Screen resistor for BF VibroChamp?

[hcorneli]

Recently I've been running my '67 VC on wall voltage (~122VAC) with (I think typical) B+ ~400, plate ~390, screen ~395, cathode 23V on a measured 465Ω bias resistor. Sounds great despite -- or because of -- the 17W, ~122% dissipation (NOS RCA). But I'm trying to figure out if I should worry about the screen drawing more current under load and should install a screen resistor.


Reading old threads here and elsewhere, the advice even from folks I trust seems to vary a lot. It's a Champ, don't worry. No, the screens are at more risk in a Champ (1K first dropper, more distortion?). Go big. Go small. Go home. Etc.


Basically, I don't want to sweeten the distortion and I don't want to increase the bias resistance. I just want to protect the tube -- if it matters. I love the way it sounds now. I'm not redplating, and I've got the heaters dropped back to 6.4V. Is the amp eating 6V6s? No, but I haven't owned it forever, and most of the time I've had it I've used a sag resistor or bucking transformer to drop B+ into the range 360-370V. Oh, and I don't play heavily distorted on this little amp.


So would it protect the screens (the tube, the amp) to add a screen resistor? I test-clipped in a 470Ω 3W, and (ow) already thought the amp sounded a bit more polite (not good), though that could be my imagination. Suggestions?

[shooter]

as a tech I ask, "what's your goal for the amp?" 
keep it original?
sound as good as I play?
bust out test equipment n learn what all the magic is about?
get the pretty girl after the last set?


look up the max current for G2, use a 100 ohm resistor, for testing, measure voltage drop, select the resistors wattage based on 80% of G2 Imax,, walla, you smell the R cook'n, it saves a $17 tube for another day

[sluckey]

It's air cooled like a VW Beetle. The faster you go the cooler it is.   

[hcorneli]

as a tech I ask, "what's your goal for the amp?" 

Thanks, good question. Answer: 1) to sound as good as it does now while 2) protecting the 6V6 against screen current runaway. I appreciate the suggestions on how to figure it out analytically, but that's probably beyond me, and I'm concerned if I *could* do all the math I might easily defeat goal #1. Still, if you have a source for an NOS RCA 6V6 for $17, please share... :-)

It's air cooled like a VW Beetle. The faster you go the cooler it is.   

Thanks. Does that translate to "It's a Champ. Don't worry..."?

[jjasilli]

Another approach is a bucking tranny or variac to drop the wall voltage down to 110 - 115VAC (or down to 90VAC if you like).

[HotBluePlates]

Recently I've been running my '67 VC on wall voltage (~122VAC) with (I think typical) B+ ~400, plate ~390, screen ~395, cathode 23V on a measured 465Ω bias resistor. Sounds great despite -- or because of -- the 17W, ~122% dissipation  ...

I used to own a 1965 Vibro Champ.  I found that voltages throughout are higher than expected unless power from ~107vac as the outlet voltage.  Why?  Because it uses the same PT as the blackface Princeton & Princeton Reverb.  (Some folks will allege the Princeton has a "wimpy Champ PT" but the truth is the Champ uses a "too-large Princeton PT")

... B+ ~400, plate ~390, screen ~395, cathode 23V on a measured 465Ω bias resistor. ... the 17W, ~122% dissipation (NOS RCA). But I'm trying to figure out if I should ... install a screen resistor.  ...

Your RCA is probably fine, and could be run as-is for many years.

But if I were you & wanted to tame the amp's idle current I would replace the 1kΩ 1w resistor at the far left-end of the board (next to the can capacitor) with a 4.7kΩ to 10kΩ 3w resistor.  This will reduce the screen voltage, which will drop idle current & plate dissipation (without tinkering the cathode bias resistor).  Which value to use & how much reduction to shoot for is your judgment call.

[hcorneli]

Another approach is a bucking tranny or variac to drop the wall voltage down to 110 - 115VAC (or down to 90VAC if you like).

Thanks, yes, the bucking transformer I mentioned got it down to 115VAC, which cuts bias down to ~14.7W, or 105%, clearly a 'safer' place. But I recently noticed it seems to sound a little gutsier at 122VAC, which as noted puts bias at (oddly) 122%.

I used to own a 1965 Vibro Champ.  I found that voltages throughout are higher than expected unless power from ~107vac as the outlet voltage.  Why?  Because it uses the same PT as the blackface Princeton & Princeton Reverb.  (Some folks will allege the Princeton has a "wimpy Champ PT" but the truth is the Champ uses a "too-large Princeton PT")

... B+ ~400, plate ~390, screen ~395, cathode 23V on a measured 465Ω bias resistor. ... the 17W, ~122% dissipation (NOS RCA). But I'm trying to figure out if I should ... install a screen resistor.  ...

Your RCA is probably fine, and could be run as-is for many years.

But if I were you & wanted to tame the amp's idle current I would replace the 1kΩ 1w resistor at the far left-end of the board (next to the can capacitor) with a 4.7kΩ to 10kΩ 3w resistor.  This will reduce the screen voltage, which will drop idle current & plate dissipation (without tinkering the cathode bias resistor).  Which value to use & how much reduction to shoot for is your judgment call.

Thanks. I'm not actually aiming to tame the idle current or plate dissipation. You're right that these run hot -- everything I can read suggests the 410V B+ on the AB764 schematic was realistic. But here I'm more worried about screen current shooting up in overdrive, and though I can read Merlin's Single Ended chapter on why this happens, and how he picks a screen resistor for an EL34, I can't do all his graphical analysis for a 6V6, and for that matter, I'm not sure I really need one at all in this amp. That's why I'm asking.

