again, not cast in stone, yet. The card-edge connector will take a little more tweaking, and will be harder to get fabbed. What was the root cause of the Mackie ribbon failures? I thought it had something to do with parts from different vendors being made of slightly different alloys, causing corrosion on the contacts. I've got an old Soundcraft board that's full of ribbon cable, and I've never had any problems with the ribbons in it. I like the concept of being able to build the harness for the thing in a matter of minutes, rather than hours.

At the monent, the layout is 4.75" by 3.5." But that's without any caps on the thing...Enthalpy, what's your preferred DC blocking scheme? Again, in the interest of flexibility, I could try to do a multi-component footprint...maybe to support lytics with film bypass, or a single larger film cap? Exactly how large?

As for metalwork, Blacet and Paia offer kits that are ready to go, right down to individual "strips." If you need fancier stuff, you could use one of those frames, with some FPE work for the channels. Anyone got a CAD/CAM turret punch handy?

 

[enthalpystudios]

the ribbon cable thing is probably a mackie thing.... just stuck with me. i really dont have a problem going with ribbons. this is really cool though, i think i like the simplicity of a ribbon. will also make metalworking a lot easier, wont really have to worry about exact depth so much as just getting clearance.

as far as dc blocking, I don't honestly have enough experience to have a real preference ;] i think it would be cool to use some nice film caps however.... i was looking around (unsuccessfully, for the most part) for a circuit example of something with monolithic or doa's that uses all film caps to get a better idea, but maybe some electrolytics are just necessary in the equation.... i'm not the one to ask though ;]

i'd just say keep it axial, but thats probably not such a hot tip ;]

hey for the pan pot, and others, what about these guys:

Panpot:
http://www.mouser.com/search/refine.aspx?Ntt=313-1250F-10K%09

Level/Aux:
http://www.mouser.com/search/refine.aspx?Ntt=313-1220F-10K%09

And in fact, I"ll have to double check, well let me do that now.... ok the ones i have are solder lug, same series however. i think they'd work fine... smaller, of course, but that shouldn't be a huge problem.. probably a benefit really.

only thing that stinks about those is that the flat section of the D shaft, if mounted the way i'm picturing it, would be -90degrees from where we'd like it to be, which would involve some set-screw trickery.

but alpha has a footprint if it helps:
http://www.taiwanalpha.com.tw/chinese/p_e_48.htm

i think they can order them with the normal alpha round shaft however.

maybe there's also another good option here.....

thanks for working on the pcb layout..... and honestly i wouldn't worry about card edge connectors. i'd stick with the ribbons. makee's steenk ;]

billy

 

[falken]

are you referring to the frac rak? that would be a great format. how would one go about cutting a slot for the fader? is this easily done?

 

[enthalpystudios]

thats a good question... i dunno, i'm going to do pots and knobs myself. i always thought faders in a rack were awkward. and takes too much space.

in fact, depending on how things work out, i'm most likely going to just put it all on one front panel... no units or modules really

i think i'll probably use it in a live rack to start with, honestly... i'd like to submix everything ahead of time, have good compression on the vocals, which quite frankly i need, and give the sound man (or lack thereof ;] a straight stereo mix.

its not typical rock band stuff, although i wish i had some people that were really into it to play with, so it will work best.....

i played in a band that had lots of stuff going on and for a lot of live guys at small-medium venues it was too much

so......... way too OT

i think a problem that may just need addressing is that of input. tranny? chip? that new THAT chip?

if tranny, onboard, offboard?

thats an important step i say....

oh wait.... addressed, it seems.....

I'm going to fiddle with the input buffer a bit, to see if I can rig it so that by stuffing a few different components, it can be the simple follower from Dave's schem, or a differential receiver.

sounds sweet

 

[analag]

The more carried away we get with the original concept that Dave presented, the less likely we are to ever stop talking about what we're gonna do to it and just build the damn thing.
In other words...stop gassing up the thread and start you soldering irons.

analag

 

[enthalpystudios]

i agree ;]

personally ijust need to pick up some dual pots in something other than 1M and 100K so i can bread a couple up.

dave did mention tx input on the schematic, but i'd be fine with a differential receiver. either way I say.

 

[kvintus]

amen to that, analag.

/A

 

[BradAvenson]

http://www.thatcorp.com/1200desc.html

Might be a good input stage. Easy and should work really well.

 

[analag]

http://www.edcorusa.com/transformers/wsm/wsm600-600.htm
http://www.edcorusa.com/transformers/wsm/wsm10k-10k.htm

Choose your weapon...these things sound great for line level apps. And M6 iron saturation curve is ten times more pleasing to the ear than nickel.
Like tube vs transistor clipping.

Fix these up to some good old 5532 chips and you have yourself a sound/ frontend that is normally reserved for the money boys. Or you could go for the That chip.

analag

 

[enthalpystudios]

Brad- thats exactly the THAT i was thinking of ;]

however, i think analag is right about using the edcors... i dont have a price on the that1200's, but i doubt it can beat the price of the edcors, and wont have a good number of it's advantages.

i've been thinking for quite a while about how i'd physically organize a mixer full of edcors.

i bet a mixer full of edcors would be sweet actually. the edcors i have used sound just fine to me ;]

actually, the U-channel frame deal on the wsm comes off easily with some needle n ose. could be mounted 4804 syle. clearance is the only issue....

i measure 1.37 inches above a pc board with tabs sticking up, with the lugs bent in, it would be a hair over an inch off the board. keep the modules around 1.5-1.75 wide, and have no probs. pc mounted wsm10/10 channel and aux return ins, mixer outs with 600/600s on the mix amp and aux master boards....

i'd also say that inserts are unnecessary. maybe just an unbalaned send/return point that can be tapped if people feel like it.

i think that would be excellent. i think the important options are the fader/pot headers for level, and maybe the DOA thing could be cool if people wanna f- with it. but a dc offset trim pot might be a good idea....

wish i knew eagle... i really need to start learning it just to start.

 

[edcorusa]

EDCOR does have a 1/2"x1/2" stack PCB mounted bobbin that could be wound the same as the 10k/10k. The WSM is a 1/2"x5/8".

 

[analag]

What now....everybody suddenly went silent.

analag

 

[enthalpystudios]

;]

big weekend

finishing 312/melcor amps and working on a cool audix 810 pre/eq module for a friend.

that pc mount edcor could be awesome... am i right in thinking that 1/4 watt is around +24dbm? or is it dbv... bah.. dont have a reference or a calc on hand, but if they have the headroom i'd be cool with trying those out for sure. and at that point, it would probably be easy enough to have a good ol groupdiy style multi transformer pads type of area to fit standard lineup.

 

[analag]

This is would be a really cool project, I was going back and forth about something like this with member on the back channel but that seemed to....
If I were to build it I would probably go Edcor into 5532 into tube summing amp out though transformer. Warm colored circuitry to put some hair on the digitized tracks. Emphasis would be placed on keeping it simple.

analag

 

[the scum]

I'm still here. I took the weekend off.

I've wasted a bunch of time trying to figure out a way to rig the DOA/Monolithic footprint in a way that Eagle won't complain about too much. I think I finally got it.

I've also been pondering a whole stack of features, and mulling over Steve Dove's articles. The simplicity thing keeps coming back to me.

My need for a line mixer is as for tape returns, doing cue mixes and confidence monitoring. I'd like my inputs to be transformerless, electronically balanced.

So I've done the input buffer so it can be stuffed for either configuration. For a transformer input, there's a load resistor and zobel network. The input and feedback resistors on the buffer opamp would be shorted, and the other two not stuffed. Or for electronically balanced input, leave the load/zobel out, add 4 resistors for a diff-amp.

The fader can hang either off the input, or the output of the buffer, selected by a jumper.

I keep pondering subgroups. For them to be useful to me, they'd need inserts, so the driver and buffer count starts to skyrocket. I'll file that away for mixer MkII.

I'm also going to prototype a pair of channels, and at least one summing amp. I'm still contemplating some of the AC coupling requirements, and a prototype will help me solve all that. It'll also let me doublecheck the component footprints.

 

[enthalpystudios]

cool man ;]

i dont really have a need for onboard trafo. I'll probably just use wsm's offboard.

have you checked out how fabio does the doa/monolithic footprints at his website?

puts the monolithic right in the center with a jumper for one of the rails i believe.....

5532 would probably be just fine and all, but it would be kind of cool to be able to experiment... honestly i couldnt think of a better way to test out different amps...... which is nice, because i'd like to try my 8x2 loaded with a different type doa throughout, and see what i like best in doa's. what works in a system for me.

thanks for doing the layout on this one scum. i have some melcor and 2520 doa's that i can experiment with on a proto if you want to share a 'running' layout with me... i'll either etch it or just go 'perfboard mechanic' (analag style ;]

but i can try things out over here and let you know what i find.

i have to make some orders to my 'usual suspect' companies, but i'm in no hurry and can hold off until there's something to do

also, any chance of having pads for a volume pot rather than a fader? i'd like to try and keep mine compact for using live or in the studio.

thanks again dude!

billy

 

[capnspoony]

analag,

I'm interested to know what you have in mind in terms of a tube summing amp?

If i could suggest something myself, I would .. but I'm not smart enough yet

I like the idea of keeping this project as simple as possible. I'd love to build an 8 channel mixer with simple panning / volume / one aux send / that sums to a master stereo ..

I just got a batch of four pair edcor 600 600 and 10k 10k .. we'll see how this thing sounds

-richie

 

[falken]

an easy way to do subroups with inserts (subs, auxes, mains, whatever) would be to have one strip containing all of the sends (8 or so volume pots down a strip) and then to have a separate input channel for each return (identical to normal input channel). What this does is (a) allow the returns to be routed into any other bus, (b) allow any pair of returns to act as master, (c) allow you to use "inserts" on your auxes, subs, and masters, (d) allow you to use patch cables in back of the mixer to "hard wire" your subs back to the group fader, and (e) have extra channels on mixdown if needed and subs are not needed. a flexible setup. sorry if my communication is bad...its not easy to describe abstract ideas!!! Also for this to work each channel would also need a post-fader send. I've already got the 100mm faders....just working on the best way to actually physically build the enclosure and the channel strips.

since everyone is mentioning their own "wouldn't it be awesome" feature; mine would be a selector switch on each channel that would select between two separate panners. a full range and a 5-position. that way you could get the "wide image" by using the 5-position switch, but have full access to all panning positions and 1960's panning freakouts if needed. But since we are going to do this as simple as possible, I would settle for just the 5-position.