Rectifier backup diode mod on amps other than 5E3 ?

[DavidP]

So, I'm going under the hood of my home-brew 5Y3 to put in those rectifier backup diodes (one of RobRob's preventive maintenance items). Now, I'm thinking, if this is such good preventive maintenance, why not do it for my SFPR and SFDR, both running 5U4s.
So... inquiring minds have toask:
Why is this mod usually mentioned only with the 5E3??
If applicable beyond the 5E3, is the 1N4007 diode sufficient to handle the higher B+ seen on on these silverfaces?
Cheers, D.

 

[Finck]

You can use SS diodes in any amplifier, just take into account that the final voltage will be very higher than what you get with the tube rectifier. You have to compensate that raising/putting resistor(s) in the filter.

1N4007 handles up to 1,000V/1A, so it will be fine.

 

[Bill Moore]

Isn't the IN4007 rated for 1KV?

 

[DavidP]

Original post edited to be explicit that I'm referring to backup diodes, not switching to full SS rectification. Sorry 'bout that!

 

[Finck]

1,000V = 1kV...

What is a "backup diode"? Put a second diode in paralell with the original to have no problem if the last opens? If yes, go ahead.

 

[Andy B]

I put them in vintage amps all the time, especially when using modern rectifier tubes. I use 1N4007’s.

 

[tubeswell]

I put them in my amps as a matter of course - they are a cheap no-brainer that prevents expensive collateral damage to your power supply occurring from a shorted rectifier tube. (Incidentally, the idea was around long before robrob. A chap by the name of R.G. Keen did a series of articles in Guitar Player Magazine some years back called 'Immortal Amp mods' which featured these, alongside a number of other useful robust mods. )

 

[Wyatt]

1,000V = 1kV...

What is a "backup diode"? Put a second diode in paralell with the original to have no problem if the last opens? If yes, go ahead.

No, the diodes are placed before the rectifier tube in series. Voltage drop and sag performance is the same as if using tube alone. This is to protect the PT in case of a tube rectifier short.

Instead of me doing a hack job explaining it, I'll let the expert...

https://www.premierguitar.com/articles/the-immortal-amplifier-mod-1

 

[DavidP]

Thanks Wyatt for the backup explanation -- was about to do it but you covered me well!!
My concern is what RG notes at the end of that "Immortal" article:
"...diodes have to be able to withstand at least twice the B+ voltage in the amplifier"
and that would hit/exceed the rated max of a 1N4007.

 

[mherrcat]

I read that article on the Permier Guitar site and noted this (my underlining):

"But if it shorts, the solid-state diodes now take over the load. In fact, the amp simply keeps playing if the rectifier tube shorts. The high voltage supply actually rises by maybe 50V, so the amp gets a touch louder. You can keep playing until the show is over and you have time to put in a new rectifier."

It kind of begs the question, "If the amp continues to work, how do you know the rectifier tube has failed?"

 

[Wyatt]

My concern is what RG notes at the end of that "Immortal" article:
"...diodes have to be able to withstand at least twice the B+ voltage in the amplifier"
and that would hit/exceed the rated max of a 1N4007.

In a Silverface Princeton Reverb and Deluxe Reverb the B+ should be 410-425VDC, well below 1/2 of 1KV.

But elsewhere in the article...

For amps up to about 450V in normal operation, you can use the cheap and widely available 1N4007, which is rated for 1000V and 1 amp. For amps with B+ supplies over 450V up to about 550V, use the ST Semiconductor STTH112U, which is rated for 1A and 1200V, available from Mouser Electronics.

 

[tubeswell]

If the 1N4007’s 1kV PIV rating isn’t big enough for your amp, then squeak 2 or more diodes in series - it’s what Leo did back in the day. Otherwise put in 1.5kV or 2kV diodes.

 

[robrob]

Incidentally, the idea was around long before robrob. A chap by the name of R.G. Keen did a series of articles in Guitar Player Magazine some years back called 'Immortal Amp mods' which featured these, alongside a number of other useful robust mods.

From my website:

Protection Mods

These amp protection mods were detailed in a series of articles in Premier Guitar by R.G. Keen.

 

[robrob]

The solid state diode and the tube rectifier are in series and work like two solid state diodes in series virtually doubling the voltage handling to 2000 peak volts. So a 1N4007 is fine as a backup diode.

 

[DavidP]

Thanks tubeswell & robrob!! Time to 'open up the hood' this weekend!

 

[charlie chitlin]

Excuse my ignorance (as usual) but I thought the backup diodes were just to keep you u and running if the tube fails.
Wouldn't a properly rated fuse blow if the tube fails?

 

[D'tar]

Charlie,

The backup diodes are on the HT AC input of the rectifier and protect the transformer in the event of rect tube failure

From Robs' site

Installing two 'backup' diodes can extend the life of your rectifier tube and prevent a dying rectifier tube from frying your power transformer or amplifier circuit with high voltage AC. The backup diodes pre-rectify high voltage AC from the power transformer into pulsing DC and feed it to the tube rectifier which will power the amp and give you that loved tube rectifier voltage sag. The 20 cent 1N4007 diode has a 1000v peak-to-peak or 353v RMS peak inverse voltage (PIV) rating. The 5Y3 rectifier tube is rated at 1400vpp PIV or 495v RMS on each plate. Having two diodes in series (one solid state and one tube) will increase the rectifier circuit's peak inverse voltage to 848v RMS which should greatly increase the rectifier tube's lifespan. You're rectifier tube should outlive your amp with these backup diodes sharing the load. Here's a quote from the Valve Wizard's web page:


If we were worried about running a valve rectifier too close to its maximum Va(rms) [AC voltage rating], we could place one or more silicon diodes having a much higher peak inverse voltage in series with each anode, as protection elements. In theory, the valve rectifier can then be used with supply voltages up to twice the rated Va(rms). The silicon diodes will have no adverse effects on the normal operation of the valve rectifier.

I installed two 1N4007 diodes on the rectifier socket pins 3 to 4 and pins 5 to 6. Place the diode's stripe (cathode end) on the power outflow side (pins 4 and 6). I attached the power transformer high voltage output wires to the unused pins 3 and 5 (instead of the normal 4 and 6). The AC power then flows through the diodes to the plate pins 4 and 6. The diodes are cheap insurance and I install them in all my amps.

The stripe on the diode indicates polarity.

 

[Wyatt]

Excuse my ignorance (as usual) but I thought the backup diodes were just to keep you u and running if the tube fails.
Wouldn't a properly rated fuse blow if the tube fails?

That is neither the purpose nor how they work.

The diodes are there to protect the filter caps (and rest of the power supply) if the rectifier tube shorts AC to its cathode. In that case, yes, the amp will continue to work thanks to the diodes and "closed" tube conmnection. A fuse would not blow until the damage is done.

But typically rectifier tubes "open" when they fail. In that case, the diodes will NOT keep the amp working most of the time. The diodes are in series with and before the tube, so if the tube doesn't pass voltage, the amp won't work. That's not what they are there for.

 

[King Fan]

Gotta say. Threads like this are the glory of TDPRI. Expertise flowing and joining from so many sources. Thanks, pros.

 

[charlie chitlin]

This seems like a worthwile update even, maybe especially, on a vintage amp.
I think I would risk having 4 altered solder points to save valuable parts.
I know Gerald Weber does this on many of his builds.