Line mixer (active)

[solder_city]

the idea is already out there for you to try. why not wire something up yourself instead of constantly perstering other people to do the work for you? get yourself one of those plug-in protoboards and go to town. prototyping with ic op-amps is easy.

if you cant build something unless theres a ready made pcb with instructions then i guess youre just going to have to be patient.

 

[therecordingart]

I have been working on this slowly but surely. I don't have anywhere near the experience of the other folks, and surely I'd end up building something that they come up with.

You don't have to be rude, and I'm not trying to pester. I simply asked if there was an update.

I apologize almighty electrical god.

 

[therecordingart]

Also...is there a new PDF link? The current one is broken.

 

[lofi]

must admit, that did come accross as rather rude, some lead others follow, when you hang up th iron we will pick it up and take your place, untill then we will just follow in your shadow, if thats ok with you that is?

 

[NewYorkDave]

groupdiy.twin-x.com has been down all day. But no matter, I'm working on a new schematic with a couple of improvements to the circuit. I just need to find time to finish it up.

 

[therecordingart]

Awesome. I'm not really that good at reading schematics so I've been expirementing with caution and very slowly.

I really appreciate what you are doing.

 

[lofi]

ahem, anybody getting anyware, its just i live in awe of you guys, so want to see what you come up with, then that guy in isreal can go and build a fully proffessional version in about 8 mins and say ' here is something i have been working on this morning' igor i thinks his name, i ramble, i'm sorry, i will get my coat

 

[Brian Roth]

[quote author="NewYorkDave"]Here's something that could be useful for you guys who wanna mix out of the box using your outboard, etc. It's for when you need capabilities beyond those of a simple passive mixing box, but don't need a full-featured console.

NOTE: THIS IS NOT A READY-TO-BUILD PROJECT. THIS IS JUST THE OUTLINE OF THE CONCEPT. But I think it could be fleshed-out into a nice project.

PDF

[/quote]

I came into this "party" late in the game, and the URL posted above goes to just a generic page with a hodge-podge of stuff on it. Is there a more recent link?

Bri

 

[therecordingart]

Be careful! Don't ask how everyone is doing on this because you might get bitched at!

 

[strangeandbouncy]

Dear All,

OK - I am prepared to put my head in the lion's mouth . . . .

i am all ready to mock-up a channel and mix-amp, but. if no-one objects, I'd like to ask some questions. i do not do so to offend, or annoy, but because a simple 10 second answer will save me hours and hours of working with perf-board. i am NOT a lazy git, but a very busy man with far too many musical projects on the go, and the demands of a 9 week-old baby at home. I beg your indulgence . . . .

I am about to start with my metalwork, so i need to ascertain some things My needs are as follows. 16-24 channels, possibly expandable at a later date, 4 auxes, 2 switchable pre/post, very basic eq, an alternative stereo sub-mix bus. i DON't want to drill holes for 4 auxes if only 2 will work! I don't want to provide for a seperate sub-mix bus, if this too is not possible simply.

so,

firsty, will I have a problem driving 4 auxes post? should i buffer them with another ic first?( i intend to use DOA for the main signal path, but would not mind a tlo7x or 5534 on the auxes . . . oh, and a direct out per channel)

secondly, I intend to use 19" racks for the case, since it would be easy to expand if necessary. What I need to get is some advice on the connecting of busses from one case to another. Will decent cable in a D25 connector do? I guess i am not worried about not being balanced. Some of my favourite consoles are not balanced in and out. ie Helios, trident etc. am i going to run into trouble with the busses unbalanced on signal cable?

thirdly. how do the component values change with additional channels? do the 22k "bus" resistor values need to be increased?

fourthly, I was thinking about adding a second DOA on the input, before the channel fader, to provide a passive eq like the Op-Amp labs 425, or API 553. Then it struck me that perhaps i could use the one AFTER the fader? i know that i wouldn't get any eq on the pre auxes, but that is not too important. I was wondering if the addition of two pots caps and inductor (hf and lf, no mids) would affect the input impedance of the op-amp. I guess i could always increase the feedback and input to ground resistors to 22k or so. maybe I am just asking too much!

lastly, is it possible to switch the output after the panpot to an alternative mix-bus? will it;s disconnection affect the mix-bus? I cannot work this out from any schemos I have. i obviously can't actually mock this up, without actually building all 16/24 channels, so any advice is essential and very greatfully recieved!

I do hope that you guys don't think that I'm asking too many dumb questions!


many thanks in advance!!!!!


ANdyP

 

[the scum]

Sorry it seems like I'm off-radar. Since AES, things have been a little crazy...culminating in hip surgery tomorrow, and some downtime after that. Tinkering with a prototype of this is going to be one of my convalescent projects.

Before you get too deep into things, consider 2 important factors:
-Grounding is paramount, and avoiding ground loops in something like this is mindbending.
-The original schems didn't do anything to deal with DC offset. There need to be a handful of coupling caps added. Otherwise, the controls will be scratchy, and the switches will thump.

So, into your questions:

firsty, will I have a problem driving 4 auxes post? should i buffer them with another ic first?( i intend to use DOA for the main signal path, but would not mind a tlo7x or 5534 on the auxes . . . oh, and a direct out per channel)

You'll be pushing it for a tl07x (4 10K pots in parallel make 2k5, getting into the territory where the tl0's feel over burdened), but the 5534 will be fine.

Direct out would be as simple as taking the signal from the mute switch to your patchpoint.

secondly, I intend to use 19" racks for the case, since it would be easy to expand if necessary. What I need to get is some advice on the connecting of busses from one case to another. Will decent cable in a D25 connector do? I guess i am not worried about not being balanced. Some of my favourite consoles are not balanced in and out. ie Helios, trident etc. am i going to run into trouble with the busses unbalanced on signal cable?

Between Paul Wolff's use of DB25 on his new designs, and Bill Whitlock's recent stuff, I would vote for using "impedance balanced" or "ground-sense" busses (we discussed this earlier in the thread) on DB25s.

I had been working on cards where the inputs were selectable via jumpers: true to NYD's original with the fader hanging right off the input and an unbalanced buffer driving the auxes, or with the input opamp wired as a differential receiver, and the fader following it. You don't have to make all of the channels the same, and could experiment with either config.

thirdly. how do the component values change with additional channels? do the 22k "bus" resistor values need to be increased?

If you're using inverting summing nodes, they don't change. The summing amp holds the buss at ground potential, so each channel is looking at ground across those resistors.

Fred Forssell wrote a very good whitepaper about summing buss design issues, it's on his website.

fourthly, I was thinking about adding a second DOA on the input, before the channel fader, to provide a passive eq like the Op-Amp labs 425, or API 553. Then it struck me that perhaps i could use the one AFTER the fader? i know that i wouldn't get any eq on the pre auxes, but that is not too important. I was wondering if the addition of two pots caps and inductor (hf and lf, no mids) would affect the input impedance of the op-amp. I guess i could always increase the feedback and input to ground resistors to 22k or so. maybe I am just asking too much!

Anything in front of the input opamp is effectively buffered from it, so it shouldn't screw up the summing buss. The fader just looks like a 10K load to the EQ output.

I'd build EQs as entirely separate units from the summing buss. Then I'd wire the whole thing with patchbays so that the EQs were in line before the channel inputs, with patchpoints (AKA insert points) between the EQ outputs and summing inputs.

The Opamp/553 design isn't bad, but requires some rather large inductors. If you're looking for a 1 opamp EQ with parts that are easy to find, you might also look into the Baxandall circuit.

lastly, is it possible to switch the output after the panpot to an alternative mix-bus? will it;s disconnection affect the mix-bus? I cannot work this out from any schemos I have. i obviously can't actually mock this up, without actually building all 16/24 channels, so any advice is essential and very greatfully recieved!

Yes it's possible, no, it's not a problem. Just a switch an an additional pair of 10K resistors. This is often wired so that when the channel is disconnected, the buss feed resistors get grounded.

You might cruise the forum and the web, and track down old console schematics. They can be very handy to refer to in cases like this. The Yamaha PM1000 and Soundcraft 600 schems are in the manuals that both manufacturers let you download. Parts of the 600 look a lot like this design. I've also seen Trident 80, old API, Opamp Labs and Auditronics stuff floating around.

 

[BradAvenson]

I'll add a little bit to this discussion.

If you want to make a modular console, use balanced buses. Dual opamps are cheap, so they can be symmetrical, not just impedance balanced.

Every feature you add makes the execution a bit more tricky. Consoles are about the devilish details. It is tricky to keep track of them all when it gets complex.

Get the "Handbook for Sound Engineers" by glen ballou. Its worth the price of admission for the console chapter.

I also support the "steal older successful designs" method. But think about why they are good, that way you make sure to use the good parts of the design and can discard the odd bits that don't make sense for your project.

About grounding, I would say to make sure you know where your currents are flowing.

Engineering is about making a good set of comprimises. I would suggest using your DOA's for the large signal gains like summing and maybe EQ. Use ic's for small gains, buffering ect.

(I wish you a speedy recovery from your surgery tommorrow scum!)

 

[NewYorkDave]

I started to draw a schematic of an improved version of the circuit; I just haven't had time to finish it. But it looks like it's taken on a life of its own here.

 

[lovekraft]

[quote author="NewYorkDave"]I started to draw a schematic of an improved version of the circuit; I just haven't had time to finish it.[/quote]Notwithstanding the work that's been done so far, I for one would be very grateful for a look at your redesign, when you get the time to finish it. My needs are more in line with your original, simpler idea anyway, although I may well be in the minority.

 

[strangeandbouncy]

Dear Mr Scum,

I think I can speak for all hereabouts and wish you all the very best with your operation, and a speedy recovery. Our thoughts are with you.
Thank you so much for your reply, and also to everyone elses contribution. NewYorkDave is definately THE Man, and I look foreward with interest to his update on this.

Lets keep this one on the boil!



ANdyP

 

[Svart]

Bump.

I'm interested in this too as I have been slowly working on an active design myself...

 

[NewYorkDave]

I agree with almost everything in Brad's post--definitely read Steve Dove if you haven't already. But as far as true balanced busses go, it may not really be necessary unless the console is large. It'd certainly negate the "minimum amplifiers in the path" philosophy I was trying to follow. Perhaps a good compromise would be unbalanced channel modules feeding a balanced mix amp--say, a "Superbal" type.

I didn't show coupling caps because ideally, you could arrange to keep offsets low enough to not need them--without having to resort to servos or unnecessarily complicated/fiddly trimming schemes. But as the note on the schematic says, you may have to use them. It's the same deal with HF compensation in the feedback loop: add as needed, depending on your amplifier.

 

[Guest]

Can somebody explain please, what is advantage of balanced amps in terms of the target, I mean sonic qualities?

 

[strangeandbouncy]

Someone please correct me if I'm wrong, but I believe the sonic advantage is in noise rejection(?)


ANdyP

 

[Guest]

[quote author="strangeandbouncy"]Someone please correct me if I'm wrong, but I believe the sonic advantage is in noise rejection(?)


ANdyP[/quote]

No.

It is rejection of signals from external sources that are results of poor construction. However, it increases internal noises about 6 dB up and doubles distortions... Also, it makes distortions symmetrical that may be good for measurements and spec reports, but bad for sonic qualities.