DIYstompboxes.com

DIY Stompboxes => Building your own stompbox => Topic started by: KarenColumbo on April 23, 2017, 02:24:36 AM

Title: (As of yet) theoretical question mixing symmetrical+instrument signals
Post by: KarenColumbo on April 23, 2017, 02:24:36 AM
Hi all,
I want to build a little mixing box that mixes an instrument signal (bass guitar) with a symmetric signal (from mixer/playback device) and puts the mixed signal out again via symmetrical outputs (XLR and 6,3mm).

What I gathered so far is:

- The symmetrical input has to be "broken down". I found some explanation at Rod Elliott's site in his audio mixer project: Here's the input module. (http://sound.whsites.net/p30a_f4.gif)





- Of course the instrument signal needs to be buffered, too. (http://sound.whsites.net/p94-f1.gif) Maybe it needs some adjustable gain to compensate for different output voltages.


- the mixing stage seems to be pretty straightforward. (http://sound.whsites.net/articles/audmix-f4.gif)

- the output(s) need to be symmetric. I thought about using this IC here: http://www.thatcorp.com/1600-series_Balanced_Line_Driver_ICs.shtml (http://www.thatcorp.com/1600-series_Balanced_Line_Driver_ICs.shtml)

Is there some obvious flaw I'm missing so far?
Title: Re: (As of yet) theoretical question mixing symmetrical+instrument signals
Post by: KarenColumbo on April 23, 2017, 06:54:11 AM
Here's the thing cobbled together in PS
(http://www.mrscolumbo.com/uploads/stomps/V0.1.png)
Title: Re: (As of yet) theoretical question mixing symmetrical+instrument signals
Post by: Mark Hammer on April 23, 2017, 08:39:18 AM
Maybe it's just to early on a Sunday morning for me, but I need some clarification.

Is your objective to mix two balanced input signals, down to a mono unbalanced signal and then convert the mixed-down signal to a balanced mono output?
Title: Re: (As of yet) theoretical question mixing symmetrical+instrument signals
Post by: KarenColumbo on April 23, 2017, 09:07:19 AM
Hey Mark, thx for reacting :)

The goal'd be to mix a balanced signal (could be some playback or a line mix) with an unbalanced instrument signal (direct from a bass guitar) to output it to a personal monitor box, which requires a balanced signal at its input.

I haven't reviewed the possible inverting stages yet - but I will, btw. I just wonder if this will work out alright
Title: Re: (As of yet) theoretical question mixing symmetrical+instrument signals
Post by: ElectricDruid on April 23, 2017, 09:59:14 AM
You could probably reduce that circuit a bit by using a balanced input receiver if you're buying THAT Corp parts anyway. I'm pretty sure they have some. They might even have one that does both balanced in and balanced out on a single chip.

My other thought is to wonder if you couldn't get away with just having two mixers. The balanced signal would go into them both (one for the positive, one for the negative). You've got your balanced output straight away without needing a special chip to convert from single-ended to balanced.

The single-ended input would get mixed into the -ve input of one op-amp and the +ve input of the other (e.g. one of the mixers would be a differential amp).

The only trouble is your mix pots would have to be dual gang. Aside from that, I think it'd be doable in two op-amps.

Tom


Title: Re: (As of yet) theoretical question mixing symmetrical+instrument signals
Post by: KarenColumbo on April 23, 2017, 11:46:35 AM
Hey Tom! Could you elaborate on the unbalanced input being put in on two amps at the same time? Of course I'd love to reduce the part count a bit - but respect (and ignorance) tells me to treat those two input signals separately. Btw. I got 3 of these: http://www.thatcorp.com/1240-series_Balanced_Line_Receiver_ICs.shtml (http://www.thatcorp.com/1240-series_Balanced_Line_Receiver_ICs.shtml), but they're input only, as far as I can see, and they're balanced signal only.

But maybe I could get rid of U1 entirely by using the 1243. Sounds feasible.

Like this? Sounds too simple ...
(http://www.mrscolumbo.com/uploads/stomps/V0.2.png)
Title: Re: (As of yet) theoretical question mixing symmetrical+instrument signals
Post by: tubegeek on April 23, 2017, 12:04:47 PM
DIY is fun as hell, I know, but a dirt-cheap Behringer mini mixer will solve this problem without breaking a sweat. Hard to compete.

THAT has limitless application notes which can simplify design matters a lot - looks like you are well on your way with that, so far so good. Between them and Rod Elliott you are in good hands.


Title: Re: (As of yet) theoretical question mixing symmetrical+instrument signals
Post by: ElectricDruid on April 23, 2017, 01:36:10 PM
Quote from: KarenColumbo on April 23, 2017, 11:46:35 AM
But maybe I could get rid of U1 entirely by using the 1243. Sounds feasible.

Like this? Sounds too simple ...
(http://www.mrscolumbo.com/uploads/stomps/V0.2.png)

That looks like a good start! The 1243 datasheet shows various options for how to wire that bit up - depends on what signal level you're expecting.
That's already a ton simpler.

What I had in mind is something like this:

(http://electricdruid.net/images/BalancedMixer.png)

I don't know if this works, but I don't see why it shouldn't. The top op-amp is a simple mixer, no problems there. The bottom op-amp is a differential amp, used as a mixer. The IN2 signal is fed into the inverting input of one amp, and the non-inv input of the other, so you finish up with a differential (balanced) signal in the mix. Note the whole thing is inverting, but I suppose you could swap the OUT+ and OUT- labels over to fix that.

Am I crazy, or would that work?

Tom



Title: Re: (As of yet) theoretical question mixing symmetrical+instrument signals
Post by: Kipper4 on April 23, 2017, 04:34:39 PM
I can't answer that but I hope it does Tom.:)
Title: Re: (As of yet) theoretical question mixing symmetrical+instrument signals
Post by: anotherjim on April 23, 2017, 05:15:03 PM
Tom's idea looks good to me. Assuming bipolar power supplies.
The unbalanced input might need buffering - don't know what will be driving it.

You can be cheaper than this, if you know something about the output stage of the balanced source and can be certain a particular one is always going to be feeding it. Some can be taken to unbalanced very easily by either ignoring the cold wire or grounding it. Depending, you might lose some or all of the noise reduction benefit of balanced cable, but not every situation is that critical.
Balanced output can also be cheated with a single op-amp and impedance balancing to the cable out. x2 gain needed to make up for losing a real cold feed.
Title: Re: (As of yet) theoretical question mixing symmetrical+instrument signals
Post by: KarenColumbo on April 23, 2017, 05:26:34 PM
The balanced input signal is almost certainly a monitor mix from the FOH mixer. The unbalanced one will be one of a few heavily modded active or passive bass guitars. I will inquire further details! Thanks so much!
Title: Re: (As of yet) theoretical question mixing symmetrical+instrument signals
Post by: PRR on April 23, 2017, 06:11:55 PM
> balanced input signal is almost certainly a monitor mix from the FOH mixer.

> Elliott's site in his audio mixer project: Here's the input module.

The FOH feed is high level.

Elliot's stage is for LOW level, microphone, inputs. Figure the gain: up-to 200. Take a 1V level from FOH, boost it to 200V.... sumthin will burst.

Line receivers are VERY standard things, on Elliot's site or a billion others. Copy/steal from commercial products. While Elliot's plans are very good (if you read and understand), other web-"authorities" may be less correct. If you find say Mackie's "Bal Line In" plan, you can be pretty sure what it is and that it works.

You can do the bal:unbal with a transformer, or one two or three opamps. The 2-op plans are better at an extreme. The 1-op plan is fine for almost all stage-work.
Title: Re: (As of yet) theoretical question mixing symmetrical+instrument signals
Post by: KarenColumbo on April 24, 2017, 10:01:35 AM
My buddy and I will have beer tonight and I'll ask him what's actually put in here. From what i gathered so far there should be the option to adjust the gain at both inputs, but then it will be set like this forever. It's "upper left" on his pedal board and he's using it, whenever there's a venue where he don't trust the technician or there's no technician at all.

Do you think the setup with the THAT ICs is too much? I mean, they're specially made for occasions like this, I thought ... well, not really, 'cos my setup here is far from professional audio.

I'll have some more reading. Thank you all, of course - if I finish it, there will be some PCBs made for those who want them, I promise!
Title: Re: (As of yet) theoretical question mixing symmetrical+instrument signals
Post by: KarenColumbo on April 24, 2017, 10:09:27 AM
Quote from: PRR on April 23, 2017, 06:11:55 PM
> balanced input signal is almost certainly a monitor mix from the FOH mixer.

> Elliott's site in his audio mixer project: Here's the input module.

The FOH feed is high level.

Elliot's stage is for LOW level, microphone, inputs. Figure the gain: up-to 200. Take a 1V level from FOH, boost it to 200V.... sumthin will burst.
Couldn't I attenuate the input with a pot at the input? And leave it there until it's needed?

Quote
Line receivers are VERY standard things, on Elliot's site or a billion others. Copy/steal from commercial products. While Elliot's plans are very good (if you read and understand), other web-"authorities" may be less correct. If you find say Mackie's "Bal Line In" plan, you can be pretty sure what it is and that it works.
I had a good long Google at this - and they're all more or less alike. I just settled for those THAT chips by links I found on the pages that had those "classic" schematics. Do you think I'm on a wrong track here?

Quote
You can do the bal:unbal with a transformer, or one two or three opamps. The 2-op plans are better at an extreme. The 1-op plan is fine for almost all stage-work.
I will have a look at this. The transformer solution probably is the best way from what I read. Like a DI with a mixing stage. But what if the inputs already come from a DI box?