power section fuses

[deci belle]

I read something recently here about fusing amps and merlin listed about 5 basic things to protect the amp.

One was fusing both primary wires from the PT to the rectifier.

Say my PT has 350v @ 200ma output; Should I use two 250ma fuses?

[tubeswell]

I read something recently here about fusing amps and merlin listed about 5 basic things to protect the amp.

One was fusing both primary wires from the PT to the rectifier.

The PT windings that supply the rectifier are secondary windings (just a small detail ;-) )

Say my PT has 350v @ 200ma output; Should I use two 250ma fuses?

Ideally you want the fuses to blow just at/slightly above the current draw rating of the winding.

R.G. did a good article on this a while back (see attached .pdf)

[trobbins]

Most old (and new ?) power transformers gave their HT output windings with a DC load current "rating", which unfortunately does not relate directly with fuse ratings for the winding(s).

For secondary HT fusing, the fuse rating depends on the type of filter - choke input or capacitor input. The fuse for choke input is pretty closely defined, because the rms current in the windings is closely related to the DC output current. But with capacitor input, the rms current through the fuse is a complex relation depending on circuit resistances and capacitances, and ss rectifiers introduce surge effects as well. One technique to determine fusing is to replace the fuse with a sense resistor and measure the rms sense voltage (ie. current) at full cranked output.

You can fuse the CT or each winding arm in a typical HT configuration. The rms current of a CT will be nominally root(2) times the rating of an arm fuse. I prefer to fuse the CT for simplicity, but I guess there are some failure modes for valve rectifiers where arm fusing would have an advantage.

[deci belle]

The PT windings that supply the rectifier are secondary windings (just a small detail ;-) )

haha!! There wasn't a face for "embarrass", so I still have to use the next best thing, I still qualify for that one, I think~
Thank you so much, Tubeswell!! I'm going to read that article now!!

You can fuse the CT or each winding arm in a typical HT configuration

Yes, trobbins!! I do have my CT fused now, but I want to learn the more complicated way next. Thank you for describing the equation between the CT and arm fuse relationships!! That helps because I know what is correct for CT fuse I'm using now.

Yay!!

I read R. G.'s article and understand that for my application (PT secondary wires to the rectifier), I must play the amp LONG and LOUD starting with low-value slow-blow type fuses and keep going up in value until they don't blow anymore!

[deci belle]

My next question is about fusing the power tubes' cathode. In my instance, with fixed bias, I have a bias checking resistor between each power tube's cathode pin and ground.

I would use a single fuse in series between them and ground as merlin stated as adequate. His website gives no information on what values to use for this method of fusing. The hypothetical values I've seen in my research are between 250~315ma— but I don't know what the application for that recommendation.

Has anyone actually fused their amp's cathodes in this way?

[trobbins]

I've put fuses in for high power amps with paralleled tubes where I want to save the output transformer more than anything else. You can estimate rms current level from the valves loadline - about 70% of peak anode current would be a reasonable lower limit for a fuse, so that if you chose the next standard fuse size up then that should give a reasonable margin on fuse rating and allow cranked 'square-wave' operation. A good idea is to parallel the fuse with a approx 10k resistor to allow the valve to go cold bias if the fuse blows, but not stress the cathode-heater if the tube isn't the problem.

[loudthud]

This thread will give you an idea of the current in an output tube of a 5F6A re-issue.

X-Y Plots of Output Tubes

[deci belle]

about 70% of peak anode current would be a reasonable lower limit for a fuse, so that if you chose the next standard fuse size up then that should give a reasonable margin on fuse rating and allow cranked 'square-wave' operation. A good idea is to parallel the fuse with a approx 10k resistor to allow the valve to go cold bias if the fuse blows, but not stress the cathode-heater if the tube isn't the problem.

Great link, loudthud!! Those traces are wild!

Wow!! You guys are great! Thank you trobbins and loudthud!!

[g-one]

Has anyone actually fused their amp's cathodes in this way?

Various manufacturers have done this, the first one that comes to mind is Marshall. Have a look at JCM900 schematics to see their method and fuse values. They even have a LED to show you the fuse has blown, in case you don't notice .

