|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
24.17.48.137
In Reply to: RE: Low Voltage to Heaters posted by mashley on July 18, 2015 at 16:45:19
You've picked the wrong diode for the job.
I would put the amp back to stock.
Follow Ups:
"You've picked the wrong diode for the job."
Why do you say that? Just from a quick look at the specs, I don't see a problem.
--------------------------
Buy Chinese. Bury freedom.
One could get less forward voltage drop from a diode rated for lower voltages.
There are also a fair number of tubes being heated by the device, as well as the cap input filter, so one must do a little head scratching at how hot those things are going to get at ~2A current. The 502 will pop up about 70C per Watt, so one would expect a temperature rise of more than 100C, which will simply cook the diodes where they sit.
A 1N5820 will exhibit just a hair more than half these losses.
The KBU will deal with the power much more gracefully, as the package will exhibit much lower temperature rise.
The OP has resolved the problem now, but I want to expand on what you've said. I still don't see a problem substituting the Schottky diodes."One could get less forward voltage drop from a diode rated for lower voltages."
The Schottky replacements exhibit a voltage drop roughly the same as the diodes in the original bridge. Lower drop diodes could be used, but this filament supply isn't regulated. It would seem beneficial not to install parts that might raise the voltage above that of the original design.
"The 502 will pop up about 70C per Watt, so one would expect a temperature rise of more than 100C, which will simply cook the diodes where they sit."
Unless I'm misinterpreting the modification, each bridge that was replaced is powering two 6SN7s. That's 1.2A. Each diode of the bridge sees less than that (average), but even using the full current for calculations, that's about 0.8W per diode. This will cause the VT5202 junction to rise to about 69C in free air if ambient is 25C. The junction is rated to 175C.
"A 1N5820 will exhibit just a hair more than half these losses."
The 1N5820 will drop 0.475V at it rating of 3A. That's a dissipation of 1.425W. The VT5202 Schottky datasheet shows a dissipation of about 1.95W at that current. That means the 1N5820 will dissipate roughly 73% of the power of the Schottky. That doesn't seem a very significant difference to me, particularly given that A) the bridge is operating well under 3A, so real dissipation differences are very small, and B) the VT5202 Schottkys will exhibit power dissipation virtually identical to the bridge they replaced.
If you think any of this is in error, let me know. I just don't see anything here that justifies telling the OP he selected the wrong diodes.
--------------------------
Buy Chinese. Bury freedom.
Edits: 07/24/15
CB, I Followed your advice and put the KBU4's back in there. As well as all the 1R resistors on the (low voltage/problem) amp.
The good amp is up to 12.57v after the last resistor so I've gained 0.5v.
The other amp, with the 1R resistors now(3) back in I'm only getting 9.5v?
I measured the current draw, good amp 0.6A, bad amp 0.5A (lifted the wire from the final resistor and used crocodile clips to run through my mm ( hope this was right?)
I had both amps up on the bench and tested every resistor in the linestage and the two caps and they all are correct as well as reading the same both sides, I cant check the 4700uf cap as my esr meter goes out of range.
I need some help here, have no idea what to do next?
(Just fyi am running film caps 100uf in the linestage, was advised this would be fine, removed lytics)I bypassed these with soviet silver mica's.
"I measured the current draw, good amp 0.6A, bad amp 0.5A (lifted the wire from the final resistor and used crocodile clips to run through my mm"
Are you unable to measure voltages on all the resistors like I suggested earlier? That's the way to get to the bottom of this problem quickly. Really, reinstalling the original diodes was a complete waste of time without making those measurements first. At this point, it appears one of the resistors ahead of the bridge in the bad channel might have increased in value. It's also possible the 4,700u in that channel is failing. Lacking an ESR meter, you could swap the caps and check whether the problem changes channels.
--------------------------
Buy Chinese. Bury freedom.
The resistors are all good, brand new mills MRA's tested!
That only leaves the 4700uf cap, I'll drop in another lytic in a few days just to see if the voltage returns to normal, if so I'll order "another" mundorf m-lytic, I didn't suspect this cap "too" much because I thought these m-lytics were supposed to have a very long lifespan/were bulletproof.
Yay, I hope this is it....
Will post results..
"You've picked the wrong diode for the job."
I bloody have? That is so disappointing!
Hmm ok then, I'm so done with this schottky attempt it's just kicked my ass one last time, back up on the bench tomorrow...
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: