PCB for differential line input

[Matt C]

I could design this board for you if you need something small and a bunch of them. Send me a pm with the size, quantity, and circuit you want. Or I can just use the big standard one.

tried to send you a message buy your inbox is full.  send me a PM if you get a chance.

That's the basics of the circuit . But you need to add decoupling as advised. Also consider whether you want a small resistor in the voltage lines to give RC filtering and Short Circuit protection for the supply if needed (Use non flammable NFR resistors). Also if you need overvoltage protection on the inputs. Any rfi countermeasures ? And, of course, the connection type detail...

yes I was intending to include RF filters, coupling caps, PSU decoupling, etc...

 

[atavacron]

Digging up this thread! What did you come up with? I have an application for 32 balanced line receivers, the power supply may end up having to be 24V, and was thinking an 8-channel board would be nice.

 

[Matt C]

board member iampoor1 designed a set of PCBs for me that are working great. Each board is 8 channels and fits well under the meter bridge in my console.  It's a pretty "standard" balanced line input configuration using a surface mount OPA604.  Let me know if you want more details.

 

[JW]

Is this for balanced returns on your PM2K?

 

[atavacron]

board member iampoor1 designed a set of PCBs for me that are working great. Each board is 8 channels and fits well under the meter bridge in my console.  It's a pretty "standard" balanced line input configuration using a surface mount OPA604.  Let me know if you want more details.

I do! I will PM you.

Also I am checking out the Regurgitator from JLM, which looks handy. The PCBs come connected 8 at a time.

 

[Matt C]

Is this for balanced returns on your PM2K?

Yes, I'm balancing all the insert returns. I'm mounting the boards right above the big array of transformers that lives behind the mothetboard/under the meter bridge.

 

[iampoor1]

I would be happy to do another run of the boards. I can also modify them if you have any special requirements and/or need a few!

 

[atavacron]

I would be happy to do another run of the boards. I can also modify them if you have any special requirements and/or need a few!

Sorry, just saw this. Hey you're in California, that's cool. I'm in Oakland.

 

[Rocinante]

I used the regurgitater in several projects. Great pcb, but if iampoor can break it down smaller and do a 8 channel board I'd be keen on that. Opa2604 flavor would be nice since I have some BB's left over but something +- 15/18v would probably be more rounded. Those 1200s have great specs.

 

[iampoor1]

Sorry, just saw this. Hey you're in California, that's cool. I'm in Oakland.

Sweet! Im in Modesto, so pretty close!

I used the regurgitater in several projects. Great pcb, but if iampoor can break it down smaller and do a 8 channel board I'd be keen on that. Opa2604 flavor would be nice since I have some BB's left over but something +- 15/18v would probably be more rounded. Those 1200s have great specs.

I have the 8 channel board I did for matt, But I also have an 8/16/32 channel board using db-25 IO and That balanced line receiver if that more your style. 

 

[atavacron]

This is going in a really cool direction.

Matt was kind enough to send me the files for the 8-way OPA604 board. It's really nice! Clean layout for sure, plenty of space for anyone to assemble. That one uses the SOIC package, but obviously a socketed DIP would be the icing on the cake.  I'm not gonna go using it or posting it without permission obviously, but am glad to have an understanding of scale and construction practicalities. I think if it were to be something that more folks could apply to projects, the individual channels might want to get compacted a wee bit, the sections would want to be all in a row like the Regurgitators, and the audio I/O would want to turn into two 16-way .100" headers. Good ±24V option.

For ±15-18V yes the THAT 120X seems the way to go. I'd love to see what that version looks like. In my application it's not quite straight ahead DB25 input. Inputs are on DB25s, but to accommodate the routing downstream of the receivers, the DB25s want to get collated by the time they hit the board. So for example if Inputs 1-8 are on one D sub and Inputs 9-16 are on a second D sub, they are organized as 1 / 9 / 2 / 10 / 3 / 11 et cetera by the time they hit the board. The practical reason for this is because they are pre and post inputs to an auxiliary send matrix mixer. Once unbalanced, they head off to frontpanel switching that selects pre or post. There is however one DB25 that does go straight to eight receivers, it happens to be what you would consider Inputs 25-32 (the matrix is  ultimately 12x8 with 8 direct inputs to the busses).

Anyhoo, 8 THAT 120X sections in a row is surely going to take up at least 8" x 2", and it may be that a board that is any wider would be a bit cumbersome for people laying out a variety of installations. I'd definitely be interested in seeing where the design is at so far!

[EDIT] Another thought is that there might want to be a little dip switch next to each THAT chip that allows one to select positive or negative gain.

 

[atavacron]

Do people have thoughts on the Regurgitator approach (both sections of the OPA2604 in parallel) vs a single OPA604 receiver? I don't quite understand the differences but I know it has to do with output current and a sort of unity gain default. Block diagram attached.

Also is the OPA2604 still having issues or did TI fix it?

 

[guy_4]

Hello,

Here is an option for adding a balanced input, using some of the famous THAT  IC's :

https://www.sparkfun.com/products/14002

At $14.95 per unit I can't see a better solution.

They also have a same size board for Unbal-to-Bal solution.

Anybody here have experience with Sparkfun company ?

Best,
Guy

 

[atavacron]

THAT Corp just blew my mind. 1646 as a buffer to drive a 600 ohm load (not a transformer obviously). See attached.

“The headroom will be a combination of the two [chips], but you can expect 24dBu at least [with bipolar 15V supply].”