[deci belle]

Ya, I'd heard about something like that mentioned, only I didn't know which one-G!! I'm gonna study that right away~ Thank you for the telling me which model had that feature!!

I checked—
It showed the EL34 tube-pair's cathode lead to ground connected to a 500ma (type not stated) fuse in parallel with 100k R. The resistor is connected to an LED, but I can't read its label… "R/A 5ma(?)".

The schematic then shows the paralleled fuse and resistor/LED going to the ground-buss. Dang, I wish I could read what kind of LED, but it's too fuzzy to tell!

Thank you g-one— the values shown here for the fuse and resistor in parallel that trobbins had mentioned earlier helps!

[g-one]

The fuses Marshall uses here are definitely slow-blow type. The type of LED shouldn't matter, they are only in use when the fuse has blown. Whatever will light up with the 100K in series should be fine.

[deci belle]

Cool!! Thank you for filling in those blanks, g-one!!

[pdf64]

Sorry to muddy the waters but I don't see the need for slow blow fuses in the power tube cathodes; there's nothing there which has switch on surges etc, so better to use regular F quick blow, that will respond immediatley to tube shorts.
Also I found 100k resistors too high a value to light the LED; 10k was better.
Pete

[deci belle]

Hi Pete, could you please tell me which LED you used? That would be nice to know!

trobbins wrote:

One technique to determine fusing is to replace the fuse with a sense resistor and measure the rms sense voltage (ie. current) at full cranked output.

Mouser had some of those, which ones should I try, trobbins?

[pdf64]

They were likely a bog standard radiospares red 5mm type, possibly a low current variety.
It was on a vox v125 which I repaired and made a bit more robust a few years ago, so I used 4 of a 5 pack, and probably haven't keep the pack with the part #.
I don't think the LED spec is critical. I seem to remember that the EL34 cathodes sat at about 30V with the fuse open, but that may be a lot higher in the case of a tube with some kind of fault.
Pete.

[deci belle]

Thank you, Pete! 5mm LEDs… I guess the JCM900 schematic I was deciphering said 5mm (not 5ma) and I appreciate the voltage spec for the EL34 too!!

[deci belle]

This is where my 5E3 power section is at with PT 2ndary (Slo-Blo) and center tap (Fast) leads fused.

The OT primaries have the conjunctive filter between them. Chuck told me a big 10k R and a 1500pF 2kV cap would be about right for a 5E3.

Each flyback diode going to GND has a 470k R and .01uF 1kV cap in parallel so no single diode will get hit harder than any others in each string.

The rectifier has the safety diodes from the drawing on the yellow paper from another thread.

Right now I'm just using 500mA fuses so I can play the amp. I'm waiting for 250, 300, 375mA fuses to see how low I can go.

[Steve Conner]

The capacitors you're using to share voltage across your flyback diodes total up to about 1500pF across the OT primaries, the same again as your conjunctive filter, but without the series resistance. I think they might kill the high end somewhat.

[deci belle]

Ok, Steve!

To be honest, I don't even know how the caps and resistors do what they do (or how they hurt higher frequencies, for that matter)— it's just a mod I did to my bf Showman's SS rectifier.

I can just run those diodes off the plates by themselves then.

When I build the safety circuit for my bigger amp, I'll use the Gas Discharge Surge Suppressor instead of the conjunctive filter to try that out too. There are already wires under the board to eyelets hidden under the big 10k R for the surge suppressor for when I want to try that on the 5E3.

Is it advisable to use a conjunctive filter AND surge suppressor?

I'm also questioning the fused CT on the PT. Maybe I should rather fuse the OT's CT.

Anyway, thank you for the tip!!

[deci belle]

I made the fuse-board for my Showman on a pedestal using the choke's chassis mounting holes.

The tandem fuse-blocks are for the PT's 2ndary; the one on the upper right is the OT-CT; the pwr tubes' cathode fuse has that 10k 3w in parallel.

Hey you were right about the high-end getting squashed, Steve. I found out when I removed those caps and resistors from the diode-strings!



Thank you all for the help!