When you click on the tab ‘Performance Graphs’, the second graph here, shows that the distortion skyrockets above 22 dBu. 22dBu is about 9.75 Vrms. 9.75 Vrms is about 13.8V peak. So that is 27.6 Vpp.
25 watts RMS/8 ohms for a standard F4 is 14.14 Vrms. That’s the usual output power of a stereo (with single ended input) F4 per the design document.
~10V RMS that the Universal Buffer can output on it’s best day with the highest voltage rails (+/- 17V DC) will make the stereo F4 output approximately ~ 12 watts RMS/ch. In other words, if you want to reach 25 watts RMS of Class A output power from the F4 you will run out of voltage swing from the Universal Buffer per my calculations.
If you want to build an F4 with two boards per channel, i.e. a balanced input and balanced output then you will need 28V RMS or 80Vpp to achieve 100 watts into 8 ohms with a distortion of roughly 1%.
Sorry if the above was a bit long winded, but I wasn’t clear if you were going to use the Universal Buffer board as a balanced to unbalanced converter for a standard F4 stereo build or if you were trying to go for a balanced input AND balanced output version that would be built ideally as monoblocks.
Best,
Anand.
With regards to the above analysis, realize I
am being super conservative. Even at 27dBu we are only talking about 0.1% distortion!! 27dBu is 17Vrms.
17 Vrms > 14Vrms, so you may be happy with that in a standard F4 implementation. It’s entirely likely that the UB will have less distortion than the power buffer F4 at that level. In fact, Pass spec’s the F4 for 0.5% THD @ 25 watts.
Best,
Anand.
am being super conservative. Even at 27dBu we are only talking about 0.1% distortion!! 27dBu is 17Vrms.
17 Vrms > 14Vrms, so you may be happy with that in a standard F4 implementation. It’s entirely likely that the UB will have less distortion than the power buffer F4 at that level. In fact, Pass spec’s the F4 for 0.5% THD @ 25 watts.
Best,
Anand.
Last edited:
Thanks for the careful response Anand.
It is not an F4, but a new design (balanced, 0dB) that I'm working on (which like the F4 should be able to do 25W into 8R).
Unfortunately, I don't have any front-end / pre-amp to test this and I guess that the universal buffer (balanced in, balanced out) can't help me here as I do need some 15 Vrms.
It is not an F4, but a new design (balanced, 0dB) that I'm working on (which like the F4 should be able to do 25W into 8R).
Unfortunately, I don't have any front-end / pre-amp to test this and I guess that the universal buffer (balanced in, balanced out) can't help me here as I do need some 15 Vrms.
15 V RMS = 21.2 V peak. That's hard to do on a 17 V rail, even for relatively large values of 17...
Thanks Anand for helping out. I was just about to dig out that graph you mention.
Tom
Hi
I am not an electronics person but this is the recommended board for using unbalanced amps with the Okto DAC8 Pro so I am willing to have a bash at this.
I am hoping that my few questions won't be too silly.
Q1. Can I wire up two RCA phono sockets per channel (output) so that I can use a stereo amp to bi-amp each speaker.
Q2. Is it as simple as connecting all the wires for balanced XLR input and RCA unbalanced output plus 15V DC power or is there more to it
Q3. Basic phone charger style power plugs look to be 2 wire (+/-) and no GND. Is this a problem. Do I need a power supply with a GND as well.
Thanks all.
I am not an electronics person but this is the recommended board for using unbalanced amps with the Okto DAC8 Pro so I am willing to have a bash at this.
I am hoping that my few questions won't be too silly.
Q1. Can I wire up two RCA phono sockets per channel (output) so that I can use a stereo amp to bi-amp each speaker.
Q2. Is it as simple as connecting all the wires for balanced XLR input and RCA unbalanced output plus 15V DC power or is there more to it
Q3. Basic phone charger style power plugs look to be 2 wire (+/-) and no GND. Is this a problem. Do I need a power supply with a GND as well.
Thanks all.
Yep. The Universal Buffer is a stereo board so you can wire left and right balanced output from the DAC to it and get left and right RCA output.
The power supply is a bipolar/split supply. So you need to provide V+, GND, and V-. I run it on ±12 V from a Preamp Power Supply. The buffer will work from ±2.5 V to ±17 V, but naturally it can only provide as much voltage on the output as you apply at the power supply so I recommend ±12 to ±17 V.
Connecting input and output is exactly as you describe. You just hook up the XLR and RCA connectors and you're ready to go.
Yep. You need a power supply that provides a split supply. So all three: V+, GND, V-.
Tom
My adventures in HiFi continue… so I’m now using a PSAudio directstream into a pair of clone audio note Shinri 300b. The input sensitivity on these is ~250mV for full output.
The directstream can easily directly drive the amps to very loud levels, but you know what I still prefer going XLR into my Zen mod iron pumpkin, zero gain on the diamond buffers, then out SE RCA to the amps. Sounds much more dynamic and fully fleshed out. But it’s a bit of an overkill for acting as just a buffer, and I always prefer the simplest solution (and one less box).
So I have three ideas:
1. XLR into universal buffer and then short leg of rca into each of the monoblocks (maybe one universal buffer per amp, then I could have ultra short speaker cables). But is this best, or would it be better to drive the cable by placing the universal buffer at the source? PSAudio directstream output impedance is 100/200ohms. I could maybe even actively drive both ends, but my cable lengths are not that long.
2. Purchase some silver OCC cables and have no buffer, and see if the lower resistance gives the same additive effect of having a preamp/buffer in the chain. But I’m not sure how good DAC level output transformers are at driving cable/amps. Quasi-balanced XLR to RCA.
3. Install a passive input transformer in my monoblocks, to allow XLR direct… perhaps step down to give more current/punch? But transformer driven into transformer driven… hmm.
Anyone else tried these experiments?
The directstream can easily directly drive the amps to very loud levels, but you know what I still prefer going XLR into my Zen mod iron pumpkin, zero gain on the diamond buffers, then out SE RCA to the amps. Sounds much more dynamic and fully fleshed out. But it’s a bit of an overkill for acting as just a buffer, and I always prefer the simplest solution (and one less box).
So I have three ideas:
1. XLR into universal buffer and then short leg of rca into each of the monoblocks (maybe one universal buffer per amp, then I could have ultra short speaker cables). But is this best, or would it be better to drive the cable by placing the universal buffer at the source? PSAudio directstream output impedance is 100/200ohms. I could maybe even actively drive both ends, but my cable lengths are not that long.
2. Purchase some silver OCC cables and have no buffer, and see if the lower resistance gives the same additive effect of having a preamp/buffer in the chain. But I’m not sure how good DAC level output transformers are at driving cable/amps. Quasi-balanced XLR to RCA.
3. Install a passive input transformer in my monoblocks, to allow XLR direct… perhaps step down to give more current/punch? But transformer driven into transformer driven… hmm.
Anyone else tried these experiments?
Last edited:
I'm in a similar situation, Stretchneck. My pre-amp is the Benchmark HPA4, which has a set of RCA outputs, another set of XLRs, and an mono-XLR for a subwoofer. I was considering bi-amping my speakers (I have two many amps around here), and they're both single-ended, as is my subwoofer. I was able to make things work using Neurochrome's Universal Buffer. I fed XLR from the Benchmark over to the UB, then to three RCA output pairs for two amps and subwoofer. It worked really well.
@stretchneck
I have a colleague of mine who has a Universal Buffer in it’s own enclosure (set for +12dB gain) and we have done some of your experiments you have described. In one particular experiment, we used it in between a Holo Audio May KTE dac and the Modulus 686. As good as this combo sounded, it was a smidgen better, with a little more ‘drive’ using the UB. No real explanation for exactly why, given that it is an additional component (with an extra power supply, power cord and interconnect to boot!). Most likely there is a little more 2nd harmonic particularly in the midbass which gives the impression of a slightly more ‘riper’ bass. Whether that is preferred or not really depends on the system overall. The other explanation is that it was all in our heads given that the test wasn’t blind.
With regards to your 300B amplifiers, my inclination would be to NOT use an input transformer on the 300B. I would use the UB in there instead. The UB is truly quite transparent and if you already like the tonality of what you have and don’t want any tremendous changes then the UB is the ticket. Every time I have used transformers (even well regarded, wide bandwidth based, etc…) there is still a little 2nd harmonic left on the table. Not that 2nd harmonic is a bad thing. And this audible effect varies tremendously depending on the design of the transformer. This is where art meets science (as much as this offends some folks!).
Do realize that the maximum amount of fidelity, transparency and detail of the overall system is restricted by the choice of your DAC. In this particular case, I am assuming that you have the PS Audio DirectStream Mk 1. I have listened to this dac on numerous occasions in both my system and my colleagues’ system. Although the FPGA upgrades by Jim Smith were audible, the impediment is in the analog output stage particularly the output transformer which is a rather inexpensive version of an Edcor 1:1 signal transformer. The dac was designed in 2014 (and cost $6K at the time!). Modern modifications of this dac are numerous (just go to the PS Audio forum for details and AudioGon) and usually the first replacement is the output transformer. I would consider that if I were you. Of course if you have the Mk2 version, the transformer is already upgraded along with other improvements in the power supply and digital circuitry. Sadly the Mk 2 doesn’t measure that great but enthusiasts are pretty happy with it.
Of course, trying a UB on the output of the Mk1 version instead of using a signal transformer is a great idea but implementation is another story and I have no experience with that.
I realize the above discussion is a bit off topic, but in summary I would like to say that the UB is the most transparent buffer and is an excellent tool to have in your diy pocket for the finest system configurations. The price is an absolute bargain given the exacting measurements, guarantees and build quality.
Best,
Anand.
I have a colleague of mine who has a Universal Buffer in it’s own enclosure (set for +12dB gain) and we have done some of your experiments you have described. In one particular experiment, we used it in between a Holo Audio May KTE dac and the Modulus 686. As good as this combo sounded, it was a smidgen better, with a little more ‘drive’ using the UB. No real explanation for exactly why, given that it is an additional component (with an extra power supply, power cord and interconnect to boot!). Most likely there is a little more 2nd harmonic particularly in the midbass which gives the impression of a slightly more ‘riper’ bass. Whether that is preferred or not really depends on the system overall. The other explanation is that it was all in our heads given that the test wasn’t blind.
With regards to your 300B amplifiers, my inclination would be to NOT use an input transformer on the 300B. I would use the UB in there instead. The UB is truly quite transparent and if you already like the tonality of what you have and don’t want any tremendous changes then the UB is the ticket. Every time I have used transformers (even well regarded, wide bandwidth based, etc…) there is still a little 2nd harmonic left on the table. Not that 2nd harmonic is a bad thing. And this audible effect varies tremendously depending on the design of the transformer. This is where art meets science (as much as this offends some folks!).
Do realize that the maximum amount of fidelity, transparency and detail of the overall system is restricted by the choice of your DAC. In this particular case, I am assuming that you have the PS Audio DirectStream Mk 1. I have listened to this dac on numerous occasions in both my system and my colleagues’ system. Although the FPGA upgrades by Jim Smith were audible, the impediment is in the analog output stage particularly the output transformer which is a rather inexpensive version of an Edcor 1:1 signal transformer. The dac was designed in 2014 (and cost $6K at the time!). Modern modifications of this dac are numerous (just go to the PS Audio forum for details and AudioGon) and usually the first replacement is the output transformer. I would consider that if I were you. Of course if you have the Mk2 version, the transformer is already upgraded along with other improvements in the power supply and digital circuitry. Sadly the Mk 2 doesn’t measure that great but enthusiasts are pretty happy with it.
Of course, trying a UB on the output of the Mk1 version instead of using a signal transformer is a great idea but implementation is another story and I have no experience with that.
I realize the above discussion is a bit off topic, but in summary I would like to say that the UB is the most transparent buffer and is an excellent tool to have in your diy pocket for the finest system configurations. The price is an absolute bargain given the exacting measurements, guarantees and build quality.
Best,
Anand.
Of the three options, the Universal Buffer is the one that will give you the most bang for your buck. Input transformers offer great common-mode rejection ratio, but they are more likely to change the sonic character of your amp. The transformers by Jensen and CineMag are excellent, but even they add distortion that's at least 40-60 dB (100-1000x) higher than the Universal Buffer and a pair of transformers will cost more than the Universal Buffer. Silver wire does a wonderful job of draining your wallet in a hurry ... and that's about it.
In my experience it's better that the longest wire run is a differential link, so if your source has XLR outputs, I would integrate the Buffer into the amp. You could also, as you point out, have the Buffer in a box near the amp and just run a pair of short interconnects to the amp. If your amp has a balanced input as well you could even experiment with balanced vs unbalanced as the Universal Buffer has both types of outputs. See what you like best.
Tom
In my experience it's better that the longest wire run is a differential link, so if your source has XLR outputs, I would integrate the Buffer into the amp. You could also, as you point out, have the Buffer in a box near the amp and just run a pair of short interconnects to the amp. If your amp has a balanced input as well you could even experiment with balanced vs unbalanced as the Universal Buffer has both types of outputs. See what you like best.
Tom