Question about rect tube 5Y3

[catnine]

I had a AA764 , 73 SF champ that had a 40-20-20uf can cap and replaced it with the same .

I later build the same amp circuit using a F&T axial caps at 47uf , 22uf & 22 uf .

Now I find out that you can cause an arc or short a 5Y3 with a surge charging anything over a 20 uf first filter cap.

So far during a few componant changes I have drained the cap s at least 10 times and on power up with the 5Y3 eye level never saw arcing which I assume I should have .

I never knew about the 20uf /5Y3 spec . The bronco fender schematics show a 40-20-20 uf can and the 6G2 shows a 30-30-30 uf can cap both with a 5Y3 .

Any suggestions? Do I need to change the cap value or is there a way around this like a choke between the 5Y3 B+ and the 47uf cap .

[Ronsonic]

Mostly you can just get away with anything under 50uF.

If you want to be kind to the tube, just swap some leads around so the first cap is a 22, and the one after the resistor is the 40.

[SatelliteAmps]

The RCA tube manual states a value of 20uf as the first cap as being the high value. Higher values can be used, to a degree. Some amps put more stress on a rectifier tube than a Champ or Bronco circuit, and it shouldn't be done with those amps. In this case, it should be fine.

And you are correct about the schematics. The AA764 should have a 20uf as the first cap to be schematically correct. The Bronco has a 40uf as the first cap, and has no problems running it.

All of that being said, I wouldn't go any higher.

[SatelliteAmps]

Excellent suggestion.

[catnine]

There is actually a way to improve that situation. It will cost about 95 cents and 10 minutes.

Add a pair of diodes to the HV leads just before the rectifier tube as shown below in rough schematic and layout drawing.

Fender uses the BYD33V (http://radiocomplect.com/nvss/download/476.pdf) (shown below) In there Deluxe RI to help prevent rectifier arcs when flipping the standby switch.
These are Fast Soft Recovery diodes.


Using UF4007 in this same topology is very effective at preventing arc-over in the 5AR4 and other rectifiers with close cathode to plate spacing. I have not had a single rectifier tube failure due to arc over since I started doing this.

Special diodes don't seem to be required for this purpose. I have not had a problem with anything rated at least 1KPIV or above. (1N4007/UF4007, and of other types )

It also helps to use a good quality NOS rectifier, but either way, the diode setup is a good solution.



http://home.mchsi.com/~my_mcintosh_240/diodes.jpg

http://home.mchsi.com/~my_mcintosh_240/diodes2.jpg


If I use these 1N5408/1000V 3 amp diodes will these be ok? Just 3 amp versions of the 1N4007 's .

I don't understand how this works with the diodes as far as stopping the arcing . Does it slow the voltage surge from the PT secondaries to the rectifier tube ?

[catnine]

1N5408/1000V 3 amp should work ok, I generally use the fast soft recovery types but as mentioned above most other types work just fine.

As far as surge goes, you should really not be having much of a surge issue unless you have a standby switch installed. 47uf is indeed a bit on the high side for the 1st cap with a 5Y3 but I have seen other guys do it in the past often with mixed results.

You are suppressing transients more than actual operating voltage surges.

Some rectifiers are more prone to internal arcs than others, I once had some old GE's that would arc with a 16uf cap, The diode trick cured it. Most likely those GE's were not to proper spec and could have been rejects.

I do have a standby switch . I have a westinghouse NOS 5Y3 in this amp . Perhaps it might be best if I lower the value of this first cap to the 22uf as the other two are of at least to a 30uf hoping to retain the bass response the 47uf allows.

I don't see any harm adding the diodes .

If i power on and flip the standby to play watching the 5Y3 I see what looks like a faint blue eye blink flash even after the amp was played for an hour and off for an hour , I don't know if that's arcing or normal but the caps should be still charged after sitting off for only an hour.