Hey all, building a SSM2142 based Line Driver to convert Unbalanced to Balanced. I have a simple PCB that has input NE5534 Opamp Buffer going into the SSM2142. Couple questions...
1. There is a 1uf cap at input of the NE5534. My unbalanced out preamp is an 6922 / 6H30 Aikido with output caps already - do I need these 1uf caps if I already have the DC blocking caps?
2. The NE5534 is a single opamp and I have some subs - OPA627, OPA637 amoung others...I know these are pretty much not same league as the NE5534, but I have them available...worthwhile to swap?
3. Each output of the SSM4124 is a 33uf bi polar cap - I would think it is worth while to use best cap here since the signal is going thru? I have Silmic II 47uf that I was thinking of trying, or just stack twelve 10uf PP caps there?
Thanks!!
1. There is a 1uf cap at input of the NE5534. My unbalanced out preamp is an 6922 / 6H30 Aikido with output caps already - do I need these 1uf caps if I already have the DC blocking caps?
2. The NE5534 is a single opamp and I have some subs - OPA627, OPA637 amoung others...I know these are pretty much not same league as the NE5534, but I have them available...worthwhile to swap?
3. Each output of the SSM4124 is a 33uf bi polar cap - I would think it is worth while to use best cap here since the signal is going thru? I have Silmic II 47uf that I was thinking of trying, or just stack twelve 10uf PP caps there?
Thanks!!
Attachments
The ssm2142 already has a specified 10k input impedance, presumably to the inverter opamp.
Why do you think you need a Buffer opamp?
This is the spec for the Source
Why do you think you need a Buffer opamp?
This is the spec for the Source
Last edited:
Looks like more of a preamp than a buffer, going by the number of resistors (and lack of compensation capacitor). Noninverting, I'd say. And the OP is clearly not the board designer.
Are the resistors the same as on the picture? I don't think 3x 31.6k is particularly realistic or refined, possibly those were just populated for the shot and would be different values on the real board.
You could swap the trusty NE5534s for OPA627 but I don't think it'll make much of a difference. Capacitors seem adequate as-is.
The way you hook up your jacks and ground is likely to be far more important than any of this stuff. A variant that works could look like this:
Power ground comes from power supply, where case is connected to star ground.
Audio in terminates in two RCA jacks that are insulated from the case, with just a ~10 nF ceramic capacitor coupling ground to case.
Balanced audio out goes to pins 2,3 of XLR jacks.
Pin 1 of XLR jacks connects directly to case (low resistance and inductance). Do not connect to small-signal audio ground.
Note that even if the converter box is kept floating, the equipment it connects to is quite likely to be grounded, and as such may introduce a ground loop over the unbalanced input if the source is grounded as well. It probably is best integrated into a source component, where - again - proper internal grounding is to be observed as usual.
Balanced, if you want to do it right, is pretty much an all-or-nothing affair. Well, that, or you may need a bunch of ground loop isolators (hint: don't use the cheap and nasty ones).
The smart part with balanced connections is not so much the transmitter but rather the receiver anyway. You can always turn an unbalanced output into a properly balanced one with just one or two components added. Of course you may need some extra gain, and achievable output levels at a given supply voltage will be about 6 dB less than a "full-fledged" push-pull type output, and some balanced inputs perform a fair bit worse with such a source, but still, it is properly impedance-balanced. Basically what you do is hook up unbal output to "hot", and then replicate its source impedance between unbalanced return and "cold" with identical components (typically a series resistor and capacitor).
Are the resistors the same as on the picture? I don't think 3x 31.6k is particularly realistic or refined, possibly those were just populated for the shot and would be different values on the real board.
You could swap the trusty NE5534s for OPA627 but I don't think it'll make much of a difference. Capacitors seem adequate as-is.
The way you hook up your jacks and ground is likely to be far more important than any of this stuff. A variant that works could look like this:
Power ground comes from power supply, where case is connected to star ground.
Audio in terminates in two RCA jacks that are insulated from the case, with just a ~10 nF ceramic capacitor coupling ground to case.
Balanced audio out goes to pins 2,3 of XLR jacks.
Pin 1 of XLR jacks connects directly to case (low resistance and inductance). Do not connect to small-signal audio ground.
Note that even if the converter box is kept floating, the equipment it connects to is quite likely to be grounded, and as such may introduce a ground loop over the unbalanced input if the source is grounded as well. It probably is best integrated into a source component, where - again - proper internal grounding is to be observed as usual.
Balanced, if you want to do it right, is pretty much an all-or-nothing affair. Well, that, or you may need a bunch of ground loop isolators (hint: don't use the cheap and nasty ones).
The smart part with balanced connections is not so much the transmitter but rather the receiver anyway. You can always turn an unbalanced output into a properly balanced one with just one or two components added. Of course you may need some extra gain, and achievable output levels at a given supply voltage will be about 6 dB less than a "full-fledged" push-pull type output, and some balanced inputs perform a fair bit worse with such a source, but still, it is properly impedance-balanced. Basically what you do is hook up unbal output to "hot", and then replicate its source impedance between unbalanced return and "cold" with identical components (typically a series resistor and capacitor).
No, I did not design the pcb. Bought it off the 'bay - just pcb - and populating with quality parts now.
If it sounds as good as I am reading, I was going to just drop it into my DIY Aikido Preamp chassis. Hence the reason for asking if I really needed that input cap, and the quality output caps.
Dropping in the OPA627 here seems a bit of a waste perhaps...
If it sounds as good as I am reading, I was going to just drop it into my DIY Aikido Preamp chassis. Hence the reason for asking if I really needed that input cap, and the quality output caps.
Dropping in the OPA627 here seems a bit of a waste perhaps...
Obviously you wouldn't need two coupling caps in series. A jumper in place of the 1µs would do.
Output caps are not likely to ever see much DC and should be either bipolar or low-leakage types. Otherwise long-term performance is hard to predict. Behringer might use standard polars, but that's just because they're cheap.
Output caps are not likely to ever see much DC and should be either bipolar or low-leakage types. Otherwise long-term performance is hard to predict. Behringer might use standard polars, but that's just because they're cheap.
- Status
- Not open for further replies.