impedance balancing resistor values

[hyde maintenance]

I'm in the process of recapping my PM1000 and adding direct outs. I like the sound straight out of the strip with no OT and was going to just run the lines out unbalanced. After testing things out a few times I found that I occasionally get some interference in the line so I figured I'd have to spring for some transformers until I did some reading about impedance balancing and tried it out- seems to work like a charm.

I have read that 33ohm resistors are good in this application but I don't have any in the house. So for shits and giggles I tried it with some 330 ohm resistors I had laying around and it seems to work ok running into my Audiofire12.

Here's my question- Is this value technically too high for this application? I live in a town where I can't just run out and buy a package of good resistors so if possible I'd like to use what I have instead of rushing my next mouser order.

[Scodiddly]

You're trying to match the unbalanced output with a pin-3 resistor, right? In that case you really want to try to match the unbalanced output impedance, which somebody has apparently pegged as 33 ohms. In this case 330 is probably better than infinity, so you're getting some good results.

How about 3 100-ohm resistors in parallel? If it works out, you can order a bag of 33-ohms with confidence.

[Scodiddly]

You're trying to match the unbalanced output with a pin-3 resistor, right? In that case you really want to try to match the unbalanced output impedance, which somebody has apparently pegged as 33 ohms. In this case 330 is probably better than infinity, so you're getting some good results.

How about 3 100-ohm resistors in parallel? If it works out, you can order a bag of 33-ohms with confidence.

[hyde maintenance]

Thanks for the reply. I think we may be talking about different things but I'm pretty green so see if this makes sense.
I'm running a resistor between the board out and pin 2, and then the same value off of ground to 3, then straight ground on 3.
Here's a link
http://web1.prosoundweb.com/install/sac/n27/rick/balout-1.shtml

It doesn't seem to agree with my older equipment, but with the Audiofire it's fine and gets rid of the broadcast hash that I had creeping in. Seeming as this is for direct recording it's ok if the only compatibility is with my convertors.

[hyde maintenance]

BTW all I have are 330's in bulk, so maybe I'll try 10 of them paralleled to see how 33ohms sounds.
Could get ugly but should work .
Thanks.

[The Scum]

The output impedance depends on where you're grabbing the signal from. If you're taking it from the fader buffer output, or the lee side of C37, then the circuit output impedance is roughly 22 ohms in series with 33 uf (emitter resistor & coupling cap...neglecting any resistance in the emitter itself.).

So you'd want a 22 ohm resistor, and a 33 uf cap.

If you're adding an extra series resistor on the hot side, then the cold side should have a resistor of that value, plus the emitter resistor and coupling cap values. You want both paths to look the same from the output pin to a low impedance node (ground on one side, amp output on the other).

As for loading, it's probably negligible with 330...you can always solve the divider presented with 330 in series with 10K in series with 330 (assuming you're driving a 10K input). Since 660 is less than 10% of 10K, the general guideline for loading deems this reasonable.

[Jim Williams]

Use a 220 uf output cap, not 33 uf.

Go to Radio Shack and buy a bag of resistors.
_________________
Jim Williams
Audio Upgrades

[hyde maintenance]

I've already increased the value of the output cap to 220uF as you advise elsewhere Jim.
So as per Scum's suggestion I will try this value for the cold side series cap.

As for the resistors, I was thinking 1% metal film were the norm in audio, but if you think rat shack 5% carbon are just fine Jim then I'll listen to your advice.

Sorry if these are stupid questions but I'm pretty green with all this, in a location where good parts aren't as readily available as one might hope, and I'm tired of putting projects aside until the next order from mouser arrives.

Thanks though for all the advice so far. It is appreciated.

[nate]

Match up a couple of the 5% resistors with a multimeter before you put them in. It'll give you a better CMRR.
_________________
http://www.natedort.com

[hyde maintenance]

Rat Shack here is now called The Source, and The Source isn't much of a source for anything besides crappy consumer electronics. The only resistors they had were the assorted packs with 2 pcs of 30 or so values all piled together- man o man.

I picked up some 1%ers out of town this weekend and will go from there.
Thanks everyone.

[The Scum]

There are so many other 5% carbon resistors in this circuit that one more isn't going to really be a problem. Certainly not in an output impedance build-out application, where it's not going to be followed by buckets of gain.

[A.David.MacKinnon]

[hyde maintenance]

I'm in Guelph...so not tooo far. I don't head in to Toronto much these days but it looks like that may change soon.
If you have suggestions Dave for next time I'm there it would be great.

[mixedupsteve]

Rat Shack is still Rat Shack down here in MD.
Those cheap 5% resistors they sell are metal film I'm pretty sure not carbon.
_________________
Steve

www.myspace.com/rollingrecorders