|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
207.237.102.34
So I want to do the Hexfred mod to the Cary SLP98. First thing I see when I take apart the power supply is that they have 3 diodes in series on each leg of the half wave rectifier. (1N5408s) Are they doing this to get a voltage drop similar to two regular rectifiers? Do I need to put enough Hexfreds in series to get the same drop? Seems pretty persnickity as we're talking about one volt out of a couple hundred.
It looks like Cary wants you to buy four at a time. To put in series?
Thanks for any advice.
Regards,
Jerry
Follow Ups:
Ah - wait a minute I think the series diodes are for peak voltages expected (from positive cap charge to negative swing) 340VAC out of the xfrmr, 2 * sqrt(2) * 340 = 1000V. Still it is odd they used 3 1000V diodes in series. Hmmm...
It's not unreasonable to use diodes rated for 3X the AC voltage that they see from the transformer.
And yes, if you don't want to change the operating points, you would need to use 3 hexfreds. No big deal. But why bother with Hexfreds when Schottkys sound better? Or rather, Schottkys are more transparent.
Schottky diodes do have less OOB EHF harmonics, yes. but that's not in the audio band.
Warmest
Tim Bailey
Skeptical Measurer & Audio Scrounger
Tim, I cannot tell you "why", except for spouting the known facts about switching noise, etc. But for years I used hexfreds in my home brew or tweaked commercial gear, when tube rectifiers were impractical. Then when Schottkys became generally available, I switched to Schottkys, went back and changed over from Hexfreds to Schottkys in a few preamplifiers and in my big OTL amplifier monoblocks. I know there is a lot of self-deception built into this hobby, and I am very much a believer in listener bias, but all of that said, I hear something less in the way of an "electronic" sound, for want of a better word, with Schottkys vs Hexfreds. Plus, they are no more expensive than freds in most cases. The difference is subtle but real to me. Schottkys are much closer to tube rectifier sound (whatever that is), too.
I will at some stage get around to revisiting the PSUs in my two * 'Once Were' LEAK Stereo 20s which have high storage PSUs. I will be looking at putting in chokes and PP filter caps. CLCLCC or LCLCC if that's sensible, and get rid of the Rs that are there now except for one bleeder R, IF my techie insists on one.And, I will put Schottkys in to replace the (R&C) snubbed SR diodes that went in in the 1990s. They certainly are measurably and audibly quieter than anything else around.
For me, the limits that valve rectifiers impose on the amount of energy a PSU can store make them unnecessary.
The hole where the GZ34 was in the LEAKs now holds a 470uf 400V, dark brown, low ESR Nichicon cap. One of 4 per chassis.
The very positive effects on bass sound quality and headroom from a high storage PSU applies to all amplifiers.
Warmest
Tim Bailey
Skeptical Measurer & Audio Scrounger
Edits: 12/04/16
Thank you Lew.
In fact two 600V hexfreds are 3X. So that's what I ended up doing. This turns out to be a fantastic upgrade! Amazing detail, bass and spatial resolution improvement by replacing the three 1N5408 Diodes that were there.
Not sure what you mean by Schottky's being quiet. The Hexfreds are "soft recovery" and 18ns response. To my ears they qualify!
Have you compared Schottky's to Hexfreds? I'm sure the lower turn on voltage contributes to less switching noise, but the Hexfreds forward voltage was than a comparable Schottky (.48V @ 8A).
Anyhow - not sure about the Schottky comparison, but the Hexfred to the stock diode comparison is clear. The Hexfreds rule!
Regards,
Jerry
but please see my post above re that experience. In my stuff, I used IR Hexfreds, mostly. I think the Fairchild stuff or UF diodes may be different in some subtle way.
I added Hexfreds to a buddies SLI-300b. Thought one 1200V fred would be adequate. Lasted about 6 months before failure. Went to two in series with 4.7M voltage equalizing resistors (not sure if they are absolutely necessary) and it's been good for the last several years. I used DSE-160 (TO247) FREDs and mounted them back-to-back with a mica spacer, the two leads on one side were clipped, bent and soldered together. The resistors were soldered across the leads (parallel connection) and wire leads soldered to the two remaining leads. If you use TO220 devices be sure to use nylon hardware if you decide to mount back-to-back.
"It is better to remain silent and thought a fool, then speak and remove all doubt." A. Lincoln
Were the Hexfred diodes installed with snubber caps (usually ceramic) across them?
Yes indeed! They were 1600V 0.01uF polypropylene.
"It is better to remain silent and thought a fool, then speak and remove all doubt." A. Lincoln
It was fairly common that ham radio operators that built electronics used diodes rated 2.8x the secondary transformer voltage. It was written in ARRL Ham Radio books as a minimum safe design. Dennis Had of Cary is a ham radio operator.
How do they fail?
My experience too with other diodes. It's a hobby so I never cared to track it down and quantify but I suspect there are some really high voltage spikes that just break down the SS diodes over a relatively short period of time. Once that happens they just become shorts.
The safety margin needs to be much higher than simple calculations would predict. I was blowing 1200 V PIV parts on a ~ 450 DC B+ supply.
Mike
Short. Main PS fuse goes "poof" on turn-on.
"It is better to remain silent and thought a fool, then speak and remove all doubt." A. Lincoln
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: