Silicon Bridge Rectifier

[Lectroid]

I once bought some small components off one of our senior members here, and he sent my parts promptly, along with a double handful of these RS405L bridge rectifiers in the box too.  Kind of like packing peanuts.  It's designed to go onto a PCBs; a data sheet is attached.

1.  Could one of these replace a hand-built bridge made up of four 1N4007s in an amp with ~400V of B+?

2.  Has anyone used these this way in a tube amp?  If so, any tips on how to build it into the circuit?

Thanks!
 

[shooter]

while it may work  the 600VDC PIV is a little low, I typically want a diode at 1,000V PIV


design into a new board 4 turrets would be best, build your own "proto-board" that could hold it, glue it to the chassis....

[SEL49]

Dummyload used them for packing. Guess why. 

[acheld]

They won't have the voltage rating of 1N4007s, as the 405 is rated at 600V, but with a max average current of 4A.

So yes, they will work in a lower voltage amp with a lower margin of safety in the voltage rating but much greater amperage.

I have used this style of rectifier a number of times and like them very much -- they are small, and you can get very beefy specs for not too much.  One thing to look out for -- some of these need a heatsink to achieve their max current rating.   This one from Rectron does not.

And yes, the turret board will be a different design . . .

[Lectroid]

Thanks everyone.  I've been reading the usual suspects re: rectifiers, and you've helped cement several things that were still kind of fuzzy.

@acheld,
The build is a Hoffman 5E3 with a MV.  I'm using a Classictone 40-18021 PT (350-0-350) which has a HT center tap. The 6.3 current needed adds up to 1.5A, and the winding is rated to output 1.65A.  I don't want to lose any of that in rectification. 

So.  I think this all means that my best bet for a silicon rectifier is to use a two-phase rectifier like on the AB763 Twin Reverb, with multiple 1N4007 diodes on each leg to share the load.  For a 5E3, two diodes on each leg might be okay?   

[acheld]

The rectifier you were asking about is a bridge rectifier, as opposed to your two-phase rectifier.  They get you roughly the same performance.  (See https://www.valvewizard.co.uk/bridge.html).

With regard to using 2 1N4007s per leg, that's fine.  However, to "share" the load on the diodes, you need 10-100nF 1KV(or greater) ceramic caps in parallel.

[Lectroid]

Great!  Many thanks.   

[trobbins]

Imho, buy a strip/batch of new 1N4007 from a reputable manufacturer and solder two in series and close to each other, and don't use any parallel balancing caps.  1N4007 have a significant PIV margin above 1kV, and if two are in series and at the same temp then there is no practical need for balancing caps in this application - especially as the caps themselves impose a risk.

With respect to the RS405L, if the application used 1A diodes, then preferable use 1A diodes, as that minimises the chance that diode related glitches could extend out into your audio signal path if the transformer and layout allow.

[Lectroid]

trobbins,
That sounds like the smartest way to go, maybe two in series on each leg.  When passive parts are so cheap, I tend towards overkill.  Thanks!

[trobbins]

Two in series is not overkill imho, more related to applying reasonable margins for aspects like main voltage variation, and two parts in series.

[Merlin]

The RS405L has a Vrrm rating of 600V, so it can be used with a transformer voltage up to 350Vac (single winding), or 350Vac centre tapped. This allows some safety margin for wall voltage variations and transformer regulation.
For a single winding (i.e. bridge rectifier) this will indeed give you more than 400Vdc safely. But for a centre-tapped winding you only get half as much DC.