ultimate diy mixdown gear project?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

tripledot

Member
Joined
Oct 11, 2005
Messages
11
Location
falmouth, cornwall, uk
having messed around with cheap soundcards, bad quality a/d converters and software mixing, i've come to the conclusion that i hate it.

i don't want to start an analogue vs digital debate, so i'd appreciate some experienced opinions on my grand idea:

using a motu 24io, i plan to sequence/record all my hardware synths etc with *tiny* latency back into my pc. i like rudimentary mixing during composition to help me "visualise" how a track is going to sound when completed. so, i will record clean channels of each input and buss them to plugins for this purpose.

for final mixdowns, however, i want the tump factor.

i'm planning on building 24(!) of gyraf's calrec eq clones, and a couple of compressors to use as sends (new york school of mixing)

this seems like my most cost-effective, rewarding and ultimate-sounding solution. any thoughts? am i being plain stupid and are there other DIY alternatives that will give me a full 24 channels of sweet eq and maybe more?


cheers, tripledot.
 
...common sense would say build one mono or stereo unit first and see if you like it. Whilst it is easier and quicker to build several units at once, 24 is taking it a bit far I feel. Have you worked out how much this will cost? And more difficult to estimate...how long it will take? If your setup allows for you to have several types of EQ, then surely that is something to take advantage of, rather that go for only one type?

Bjorn
 
...and welcome, by the way...the window where you type your posts doesn't show the number of posts someone else has made so didn't see your "1". The Calrec does have many happy customers...alas I have yet to build one...

Bjorn
 
ok thanks for the welcome and quick replies guys!

yes, common sense would suggest that i just buy a desk, but i'd rather have the satisfaction of building it all myself. rather than build a complete DIY mixing desk i like the idea of a modular system..

the workload involved is huge, i know, but it's something i'd be quite happy to spend a couple of years working on - i'm patient and have much to learn about electronics, and i'm looking forward to the entire process of pain, suffering and (hopefully) reward!

i've heard the polar eq and i like it *a lot*, very transparent but musical. it'd be nice to have a whole bank of them for a predictable mixdown every time, but i take the point about having several different eqs..

i might try and build a couple of the pultec clones for a bit of variety.

if i use the highest grade connectors and short patch leads, using all-balanced connections, will i still suffer s/n loss compared to an integrated signal chain of a desk's complete channel strip?

also, any suggestions for a channel compressor design that will give me a nice range from gentle vocal compression right the way through to rock-hard limiting? low-noise hard limiting is a toughie i know, but there must be something out there that works!?
 
well, for a compressor that'll do what you want, i would suggest integrating the rnc into your setup. it'll set you back just 200 a piece.

but if you want to go all diy...a gssl?
 
Here I go souding discouraging again, but here's yet another first-time-post with grand ambitions, indicating that the poster has thought this through down to the details and with a set working practice in mind. (New York School of Mixing)

Some of this (apparently established) methodology is new to me, so I may not be the best person to ask (has anyone ever heard of compressors being used as sends before) (...and what the heck has latency got to do with this anyway? -It's an issue in certain circumstances, but why mention it here?) I think that this is a bad idea, and that the logic is fundamentally flawed... Or perhaps it's just a troll, who knows?

Myself I'd counsel that ANYONE doing this much DIY should be aware that there's a limit to the 'cost-effectiveness' of DIY'ing 24 channels of anything bearing in mind that the resale value is -in most cases- buttons compared to the real thing. You'll never sell a plainly non-genuine DIY'd piece of classic gear for anything like the cost of the real thing, but they'll cost a bundle to make 24 calrecs.. just do the math.

My advice is to buy something. If you're not already a DIY-er, you have no idea of what you're getting into.

Keith
 
I started with exactly the same idea...
But after a year I have finally come up with what I am planning to do - that is
a. feasible
b. finished before hell freezes over

The thinking about the metalwork and how they are all going to fit in is the big problem
Are you going to use 19 inch 1RU for the Calrecs... are you going to mount them vertically or horizontally? if they are vertical then that's an awfully big channel strip ...
So... you buy a smaller rack unit - maybe a half width rack unit - and then spend the next two months trying to get somebody to sell you one... only for Hammodn to say they ain't exporting them to the UK anymore.. so you get them custom built..
Now comes the crunch - the calrec is a 2 PCB operation.. now with all those 2 gang switches and trying to mount the PCBs next to each other you ain't got much room in there.. in fact you mount the PCBs vertically - and now you have a problem because the size of the electrolytic caps makes them impossible to fit inside the half width single rack unit...
And then what are you going to do with the relay boards for the illuminated push buttons - how are they going to fit?
Why illuminated pushbuttons... Think about it 24 channels of the same thing - how the hell are you going to know which has the HPF etc.. switched on - by a plastic capped switch being 3mm deeper than another one... nope...

You may have gathered I have been down this route...
And then their is the PSU and grounding to worry about...

I am coming around to the idea of still getting rid of my mixing desks (I have 3) - using a passive summing bo, a few mic pre amps, a few compressors, a few EQs (printing different EQ versions to tape... like back in my 4 track days...)
 
ok i have little diy experience, but don't knock my knowledge of studio practice just because i sound like a newbie.

new york school goes as follows..

mix with customary 3dBs of headroom on each channel minimum, get everything clean, clean, CLEAN, then for added punch just send out key elements (kick, subs, anything you want squished with these etc) to a nice hard-knee compressor. in the analogue domain there is no audible phase cancellation (i was surprised too!) and it helps bring parts of your mix to life without having to worry about setting different attacks on all your inserts so that they sit comfortably.

sure, use inserts where necessary to control horrible peaks, but the temptation to mix with everything flattened by limiters is ruining both the dynamic range and the individual character of much of today's recordings.

back to the diy problem..


unless anybody has any problems with NOT pcb-mounting the switches and pots, i don't see what the problem is. space-wise, racking 16-24 channels of channel strip pots won't be a problem. i can build cabinets happily. the pcbs will all be mounted vertically in the mixer "base" with air-cooling from below. wiring from gyraf's PCBs to the pots will be heavyweight shielded. easy.

then i just wheel my whole cabinet over to my main rack and connect up.

24 is overkill, agreed, but i don't think i can get away with fewer channels. and for the same money in sound control or whatever, i would struggle to get 8 channels of real top-notch sound.

i can do it for under 500 quid. (granted i have some parts already). sound stupid now? :D
 
er, latency has nothing to do with this at all.

sorry to be confusing. i only threw that in because when i have my motu i will be able to monitor 24 inputs and use plugins without worrying about latencies approaching 2 seconds even during recording and arrangement.

i'm fed up with latency already. hence my bitching about software mixing and one of the reasons i want to go analogue.

sorry.
 
No probs mate..
BTW - can you give us your contact for 19 inch rack enclosures then...
24 x roughly £30 each = £720 just for the cases
You might want to have a separate Big PSU rather than build a PSU inside each unit.
 
[quote author="tripledot"]i can do it for under 500 quid. (granted i have some parts already). sound stupid now? :D[/quote]

Yep.

24 Calrecs for under £500? -That's less than £20.84 each. At todays exchange rate that's under $37 each.

Better wake up. If you think you can build this for that price, you are seriously delusional.

Keith
 
aha now when i said 500 quid i was talking about what my rapid electronics basket was totalling earlier...

i was ordering parts i need to build 8 calrecs!

hehehe so maybe i'll work my way up to 24


as regards rack enclosures, i wasn't going to rack them horizontally, but sit the pots etc. in a solid alumininium sheet and put that in an existing rack-railed wooden cabinet.

i don't know of anywhere that manufactures cheap rack enclosures right now!

sorry for the general grottiness in my posts, just i got really excited about what i thought i'd thought through very thoroughly.. now i've decided to do it in banks of 8 at a time... this way i might be able to build the whole lot into a (very very deep) 8u rack....

many options... order parts, then i'm committed to the build. i'll post up some photos when i'm really screwed later!

:)

cheers
 
I am foolish enough to want to someday own a large format analog console. I think there is misconceptions on cost but nothing wrong with wanting to bite off more then one can handle we test our limits that way. Keith you are correct. I feel it would be better to buy and mod an existing console or get a frame and build from the ground up like Amorris. IMO buy a sony 3000 analog board can get them cheap and completely mod able. :thumb:
 
triple dot
if you are doing that many calrec - best talk to Margaret at Omeg Sales to buy your pots direct... (and make sure you buy the proper ones - I didn;t on one order ... groan..)
Then you can get plastic conductive ones and not carbon pots.. (Electrovalue is the only place in the UK to get the rev dual log ones... and they are carbon)
And why not talk to Sifam direct if you are ordering that many - then you can get decent colours of knobs and caps and not what Canford stocks...
What sort of electrolytic capacitors are you putting in there? (BTW - the Stargets you get at RS compments fit in their quite nicely)

I see what you are doing with the enclosures...
Have a back plate/a front plate and a sheet of aluminium in between
So you are going to have to build a central PSU to feed all of these units... and feed it to each unit in turn...
 
triple dot,

Perhaps I'm misunderstanding what you're thinking but how are you going to sum these channels? That's not clear to me from what you've described.
 
panasonics for the caps

i thought i'd go the whole hog with resistors too - 0.1% tolerances all round

thanks for the advice on pots, i've been panicking about that too! half my build budget is going on pots and i need to know i can find top quality!

i take the point about modding an existing console keith, but i just don't have the knowledge to mess around with an existing design. i need to learn from scratch, and the gyraf DIYs have so much support and so many documented builds i feel i can learn the most from following these for starters.

but one day i'd like to get inside some of my monosynths... mmmmmm :)

i'm gonna turn my sh-2000 into a modular one of these days!



- chris
 
channel summing, oh yes..

that's something to think about later.. for now i will use the motu inputs and set levels in software (grrrr) but if i can bring myself to sell my scirocco i might invest in one of these:

http://www.cranesong.com/spider.html


drooling already
 
no way man go for the Shadowhills gamma delux. :green:
Not to pimp but I have to on this one :twisted: Stereo mic pre w/built in 32 channel passive summing. Selectable outputs between 2 custom trafo and our trafoless a discrete output plus custom input trafo's on the pres. Gotta love it. fook me I better stop this right now.
 
Ah. Eight is a little closer, but you might still be surprised by the total costs.

Even then, all of these will be useless for analog mxing without summing. You should consider level control and panning/matrixing also... otherwise you're stuck with something that isn't much use, specifically stand-alone.

Keith
 
I would make two or four channels of each first just to learn and evaluate the project and everything that goes into the process...

Secondly, you might want to consider trying out different components (really nice stuff... Wima, Nichicon Gold, Elna Silk, and a bunch of others)
The reason is if your going to put so much time into a project on this scale, you might as well put the best parts in the units.. if not just buy some manufactured gear.

Are you going to have external PSU to power 24 calrecs? if so... can you build it yourself? there's no DIY Pcbs for making your own external PSU...

if you add a couple GSSLs, VU meters, and preAmps into your project are you going to need a more voltage, those things probably have different voltage requirements?

Then you have to think about the metal work... Its pricey to have someone else do it right.... If you can get the design PERFECT. Or maybe you have a metal shop in your closet to make everything yourself....

The reason I'm saying this is because I had the same ambitions when I found this place... and I had a lot of gear lust... and didn't realize all the work and the expensive learning curve involved.

After building two Gssls, two Calrecs, One Pultec, One La2a, One G9, one 1176, Green Pre, 4 Chae Neve-ish 1272 type Pre... I am Very happy! and glad I didn't buy 24 pcbs of one thing.... expensive!

I'm not trying to discourage you but I agree with sslTech..

At least do a 2 or 4 channel prototype first.

Or better yet maybe buy an old desk and restore it, you'll learn a lot.
 
Back
Top