You may find this thread interesting: http://www.diyaudio.com/forums/analog-line-level/271615-volume-control-modulus86.html
If you are using a single-ended source, the best solution is to drive a pot with the source and feed the output of the pot into a single-ended to differential driver. Interestingly enough, I just happen to be developing such a single-ended to differential driver. I should have it ready in June.
Tom
The deal is that that the Modulous86 and Parallel86 have their sonic signatures governed by small signal op-amps and as low current devices, don't vary greatly in operating temperature and may be assumed to have settled within a timeframe in the general range of fraction of a second, up to the duration of only one song, at most.
Tom has worked very hard to make sure that the "warm up" time (settling time) has been minimized when it relates to performance. Therefore, warm-up comments may have a shortfall of relevancy. And, that is by design.
You're right that long warm-up times DO exist in some solid state amplifiers; however, this isn't one of them. I have definitely had problems with some solid state FET based amplifiers drifting during warm-up (mainly irked that it sounded good until it warmed up and then it didn't) and some older BJT amplifiers with unsteady bias may also take an overly long time to settle (warm up). With a considerably different circuit design, those problems aren't relevant to Tom's amplifiers.
Due to the topology, one should assume that the warm up time is similar to a chip-based preamplifier, which is no more than a handfull of minutes, and that is ever so close to a moot point.
Therefore, if this was your favorite topic, that's alright, except that you're on the wrong page.
Last edited:
Thoughts, sure. No real action yet, though...
I am curious about the requirements for a USB DAC. What sort of features are needed? Which form factor should the DAC be (i.e. a stand-alone piece of HiFi equipment or more like a small box to tuck away behind the amps)? What kind of user interface is expected?
Rather than cluttering up this thread with the responses, please shoot me a PM with your answers to these. Note that these are top level type questions. Details (which DAC IC, etc.) can be worked out later.
Tom
So... I took delivery of a precision 1 kHz oscillator and just happened to find another zero on the THD: http://www.diyaudio.com/forums/vend...r-achieving-0-0004-thd-n-131.html#post4335396
Modulus-86 R2.0 @ 1 W, 8 Ω, 1 kHz: 0.000067 % (-123.5 dB) THD.
Tom
Modulus-86 R2.0 @ 1 W, 8 Ω, 1 kHz: 0.000067 % (-123.5 dB) THD.
Tom
Last edited:
If I recall the specs correctly, AP specs the APx525 at -112 dB residual THD. The APx525 I have beats this by about 3 dB under some conditions and about 0.9 dB overall. I'm pretty happy...
Something will always limit the test equipment performance. If the signal source was -140 dB THD, you'd argue that the -130 dB THD ADC was under-speced.
Personally, I choose to deal with reality rather than get hung up in the what-ifs.
Tom
Last edited:
Victor Oscillator THD
Here's the precision oscillator I used. I think it's from a DIY Audio member, actually. I found it on eBay (following the recommendations of a DIYA member).
The THD in this measurement is limited by the THD of the ADC of the APx525 (-130 dB THD). To measure the THD of the oscillator itself, a precision notch filter is needed to reduce the 1 kHz fundamental. I've seen measurements done this way that back up Victor's claims of harmonics below -145 dBc.
Tom
Here's the precision oscillator I used. I think it's from a DIY Audio member, actually. I found it on eBay (following the recommendations of a DIYA member).
The THD in this measurement is limited by the THD of the ADC of the APx525 (-130 dB THD). To measure the THD of the oscillator itself, a precision notch filter is needed to reduce the 1 kHz fundamental. I've seen measurements done this way that back up Victor's claims of harmonics below -145 dBc.
Tom
Attachments
Last edited:
The same argument as one would use when designing for really low THD.
But from experience reducing 20KHz breakup mode level on a driver, there is a significant difference audibly. Actually a few patents were generated in the process to obtain better wide band performance while taming the breakup modes.
20KHz is a measurement point, you could also choose some other high frequency point for amplifier as long as you feel the measurement will over enough ground to ensure predictable performance over a wide enough range where distortion variation with frequency will remain predictable without large changes in variation trend over a range of frequency. Maybe 10KHz?
But from experience reducing 20KHz breakup mode level on a driver, there is a significant difference audibly. Actually a few patents were generated in the process to obtain better wide band performance while taming the breakup modes.
20KHz is a measurement point, you could also choose some other high frequency point for amplifier as long as you feel the measurement will over enough ground to ensure predictable performance over a wide enough range where distortion variation with frequency will remain predictable without large changes in variation trend over a range of frequency. Maybe 10KHz?
Last edited:
Last edited:
Sure: Ultra Low Distortion 0 00001 1kHz Sine Generator Assembled and Tested PCB | eBay
Search for "Ultra Low Distortion 1kHz Sine Generator" on eBay. The seller's ID is michevich
Tom