|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
206.255.208.188
In Reply to: RE: Ever auditioned posted by Bold Eagle on September 11, 2016 at 07:44:14
As in: why do two preamps that measure the same for frequency response and output impedance have noticeably different low bass with the same power amp and speakers?...If I knew why, maybe I could fix the weak one.Very good question. From my experience with both tinkering with existing designs and comparing related ones, I'd venture to say the difference lies with the power supply. FWIW, the Pass Labs Xs preamp uses 100,000 uf of capacitance. :)
I find that "conventional wisdom" as to how stiff one should be seems to be underestimated using sine wave testing. The very best sounding gear I've heard employs heroic sized ones. :)
edit: I think I might have previously relayed this story, but I think it bears repeating. A Frank Van Alstine modified FET-5 preamp replaced my Citation 11 in 1976. Along with upgrading the quality of the passive parts and using LF356 op amps in place of LM301s, he greatly beefed up the power supply. That was also the era of "double Dyna 400" where he not only increased the number of output devices, but significantly boosted the size of the power supply via an external cabinet. As an inquisitive teenager passionate about sound quality, I followed his lead with an Audire One power amp acquired about the same time. It ran 20,000 uf of capacitance using ~ 50V rails for 100 watt/channel output. Julius Siknius' designs were clean, but I wanted more. So, I replaced the bridge with a 30A unit and added another 60,000 uF in an external cabinet. That increased the energy from 25 joules to 100 - which coincidentally was the value of my later 1981 Threshold Stasis 3 acquisition with comparable wattage.
Sonic difference? Much improved bass punch and dynamics with my Magnepan MG-II speakers. And, the music played on for about thirty seconds following power down. :)
Edits: 09/11/16 09/11/16Follow Ups:
I've had similar improvements with beefed up; but less heroic power supplies. Adding extra caps in parallel increases the energy stored; but it also decreases the power supply impedance, allowing greater current flow rate. One of the design goals of the Adcom GTP-400, according to the designers, was to decrease the power supply impedance. The weaker sounding Rotel RTC940AX actually has 2x larger primary filter caps; but there is a lot of other circuitry between them and the output OP amps. The Adcom also uses OP amps; but there is nothing in the path between the filter caps and the OP amp. I may try adding some filter caps right at the output OP amp's power supply connections.
I also had a related experience with my Onkyo TX870 receiver, and an Onkyo A8190 integrated amp. The TX870 had very robust bass. It had a pair of 20,000 uF caps in the power supply, and in their literature, they touted the caps as having been selected for low ESR. The A8190 was from the same era; but used 12,000 uF caps and had an unusual "Real Phase Transformer" between the main power transformer and the rectifiers and caps. Both were nominally 100 W/ch; but the A8190 had noticeably less robust bass. So, on a hunch, I removed the Real Phase transformer, and the bass improved; but still not up to the TX870. Then, with the real phase transformer gone, there was room for another pair of caps of 8600 uF I had, so I put those in parallel with the 12,000 uF caps. Now the bass was quite close to the TX870. Next, I tried adding the real phase transformer to the TX870, and Poof!, the bass went wimpy.
The lesson I took from this little experiment was that the A8190's caps were too small, and lacked the special low ESR construction of the TX870's; and the extra windings of the real phase transformer added more series impedance to the power supply. Further, the added caps I put in were a low ESR type, so I don't know if it was the extra capacitance, or the low ESR, or both that did the trick. The NAD C350 I have also uses a pair of 22,000 uF caps; but again their literature talks about special low ESR types, and the C350 has a very solid low end, close to the bass power of the TX870.
Jerry
power supplies make a big difference in sound quality.
With my older Squeezebox Touch network player, I replaced the crappy SMPS with a linear (red box in pic) and enjoyed more resolution and clarity throughout the range.
With my current Sonore microRendu player (silver box at top center), I opted for a particularly stiff and quiet power supply for the same reason (black box at bottom right). Kinda dwarfs the player. :)
I am convinced that the overbuilt power supply is one of the reasons my Yamaha C2x sounds so good. This preamp is heavier than a few of my power amps!
Dave
When you think about it, devices that amplify are merely controls for the power supply. :)
Sure enough. EE courses used to actually teach that - I wonder if they still do? So much effort spent on signal path designs, and often the PS is neglected. Take a look at the Dynaco Stereo 70, for example. All kinds of mods to the PC board and different combinations of tubes; but the freak'n power transformer is undersized. That was also true of my Eico HF20 integrated amp. The power transformer got really hot in use; but that's another story.
Jerry
Take a look at the Dynaco Stereo 70, for example. All kinds of mods to the PC board and different combinations of tubes; but the freak'n power transformer is undersized.
FVA takes care of that today with his Ultravalve amp. :)
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: