Would anybody be interested in a super balanced PCB that would work with this DAC? You could put NE5532 compatible op-amps in (5532 is a good choice IMO) such as LM4562. The improvement is significant over a single op-amp output such as that built into the kit.
I have a plan to produce this PCB for another DAC so it would be easy to do it for this at the same time.
See the simple difference between a single NE5532 (Green) and the superbal config using the same op-amps (Red).
(3rd Harmonic is not higher in Red, just equal)
I have a plan to produce this PCB for another DAC so it would be easy to do it for this at the same time.
See the simple difference between a single NE5532 (Green) and the superbal config using the same op-amps (Red).
An externally hosted image should be here but it was not working when we last tested it.
(3rd Harmonic is not higher in Red, just equal)
Yes since the AK4393/96 have differential outputs.
Which would only be of benifit if you were driving the inputs of a balanced preamp.
Hi,
Not at all, it would give single ended outputs (only).
Super-balanced referrers to the input of the circuit which significantly reduces common mode distortion over using the + and - inputs of a normal op-amp arrangement.
My preamp (UGS) has inputs rca and xlr and is based on a symetric design(Nelson Pass), i think it's better to use the XLRs. Could someone help me with the wiring from dac to transfo and transfo to xlrs. Transfos are TAMradio TD-1 (600R primary and 600R sec with a centertap on each side). transfo shematic shows a ground between prim and sec ?
Thanks.
Thanks.
In that case let me be a little more clear for you. You can remove the resistors near the connections. No point driving the op-amp output if you don't use it.
Connecting to Gnd is optional by the way, as long as the cable is grounded at the pre-amp side. Probably better not to ground in that case.
Uploaded with ImageShack.us
Connecting to Gnd is optional by the way, as long as the cable is grounded at the pre-amp side. Probably better not to ground in that case.
An externally hosted image should be here but it was not working when we last tested it.
Uploaded with ImageShack.us
I don't see how that would improve much vs summing into a single op-amp.Since to take full advantage of any CMRR of distortion and noise would requires keeping the signal in it's differential form through to the output.
Why?
I don't claim to be an expert however I thought common-mode distortion was mainly caused at the input stages of an op-amp.
I think it works like this; the inverting input of an op-amp is usually driven by the op-amps output via a resistor to control feedback, while the non-inverting input isn't. This means the input stages, even when driven by a differential input see non-symmetrical voltages. The super-balanced arrangement drives the non-inverting input with an inverted version of the op-amps output so that the voltages seen at each op-amp input are symmetrical. Common mode rejection is then able to do it's job and knock distortion down.
I'm not sure how this is different than rail bootstrapping in what it actually does. Can anyone enlighten me?
I don't claim to be an expert however I thought common-mode distortion was mainly caused at the input stages of an op-amp.
I think it works like this; the inverting input of an op-amp is usually driven by the op-amps output via a resistor to control feedback, while the non-inverting input isn't. This means the input stages, even when driven by a differential input see non-symmetrical voltages. The super-balanced arrangement drives the non-inverting input with an inverted version of the op-amps output so that the voltages seen at each op-amp input are symmetrical. Common mode rejection is then able to do it's job and knock distortion down.
I'm not sure how this is different than rail bootstrapping in what it actually does. Can anyone enlighten me?
Last edited: