Mark Levinson rectifier

Dear friends
I have pair of mark Levinson 33H which I am very pleased by the sounds of those.
How ever since this monoblocks uses 2 bridges high amperage 70A diodes 70HF-HFR40 stud version. Those diode are standard
Not soft and not ultra fast.
Since they are not soft mark uses 1.5uf cap in parallel of each diode.
I was thinking to upgrade those to IXYS DCG100X1200NA
Silicone carbide schottky diode.
I know they have zero recovery time but they are not soft.
So snubber must be taken.
In the other hand mark Johnson wrote in this forum that even soft diodes have a ringing
ROD ELLIOTT that snubber wouldn't change a thing but 100nf+100R can help.
I have red articles that in 50/60hz PSU ultra fast or standard diodes is not an issue.
I have play around by changing diodes in an old amp and it does change the sound of an amp...
What would you do in the case of Levinson's and why ?
many thanks for everyone who comments here !


  • TST33JR.pdf
    18.1 KB · Views: 152
I would not sweat over these questions too much. Rectifiers sound different and what suits your taste is anyone's guess. Snubbering, to be effective has to match the particular power supply you have and in the absence of measurements is pretty much a shot in the dark.

So, if you want to spend time and effort, best is to test several different rectifiers, allow for break in and choose whatever suits your ears.

Otoh, sync rectification sounds dramatically better than even the fanciest rectifier. And it doesn't need snubbering. If your rail voltage is below 80v LT4320 based rectifier boards is all you need.

Have you already recapped? These amps need recapping much more urgently than a change of rectification.
Tnx for your answer
I got the caps last week nichicon NR
As for the rail is 80V
You are right one of the problem with this amp that it's need a vent on top to pull the heat from the inside which can be done easily.
The problem of testing diode bridge over this amp is that the bridge is hidden in the middle of the amp and each time is to disassemble it to it's boards:mad:
Power supply rectifiers have to handle switch-on surges, so unless the amp has a soft-start circuit the size of the transformer and filter caps tend to determine what the diode ratings must be, not so much the load.

SiC diodes have a larger forward voltage drop which will lose you some supply voltage.
Tnx mark
You got a point
Yes the amp has a soft start
The transformer is 3.5KVA
There are 2 diode bridges
Each bridge charge 2x 68000uf filter caps
Regarding the larger forward voltage drop
Maybe is worth to wait for the galium arsenid schottky diode
Which I not know it's character but it's seems the the future
Is going to those GaAs.
One of the problems is that the original diodes are stud version
Which are bolted straight to the copper bar
Unfortunately by looking all over the net I didn't find any brand that makes better stud diode that can replace the original
What are the parameters of the transformer of Mark Levinson NO.33H ?

NO.432 BYQ.jpg