Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Anyone has noticed a new power supply design from Tangent's website?
New Posts  All Forums:Forum Nav:

Anyone has noticed a new power supply design from Tangent's website?

post #1 of 17
6/16/08 at 8:09pm
Thread Starter 
His STEPS is gone, and YJPS he mentioned on the power supply test page appeared.

The first mentioned here, a long time ago.
Power Supply Tests


Recently he made this page, and STEPS page seems gone, and this appears...
Young-Jung Power Supply

It seems you can buy the board from his shop, but he says that there would be improvement over sometime.

Hmmm, has anyone ever build this? The hard part seems soldering surface mount parts.
Reply
post #2 of 17
6/16/08 at 8:13pm
I have. I'm glad he's making the change. There wasn't much difference between a STEPS and a TREAD. The YJPS is something else entirely. Jung super regulators have always been nice.
Reply
post #3 of 17
6/16/08 at 8:32pm
Thread Starter 
Good to hear.

given the transformer used, probably it would be difficult use this with my beta22, but probably great choice for PPA, M3, and other similar amps.
Reply
post #4 of 17
6/16/08 at 8:56pm
It's a single supply, so it's not meant to be used with a Beta22.

It does look pretty cool though, I'll definitely consider it for my next single supply (virtual ground) amp; thanks for the heads-up.
Reply
post #5 of 17
6/16/08 at 9:05pm
You should be able to build a negative rail version as well, but I'm not 100% sure. I would have to take a closer look. But that means 2x yjps.
Reply
post #6 of 17
6/16/08 at 9:17pm
Looks quite similar to the JSRxx regulators by Per-Anders Sjöström, e.g. JSR03. He's got a negative layout available as well.

Good to see this design up there. The LM317 is nice, but you can certainly do better than spending a ton of money on filter caps and end up with a nicer supply. This is the result, I suppose - it should cost about the same as a STEPS and be significantly better.

Edit: The transistors seem to be missing from the parts list page...
Reply
post #7 of 17
6/16/08 at 9:18pm
Thread Starter 
We need some more information I guess.....
Reply
post #8 of 17
6/16/08 at 10:59pm
From the schematic, the YJPS appears nearly identical to the unregulated portion of the STEPS supply. Why would this unregulated PS be better than the STEPS?
Reply
post #9 of 17
6/16/08 at 11:04pm
Quote:
Originally Posted by V-DiV View Post
From the schematic, the YJPS appears nearly identical to the unregulated portion of the STEPS supply. Why would this unregulated PS be better than the STEPS?
gotta look at page 2.

I have seen one of these, but didnt think it was going to go to production.
Reply
post #10 of 17
6/16/08 at 11:23pm
Quote:
Originally Posted by nikongod View Post
gotta look at page 2.
Duh! That's definitely more elaborate than STEPS.
Reply
post #11 of 17
6/17/08 at 4:10am
I'd be interested to see how this compares to AMB's σ11. I've only just started reading about the Jung Super Regulator, but the principle seems vaguely similar to my young electronics brain....... reference voltage, error amplifier, pass transistor.

The partially regulated input of the Jung seems like a very nice touch though.
Reply
post #12 of 17
6/17/08 at 4:49am
I'm just getting up to speed on analog stuff myself, so comparing the various designs (and tangent's excellent article on his YJPS) is very informative. As far as I can tell, σ22 is pretty much a Jung-Kan Super-Duper Regulator. It follows the basic Jung Super Regulator design, with some modifications/improvements. Instead of a monolithic opamp as the error amp, amb has designed a discrete implementation of one. He's added constant current sources in several spots for biasing to improve PSRR and reduce noise, and switched to MOSFET pass devices. Other than those changes, the overall design seems to be that of the Jung regulator. I could be grossly wrong in this analysis, so take it with a fistful of salt.

It's hard to say which design will win out for overall performance. My guesswork makes me think amb's design is probably better where larger currents are involved, while the YJPS may be slightly lower noise, but can't cope with the same amount of current either constant or peak, and may have slightly worse load regulation.
Reply
post #13 of 17
6/17/08 at 8:22am
PCB ordered! I'll use this in my yet to be built PPA.
Reply
post #14 of 17
1/8/09 at 12:53pm
Anybody have more information, analysis, or experience regarding the YJPS? I’m most interested in the desirability of using this design with a M^3 amplifier considering the following cautionary except from amb. Thanks in advance.

Excerpt from The M³ Stereo Headphone Amplifier
The Tangent STEPS power supply, which was formerly listed as a usable alternative, has been discontinued. Its successor, the YJPS, has not been tested with the M³ and their compatibility has not yet been established.

Do not use an unregulated power supply. Not only will an unregulated supply degrade performance (such as increased hum and noise), the lack of voltage regulation may cause damage to the opamps or other parts if the actual voltage exceeds the maximum ratings of the components.

Also, do not use a power supply that has its V- output connected to the "earth" pin of the AC plug. This is because V- is used as the negative DC rail on the M³ amplifier, and the signal ground is synthesized by an onboard TLE2426 rail splitter chip. If you use this kind of power supply, and any other component in your audio chain has its signal ground connected to the AC earth, then you will cause a short circuit from the V- to signal ground.

The power switch should always be on the AC mains, not between the PSU's DC output and the amplifier. This is due to the high capacitance across the M³'s power rails, which can appear similar to a dead short to the PSU momentarily while the PSU attempts to charge the capacitors. If the power switch is at the output of the PSU, when it is turned on, the sudden current surge could damage the PSU. This is especially important for the σ11 PSU, which has no current-limiting protection. Even LM317-based PSUs (with over-current protection) have been known to be damaged in this manner.

Always measure and verify the actual output voltage of the power supply before connecting it to the amplifier for the first time, to make sure its output voltage is as expected. Do not connect the PSU to the amplifier when the PSU power is already on.
Reply
post #15 of 17
1/8/09 at 1:15pm
Quote:
Originally Posted by error401 View Post
As far as I can tell, σ22 is pretty much a Jung-Kan Super-Duper Regulator. It follows the basic Jung Super Regulator design, with some modifications/improvements.
So many modifications it cannot be said to be similar anymore. The sigma22 is optimized for high currents, the Jung is not.

- In a Jung super reg, the error amp is powered from the regulated output, not from the input.
- The sigma uses a zener and not a low-noise voltage reference as Jung's.
- In a Jung regulator, the pass transistor base is fed by a constant current source. The error opamp doesn't have to provide the bias current and operate always as a sink. This provide current limiting, which is a good thing.
- A full blown Jung allows remote sensing, while the sigma pcb makes no provision for it.

Walt Jung has made clear many times that he didn't want his name associated with anything he didn't personnally designed. Which means any design where parts were substituted or the original layout altered. (not unless it is made clear of course, as Tangent did).
Reply
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Anyone has noticed a new power supply design from Tangent's website?