Valve Itch phono

diyAudio Senior Member
Joined 2002
Hi,

Given the events sequence, the old transformer must have had a malfunction before finally failing. First the noise, the diodes blew, with new stronger diodes the Tx itself ultimately blew, with new Tx all good.

Whilst I understand the logic I doubt it's the xformer that caused the fault.
These are too simple. They either work or they don't.
IOW, if internal short they burn their windings. If wired incorrectltly they burn too or blow a fuse at best.

My advice , use the proper fuse rating and type.

Cheers, ;)
 
Member
Joined 2011
Paid Member
Silicon rectifiers usually internally breaks down the doped chisel layers into a short circuit when they fail. When running them on speclimit they get hot and one or more of the four diodes shorts out, half or the full of the AC being supplied would be shorted and heat up the tx to short its windings.

Allways fuse
 
Member
Joined 2011
Paid Member
Primary fusing is certainly mandatory and secondary fusing is healthy to do. When fusing secondary use DC fuses. You usually see DC fuses having a little spring on the melt wire. It is for pulling it apart when it melts because DC is prone to arcing otherwise and worst case form a weld.
 
The first stage from the FSP has 40dB or more gain before its passive RIAA network. It would need some modification to take the gain down by say 20dB resembling a 10X SUT function.

Would that be a problem? I'm quite positive about that I am not the one proposing the modifications :)

BTW: What tubes would one use as folowers? 40dB -> 20dB puts demands n the next level I presume...

Regards
 
Last edited:
diyAudio Chief Moderator
Joined 2002
Paid Member
Not a problem but a reminder it can not just be transferred as is. When 40dB+ of gain before an MM input is not usually needed at least. For a better noise result.
Gain arrangement and PSU voltages must be reviewed (to be at correct levels both for any different CCS load for the JFET and the common base PNP folded cascode's part too).
 
Not a problem but a reminder it can not just be transferred as is. When 40dB+ of gain before an MM input is not usually needed at least. For a better noise result.
Gain arrangement and PSU voltages must be reviewed (to be at correct levels both for any different CCS load for the JFET and the common base PNP folded cascode's part too).

Would you care to think about this and if you find it interesting enough propose a circuit? I would give it a go at once...

Sorry Salas- I forgot - Please :)

Regards
 
Last edited: