500 series faders...

[EnSkorSang]

Ok, continuing the investigation of 500 series. I've been thinking that something like this would be great:
500 series fader.png
Some fader fun, but in 500 series format - so you can have as many as you want or e.g. chain an eq into the fader or whatever. Then all the outs would go to your little summing box (ha ha) or back into digital land or whatever.
Something like this exist?
A problem with it would be the pan control...the way I understand it is that 500 series racks are mono in/out for each slot? This would require mono in/stereo out...
ideas?

[paul999]

There have been a couple of people try to get this sort of thing going but it always ends up being pretty darned expensive. If you think about $425/6 unit lunch box plus $125 for a cheap fader/ pan system. You'd be at $1000+ in a hurry and all you got was faders. Even the $425 is pretty expensive for 6 faders. Not to be a downer;-) but it looks like I am.LOL

[EnSkorSang]

Eeech, they are making a killing on those racks, no? A PCB, 6 connectors, 12 neutriks, power supply and a bit of metal...

DIY! $250 for a kit...

Nah, but really, Neve seem to manage to sell their fader pack for $2K and that really is only faders and only compatible with their summing mixer.

It's not a way to perhaps turn someone who has a couple of racks already full of 500 series modules into someone who has a full mixing system (minus the mixer :P) for just 1 extra full rack?

I dont know...It's something I would be excited about even having to pay $150 or whatever a module...Much cheaper than getting a tonelux one, (or a tree audio console :S). Of course, it wouldnt have any grouping/routing capabilities.
Is it me or did API miss a trick with the 500 series - i.e doing what tonelux are doing and having a much bigger number of pins on the connector to allow for routing options etc....of course, tonelux prices are just ridonkulous ($599 for their v-rack :S and you're looking at $1619 for an f-rack with just 6 faders - and thats only faders)...

Anyways...maybe if you could make the fader a bit more functional that it is on my pic it might be more worth it. Although Im not sure what you could put...a phase switch. Sends, but then the send/return would have to go on the front plate as the api rack wouldnt have any capabilities for that. With a knob to control the send level aswell, there probably isnt enough real estate on the front panel..

So then it boils down to creating a new rack, but then the point of this is lost - which is to take advantage of all the different 500 series modules out there...

Unless....

[garageband]

Your unlabeled button should be Phase Invert.

So, you are proposing a purpose-built console whose architecture and routing is very friendly to the myriad of 500-series modules on the market? Fantastic!

[EnSkorSang]

Haha...hmmm....or I was just brain farting all over RR. Consoles are impossibly difficult to make, over my head certainly. But, I do think using 500 series style things would make it a LOT easier. Your preamps and EQ's are sorted, so your channel strip suddenly becomes quite simple (like the one above).

What holds it back is the actual rack. At least the last rack in the chain would need to be a custom job to include some routing. To keep the 15 pin 500 series format in that rack, the routing select would need to be on the rack itself rather than the modules (there just arent enough pins to put more than 1 out on the module).

Or you could do it with a connector that has say...30 pins and keep the first 15 the same as 500 series modules so it is compatible. The last 15 pins would then purely be for buses. Then you would make a bus select module (ala tonelux) which would take advantage of the extra pins...

So, we are getting something more towards the Tonelux system (rather than e.g. the 1608 or Tree Audio) that is fully modular, but crucially it is 500 series compatible. You dont want to be limited to 1 flavour of module. And hopefully cheaper than the tonelux.

Lets think about the cost. You would need at least 1 of these custom racks with bus tracks. How many spaces in the rack? multiples of 8 would be normal (it would be the same size as the lunchbox? - if the power supply is outside the rack, giving it those 2 extra slots?) Any module wanting to take advantage of these tracks would of course need the extra pins and the summing resistors in it.

So if you wanted to start your system with a simple line mixer (no pre or eq or routing yet). It might be 1 of these racks and 8 of the modules above... Transformer balanced output and then go wild with some good IC circuitry could yield a great sounding module for not much IMO. Discrete circuitry would bump up the price...
So if a company was selling this it would be, what, $500 for the rack and then something like $150 for each module? Which puts you at $1700 for the simple line mixer. Thats a lot, but it's still cheaper than what is out there...with the advantage of say, starting with only 4 modules if you wish and building up. The real problem is the rack cost....

Ok, now I really am farting all over the place....

[EnSkorSang]

So here is the idea. A 500 series module uses a 15 pin connector a la:
306-015-520-102 EDAC Standard Card Edge Connectors

You could have a rack that uses a 36 pin connector (with same contact spacing) a la:
357-036-520-202 EDAC Standard Card Edge Connectors

Keep the first 15 pins in the same format so that 500 series modules can still be plugged in.

This leaves 21 pins for fun. That could be setup as 5 stereo buses (4 groups and master?) and 1 send. You could always have the sends on the front panel of the e.g. bus assign module or group modules....

You could have just 1 or 2 of these to turn your existing racks into a console...

Back to the fader channel. It would be better if that worked in a standard rack. So remove the pan and that can go on the bus select module...

Then maybe you can put a single aux send on the module instead. Send/return jacks and rotary wet/dry fader....

[EnSkorSang]

Well, I ploughed on with this :P, with some help from electrical dudes at uni.
First, I started thinking about the basic building block - the input channel. So what did I want it to do? :
- Receive line-level signals
- Volume and pan
- mute, phase, solo?
- sends and inserts on each channel?
- balanced output.

We've come up with something like this:


First thing is, this part isn't 500 series. It is simply not necessary and cumbersome for this module. The PCB backplanes on 500 series racks are expensive and unnecessary for this part. It's 4 RU high - 7", 500 series width - 1.75". 100mm fader, 15mm panpot, mute, phase, 2 aux sends and insert send/return.
There are some limitations or things to consider..
The actual mixer then would look something like this:


It's something between Tree Audio and tonelux with some important differences.

My idea would be that the jackfield/connections/whatever, contains only inputs/outputs to the channels and the aux sends. That way, standard 500 series racks could just be 'attached' on top of it. The outputs from the racks would just be linked to the channel inputs in whatever manner is seen fit.

No backplanes on the channel modules/group assign modules etc. Firstly, the channel modules dont need them as they arent connected. Simply providing L&R out which would go to the group assign modules. These would be connected by ribbon cables, which would finally go to the actual group modules.

Group assign modules are separate modules, 2RU high. separate for ultimate flexibility. Cause I dont know how many groups is enough. If I want more, then instead of having to change the whole channel module, backplane etc. I simply swap in a group assign module with more groups on it and a wider ribbon cable I've done this pic with 4 groups.

Then the group modules. Currently standing at 6RU in length. Pretty much the same as the channel module, except of course in stereo (so number of knobs and holes for inserts/sends is doubled).

Master section. Haven't decided really what needs to be on here yet. Ideas?
I've put a 500 series bucket above each part, but really it could just be as many as wanted. Kind of like the Lilo, except set up so that the 500 series can sit on top of it in a frame like a full blown console :P
I figured, the spaces above the channels could be filled with EQ, and above the group channels/master it could be preamps or processors for the groups/master. I dont have much need for tonnes of preamps. Plus they are generally something you might already have a lot of before getting a mixer, so you would just patch them and then all the preamps on the mixer are just sitting redundant....

So heres somethings:

- No solo on the channels. Solo switches are actually kind of hard to implement. Mute just turns that channel off...solo has to turn everything else off...
- Not sure if TRS inserts for effects on each channel should be on the front panel.
- not sure how to arrange the patchbay (for auxes and I/O). They are generally on the back...but thats annoying. Of course, the 500 racks do have them on the back, so it is perhaps the only option.
- Those group assign modules sure look sparse. With 6 groups it could be better...but then I might be running out of space to put group channels. The middle section could be made bigger...but then it's getting pretty big. It's already at 36inches (which is close to 1m) wide for 16 channels.
- I would've like to get the group channels on just 3RU to get another 500 series rack in, but there just isnt enough space for everything. It would be nice to get 6 groups in there without expanding but without eating to much into the master section real estate. hmmm.

- Some functions have been omitted to keep the simplicity high and cost down. No PFL. I dont get that. No input trim, no aux mutes or pans. No choice between inputs (e.g tape or line). Maybe there should be something in that isnt.
I'm not sure about the phase switch. I dont often switch the phase of things, and if anything is out of phase is usually a gentle nudge one way or the other (in the DAW) that is needed, rather than a complete reversal. Perhaps it's an unnecessary complication?

The electronics dudes at uni helped to put together a schematic for the input channel. Basically, stealing each block from 'Small Signal Audio' and putting it together like lego. IC's throughout. I dont understand why Tree audio and others have treated their designs with discrete opamps, hand wound transformers etc etc when all it is is a line receiver, volume controls and then just switching what path it's gonna take.
IMO, these input channels should be totally transparent, with high headroom and low noise. Modern IC designs are great for that. Plus it keeps the cost way down which is a main goal.
Transformers will possibly be put on the inputs to keep things isolated and safe...that might add mojo.
But the mojo would come from whatever 500 series is put in.
Can always make the make up gain discrete or full of valves...

Anyway, using 5534's and THAT 1240's, we found an input channel could be quite easily be made for under £100. More like £100 with a transformer.
The most expensive part of the group assign module would be the metal work for the front face. But I could do it all plastic...
The most expensive part of the whole thing would probably be some kind of frame to put it in.
Anyway, for the first 8 channels, plus groups and master, I'm thinking £1500 perhaps a bit more to build. And thats using whatever I can from uni (i.e not factoring any equipment that might need to be used, students to help designing - no wages )

The picture is plan view, but in reality the channel/group/assign modules would all be on the same, slight incline and the racks would be much more toward upright, as is normal.

Seems I'm turning this thread into my brain fart dumping ground

[garageband]

Wow, sounds like you might actually go through with this madness. In that event, I should try to help. I'm grab random ideas and say what's on my mind.

Plus they are generally something you might already have a lot of before getting a mixer, so you would just patch them and then all the preamps on the mixer are just sitting redundant....

Isn't this mixer where said preamps are supposed to live? You buy a mess of these things, to your liking, and then have an infrastructure to support them? Then you just have a Mic/Line switch to wire the preamp out.

Simply providing L&R out which would go to the group assign modules. These would be connected by ribbon cables, which would finally go to the actual group modules.

This seems sticky. Group modules can only receive L/R-assigned signal? Having more than one send/return point here seems of limited value.

no aux mutes or pans.

Un-pannable auxes is a mistake. That's mixing flexibility you need. For me that's be a deal-breaker in light of the otherwise limited busing architecture.

Transformers will possibly be put on the inputs to keep things isolated and safe...that might add mojo.
But the mojo would come from whatever 500 series is put in.

This is a very expensive "add". I wouldn't be in a hurry to do this.

No solo on the channels. Solo switches are actually kind of hard to implement.

That's no foolin'. Good call.

The electronics dudes at uni helped to put together a schematic for the input channel. Basically, stealing each block from 'Small Signal Audio' and putting it together like lego. IC's throughout. I dont understand why Tree audio and others have treated their designs with discrete opamps, hand wound transformers etc etc when all it is is a line receiver, volume controls and then just switching what path it's gonna take.

Sounds like you're in good hands. They might have some interesting circuitry embellishments that hadn't occurred to you, something simple, yet different, yet effective. Along the way, I'd sneakily might them improvise an idea of two like that.

Anyway, using 5534's and THAT 1240's, we found an input channel could be quite easily be made for under £100.

THAT reminds me (electronic nerd pun), see if they couldn't bake in a THAT 2181 for some rudimentary dynamics control on each channel. Threshold only control and fixed ratio at 3:1. Your own little SSL moment.

he most expensive part of the group assign module would be the metal work for the front face. But I could do it all plastic...

You're resourceful enough to get it handled with metal. Plastic? The shame!

IMO, these input channels should be totally transparent, with high headroom and low noise. Modern IC designs are great for that. Plus it keeps the cost way down which is a main goal.

Yes, you want this. Also, grounding, grounding, grounding.

Meter bridge? Yes, please. Might I recommend real VUs? Totally different yet more useful than LED peak meters.

Lastly, they need to design the mother of all 16V power supplies. Should need a lift truck to move it about.

[EnSkorSang]

Wow, sounds like you might actually go through with this madness. In that event, I should try to help. I'm grab random ideas and say what's on my mind.
Isn't this mixer where said preamps are supposed to live? You buy a mess of these things, to your liking, and then have an infrastructure to support them? Then you just have a Mic/Line switch to wire the preamp out.
This seems sticky. Group modules can only receive L/R-assigned signal? Having more than one send/return point here seems of limited value.
Un-pannable auxes is a mistake. That's mixing flexibility you need. For me that's be a deal-breaker in light of the otherwise limited busing architecture.
This is a very expensive "add". I wouldn't be in a hurry to do this.
That's no foolin'. Good call.
Sounds like you're in good hands. They might have some interesting circuitry embellishments that hadn't occurred to you, something simple, yet different, yet effective. Along the way, I'd sneakily might them improvise an idea of two like that.
THAT reminds me (electronic nerd pun), see if they couldn't bake in a THAT 2181 for some rudimentary dynamics control on each channel. Threshold only control and fixed ratio at 3:1. Your own little SSL moment.
You're resourceful enough to get it handled with metal. Plastic? The shame!
Yes, you want this. Also, grounding, grounding, grounding.

Meter bridge? Yes, please. Might I recommend real VUs? Totally different yet more useful than LED peak meters.

Lastly, they need to design the mother of all 16V power supplies. Should need a lift truck to move it about.

Well, you could fill the 3rd bucket with preamps. It would be good if it was possible to squeeze in another bucket, but I dont see how. You could put another row of buckets along the top, but then it's getting stupidly tall and more like the tree audio. Of course, the beauty would be that you can easily swap in whatever 500 series you want. So you could have a different setup for recording and mixing....

Dont group modules normally just receive a stereo signal?


My idea would be that the aux sends on the channel would be before pan - mono, and the aux sends on the groups would just follow the pan. But being able to pan the aux sends on the channels would be a good idea....
Reduce the aux sends on the group to 1 stereo?
1 stereo insert aswell?
This leaves a lot of empty real estate...could be where the possible dynamics goes

I have no idea where I would squeeze in the pan for the aux on the mono channels. Unless I reduce it to 1 send per channel.


Yes, transformers are expensive and not so necessary. Without them, the most expensive component would be a fader at around a mere £5-7, TRS connectors at ~£2, potentiomenters at ~£2-3. Knobs/fader caps ~£1-2. Illuminated switches ~£3-5
Single channel would be around £35-45 in components at my estimate. That would be the upper price. To get a front panel sorted from a proper place (not scrounging off uni) is around £10-15 depending on quantity. To get a piece of metal and drill myself? Around £5 for the metal and then probably another £10 for all the times I'll screw up per panel .

So a channel cost, I'm thinking no more than £60. Probably cheaper, but I'll keep it at that price to account for screwups and 'hidden' costs. (All the coke Im gonna need to solder this on a night...).

VU meters? Yes. Would be great. Cheap? hmmm... can pick up something like this for £1:
FM-136 | GES-ELECTRONICS
Looks kinda shitty. Or something like this for £8:
ANDERS ELECTRONICS|C360VU|METER, CIRCULAR, 0-200UA | Farnell United Kingdom
I would like those semi-circular ones, kinda like the ones on the tree audio console . But I dont know where to find them and they are probably uber expensive. It's not a project on which to get too fussy about looks. Perhaps a little fussy.
That reminds me. Front panel colour? Not black...

Plan of action; finalse the channel design and build it on a veroboard. Power it up (with batteries prob) and see if it works. Take it to uni and borrow the oscilloscope or something
If it is all ok, then I'll think properly about the power supply (should be huge, but should be quiet...would be good, even important, if it could also power 500 series racks) and the master section.
Then, I might just start the proper building with the VU meters...

[garageband]

if it could also power 500 series racks

I thought that was part of the plan. That seems like a "must" as far proper integration and implementation goes. I don't see why you couldn't put more effort than most console mfgrs put into making your PS quiet (which is none, I think) and get something very quiet. Have a look at the Apple Pro computer case for riffs to steal.

More later.

[EnSkorSang]

Yes, the power supply must power some racks aswell. Questions are:
1. Should there be seperate kettle leads for a number of racks - so external racks (like a lunchbox) can be included, rather than relying on internal power distribution
2. How many racks should it be able to power...

Some changes on the input channel. That chips are pretty damn expensive, so Im thinking either only use them on the group channels, or perhaps just for some dynamics. 5532/4's for the rest. Here's a block diag of where the input is now.

1 aux pre fader and 1 pre/post select. Maybe they should both be switchable. Balanced insert and mute. I'm thinking of some kind of buffered mute so it doesnt pop when engaged/disengaged.

Direct out that is selectable between pre/post fader. Maybe not necessary to have that switching option.

I did wonder whether to move the aux sends to the group assign module, to keep all the routing together and neat, but then it removes the pre/post option.
I wonder aswell whether it would be necessary to put output amps on the L/R out to the group assign module...

Schematics of the blocks to come and updated front panel? Anything I should/shouldnt include on this module?

Oh, here is an interesting circuit from THAT:
http://www.thatcorp.com/datashts/dn125.pdf

[EnSkorSang]

Continuing with the madness. Heres where I'm at. (prepare for geekiness):

Ok, I have a shitty schematic editor, but looking past that, this is pretty much the channel, minus a few problems:

1. The insert should be removed. I suddenly thought, on a channel that has no eq section, what is the point of an insert? Any inserts would just plug directly into the channel. Im also considering making every single input/output a TRS or xlr jack that goes out into the real world, so the modules could potentially be rigged up however the hell you want. Potential mine field though.

3. Mute switch is the wrong way round.

4. Those 2 1M resistors would be connected to -V

Changes from the original design are:
- phase switch gone. What a lot of circuitry for something you rarely use. This is something that will have to stay in the DAW...
- added pre/post select to one of the auxes.

Things to add:
- need a metering output between fader and panpot
- possibly a peak LED before fader?
- aux pan? This would ramp up the cost a bit..

So, the cost for components at the moment is £45 using 1% resistors, quite expensive potentiometers and illuminated switches. Its also without bulk buy discounts. It's possibly fairly expensive considering the faceplate, pcbs and knobs/fader cap have not been taken into account. I could probably save money on pots bringing the cost down to something more like £40. Then the total cost might be around £70-80. Again, pessimistic estimates. Still hitting target of a low cost, good quality 500 series mixer. It's approaching £2K for a 16 channel unloaded version. Add on a meter bridge, frame, power supply, more like £3K - 3.5K. Darn sight cheaper than tree audios ~16K, although less fully featured...

Of course, if I werent just building 1 for myself, owned a company and thus able to purchase crazy amounts of stock etc, then it might be more like £1-2K production.....why does nobody do this?

Anyho,
So the channel now looks like this:

[garageband]

Hard to go wrong with 5532s and 5534s.

Any inserts would just plug directly into the channel. Im also considering making every single input/output a TRS or xlr jack that goes out into the real world, so the modules could potentially be rigged up however the hell you want. Potential mine field though.

Seems like the only way to go. Build in more flexibility, not less. Kinda the point, it seems to me.

I still think the one gizmo (that you linked to - the one-knob deal) from the THAT people has to get shoehorned in, regardless. Trust me, you want this. You'll thank me later.

why does nobody do this?

I've wondered this earlier when 500 series modules started to become fairly ubiquitous about 5 years ago. The whole lunchbox thing just seemed like such a bare-bones approach and really not an improvement from rack gear, utility-wise.

- need a metering output between fader and panpot
- possibly a peak LED before fader?
- aux pan? This would ramp up the cost a bit..

Yes.
Yes. Definitely.
Yes. I thought we'd been over that.

I could probably save money on pots bringing the cost down to something more like £40.

This is not a good idea. Buy the best pots you can stand to pay for. This isn't something to scrimp on. The device on my left has about 800 pots. I'd be very unhappy if they were half-scratchy naff bits (that's how you guys talk, right?).

Direct out that is selectable between pre/post fader. Maybe not necessary to have that switching option.

Not necessary. Direct Out should be prefader.

[EnSkorSang]

You're thinking the THAT compressor on every channel? I was thinking only on the groups, but it is a possibility....before the pan of course...
Pre or post fader?

I'm going to look at doing a pan for the each aux. Not sure how I will squeeze in 2 extra knobs on the front plate...

The peak circuit will be easy as it can be straight out of the textbook and slips in just before the 10uF capacitor after the line amp:
p30a_f8.gif


The meter out has to come post-fade, pre-pan. I guess that can be fitted in fairly easily.

I'm having trouble fitting in more than 1 aux pan into the front plate. It would be 1 aux selectable pre/post and 1 aux pan-able. Hmmm...is this acceptable.
Whats the common use of aux sends?..Compression and reverb for me. I dont pan compression, but I do pan reverb. But this doesnt leave room for other effects, plus I might be providing some little compressor anyway.

But what I really dont want to do is use up real-estate in a way that forces me to uses crappy little behringer/mackie style knobs that are difficult to turn. I'd like no smaller than 11mm knobs...
Heres what it would look like as this configuration:


Putting the 1 knob squeezer in on every channel though, would either mean moving the mute somewhere (side of the fader or something, which is not that attractive) or going 5U long. But then, thats getting pretty long...

[EnSkorSang]

I've wondered this earlier when 500 series modules started to become fairly ubiquitous about 5 years ago. The whole lunchbox thing just seemed like such a bare-bones approach and really not an improvement from rack gear, utility-wise.

Yeah, in my mind, the main (only?) advantage of 500 series is their shape/size which lends itself to setting up your own system...except not many people have actually tackled making a 500 series mixer available. What is out there is crazy expensive. I want mojo to come from the 500 series modules, not the mixer part. I want that clean and transparent. Clean and transparent these days is pretty cheap to achieve. IC's are ideal.

Of course, I'm looking at this and thinking 'How the hell do I put this on a strip board?'. I'm also wondering how to tackle providing the power (the first, stripboard test will just be battery powered I guess) and how to get the grounding sorted. These things are easy just to plonk a schematic (you just say 'ground here), not so simple in reality...