Lenghty Unbalanced Cable run questions

[byoung]

OK so I have a possible issue cabling some equipment. In the new studio in which I'm working, there isn't room for my keyboard racks in the control room, but I wanted to run a 16 channel cable from my Rack to the studio patchbay. The length of the entire cable run from my rack to the studios mixing desk inputs is roughly 60 feet. It is actually all balanced cable, but the signals going into the cables are all unbalanced from my synths and modules. I am wondering if this will be a problem with signal degredation or if I will be alright running it this way? Or is there a way to balance the 16 lines as they leave my rack? Maybe something I can buy or build...any ideas??

I'm looking for the best options because it looks as if this will be a permanent setup at least for a while. Previously I was connecting all of my modules to a Mackie LM-3204 and running stereo balanced lines out of that, but I really need more than stereo to access the full potential of my rack.

 

[JohnRoberts]

You will get the most benefit from the balanced wiring and proper balanced/differential receiver at the receiving end. On the low send you can balance the impedance with a simple resistor (equal to the build out R on + side).

Differential drivers double the signal level for improved S/N and provide a first order cancellation of electrostatic crosstalk (+ and - cancel each other out), but at significant expense. For 60' you can probably get away with a strong SE signal.

Some gear is marginal sending signal 10' so you may need to buffer the weaker sisters in your rack.

JR

EDIT: back in the days of small format tape recorders there used to be commercial bump boxes available that contained maybe 8 channels of SE -10dBV to +4 bal output in one rack mt package. If you find one of these used it might be a good start.

 

[byoung]

What you need is a 5534 as a non inverting follower with a 100K resistor to gnd at the input.

The 5534 drives a THAT 1646 line driver. These chips are real bulldozers.

Roger,

sounds simple enough, do you have a schematic for the box you built, or would you recommend using this circuit---unbalanced line into 5534 circuit schemo into the 1646 schemo out to balanced line? or are there mods to the 5534 circuit and 1646 circuit I need to be aware of?

For 60' you can probably get away with a strong SE signal.

John,
sorry but I'm a beginner in circuits still, and lacking in some terminology, I'm not sure what the SE signal is? It does sound like it may be a simpler cheaper solution than rogers box though?

 

[pstamler]

SE presumably means single-ended, or unbalanced.

I think the 5534 and the THAT chip are the best solution, though. One thing to watch out for: the 5534 is not spec'd as stable at unity gain. Some samples are, some aren't; a 22pF capacitor between pins 5 and 8 on the DIP will solve the problem.

Peace,
Paul

 

[JohnRoberts]

Indeed 60' of quality balanced wiring is not a long distance.

If you want to get specific you can look up the capacitance per foot and even predict the expected HF loss from individual product's output/source impedance.

Yes, SE means single ended or single legged output but as I already mentioned you can run SE through balanced wiring and get most of the benefit.

This is indeed a good beginner circuit to dip you toe into the water of design. NE5532 (a unity gain stable dual opamp) is a good inexpensive general purpose opamp with decent drive capability. I remain unconvinced that you need differential or true balanced outputs for only 60' but low source impedance and decent drive level will be of use, to help any products that don't cut it. Especially important is good differential receivers on receiving end and balanced impedance in lines.

Can we assume patch bay is balanced (3 circuit?). This is all pretty basic stuff but audio is never trivial.

JR

 

[byoung]

the cables used will be redco RGS snake lines and they will be tied into a 3 conductor TT patchbay, which will then be normalled to the mixing desks line inputs(which from what I was told accepts balanced or unbalanced connections).

As I was looking at the schemo for the 5534 I noticed that TI also makes the drv134 line driver. I could be completely off my rocker but it looks like it is a all in one chip that would combine the 5534 and the 1646 into one, maybe making the circuit even simpler? Maybe not as good of a solution.

John do you think that I should just wire up my rack directly to the bay and see what happens signal wise?

 

[byoung]

wow thanks for the schematic, that was how I figured it would work out, what type of PSU would you recommend for 8-16 of these circuits?

 

[JohnRoberts]

Less is generally best, for a stable fixed application if it works ok without adding more circuitry in the path what's the benefit?

One consideration with a patch bay is that you never know what might get patched into it, so YMMV

If you have some products that clearly don't handle the (IMO not very long) run, address them, maybe build in some line drivers as needed.

JR

 

[mikep]

IMO some synths are just fine unbuffered. (Tritons are NOT). properly done impedance balancing works wonders nearly 100% of the time as long as the following input is doing its job. think of the buffer not as a new output for the synth, but as a new input for the mixer. I suspect roger's circuit would work just about as well without the line driver IC, just impedance balance the 5534 and set up 6dB of gain.

for this problem Ive had good luck with step-down DI transformers. kills hum for sure. part of the problem is output loading, so you dont want to use a 1:1 of any nominal impedance IMO. but you also don't need a gigaohm input-Z. this will drop the output-Z and the level, now the synth has a mic-level output and you have to plug it into a... MIC PRE. so you can choose the pre based on the intent. on one hand you can find a transformer that matches well with a particular synth and build it right into the case, or amass a collection of individually boxed step down transformers so you can mix and match. I only have a couple but have found this is very useful. try the 1:10 mic transformers from 70's shure solid state mic mixers wired in reverse for a synth DI. love it on a juno.

EDIT: IMO IMO IMO IMO

mikep

 

[mikep]

[quote author="RogerFoote"]Of course a feller can do whatever floats his or her boat.

I just found that many synths are a little duller sounding when loaded at 10K. [/quote]

I agree completely. I didnt mean to sound like I was dissagreeing with you. the transformers I liked tend to have a rising high end. ever try a z-vex SHO pedal? (super hard-on) nice on synths, it adds a subtle bit of brightness, and buffers.

 

[byoung]

So I was going to try and build two line drivers and see how well they work. As Mike was mentioning earlier about cable capacitance I got this from the spec sheet of the redco cable. Which values do I need to look at?

Capacitance at 1kHz, Between conductors 21pfd/ft Conductor to shield 38pfd/ft

also how would you add switched gain, not saying I need it, but just curious how it would be implemented to the circuit. Also am I correct in assuming that I could just add a switch at the output that disconnects the sheild for that possible ground loop hum issue?

Brian