[sluckey]

Quit fretting over it and just do it. There are two wires on pin 4 of the 6V6. Move both to pin 6. Put a 470Ω/1W resistor between pin 4 and pin 6. Let the resistor stand up over the pins about 1/2".

[tubeswell]

^What sluckey said^. 470R

[HotBluePlates]

... I'm not actually aiming to tame the idle current or plate dissipation. ... here I'm more worried about screen current shooting up in overdrive, and though I can read Merlin's Single Ended chapter on why this happens, and how he picks a screen resistor for an EL34, I can't do all his graphical analysis for a 6V6 ...

If you're concerned, go ahead & do what Sluckey said.

Be aware that EL34 and EL84 have very much higher screen current than 6V6 or 6L6, in any amp application.  That's because both are true-pentodes (suppressor grid instead of beam-forming plates), and their screen grid was not wound to "sit in the shadow of the control grid."  At both idle & overdrive, the screens run hot on EL34/EL84 and experience dangers that simply don't happen the same way to 6V6/6L6.  This is why you almost always see screen resistors for EL84/EL34 amps, and quite often don't see screen resistors for 6V6 amps (and earlier 6L6 amps).

[hcorneli]

Thanks, gentlemen, that's very helpful. I'll do it. One detail; I was thinking of leaving the jumper to V2b's plate right where it is on pin 4, downstream of the screen resistor, because I figure it doesn't need to idle at 394V either. Admittedly the 470Ω only knocks the idle screen voltage down a few points, but the trem seems to work fine either way. Yes/no?

[sluckey]

No! V2 needs to get it's B+ from a node with a filter cap.

[hcorneli]

No! V2 needs to get it's B+ from a node with a filter cap.

Thanks. I may be lost in the woods. I thought V2b's plate was fed off the screen node, which (I thought) has a filter cap. No doubt I'm missing something. I like your sig about including the schematic; lessee if I can post a pic here.

[sluckey]

No! V2 needs to get it's B+ from a node with a filter cap.

Thanks. I may be lost in the woods. I thought V2b's plate was fed off the screen node, which (I thought) has a filter cap.

Yes it is and it should stay that way.

[hcorneli]

Yes it is and it should stay that way.

OK, thanks, got it. I take it that putting the screen resistor upstream of the jumper interferes with the essence of being 'supplied by a filter cap'. RC vs. CR, something like that? Anyway, helpful.

[hcorneli]

... I'm not actually aiming to tame the idle current or plate dissipation. ... here I'm more worried about screen current shooting up in overdrive, and though I can read Merlin's Single Ended chapter on why this happens, and how he picks a screen resistor for an EL34, I can't do all his graphical analysis for a 6V6 ...

If you're concerned, go ahead & do what Sluckey said.

Be aware that EL34 and EL84 have very much higher screen current than 6V6 or 6L6, in any amp application.  That's because both are true-pentodes (suppressor grid instead of beam-forming plates), and their screen grid was not wound to "sit in the shadow of the control grid."  At both idle & overdrive, the screens run hot on EL34/EL84 and experience dangers that simply don't happen the same way to 6V6/6L6.  This is why you almost always see screen resistors for EL84/EL34 amps, and quite often don't see screen resistors for 6V6 amps (and earlier 6L6 amps).

By the way, I meant to thank you for the detailed (and clear) explanation of how and why. I like building amps, but I love learning stuff. So thanks!

[tubeswell]

FWIW, the filter cap at the screen supply node essentially 'bypasses' the screen supply resistor (i.e. a 'bypassed screen supply'), which keeps any B+ voltage fluctuations at the screen supply node to a minimum (the way a cathode bypass capacitor does for a cathode resistor). (or for that matter, every filter cap more-or-less bypasses the preceding B+ supply resistor.


Now if you add a further resistor between the screen supply node and the screen without adding another filter cap right at the screen grid pin (i.e. such as a screen grid resistor) , that part of the screen supply is unbypassed. Which means that a screen current feedback mechanism comes into effect - whereby as the screen current increases, the voltage across this (unbypassed) resistor also increases, reducing the overall voltage at the screen grid (which is the purpose of a screen grid resistor - i.e. to eat up excessive screen grid current). Now if you use a small-enough screen grid resistor here (e.g. between 100R to 1k), your ears won't notice much (if any) screen current feedback compression. But if you increase the screen grid to say 4k7 or 10k, you'd notice quite a bit of screen grid compression (which you may or may  not like) alongside loss of 'power' when playing hard.


Note that older champ and deluxe amps had a big screen supply dropper resistor (5k to 10k), which keeps the screen voltage sitting at about 50V below the plates. In this condition, a screen grid resistor isn't so important, because the biggest possible swings in plate current won't be big enough to cause excessive screen grid current - all other things (i.e. reflected load etc) being equal.

[hcorneli]

FWIW, the filter cap at the screen supply node essentially 'bypasses' the screen supply resistor (i.e. a 'bypassed screen supply'), which keeps any B+ voltage fluctuations at the screen supply node to a minimum (the way a cathode bypass capacitor does for a cathode resistor). (or for that matter, every filter cap more-or-less bypasses the preceding B+ supply resistor.

Thanks, tubeswell, that's extremely helpful. I'd read that the screen resistor could 'sweeten' the distortion, but had no idea how that worked or how upping the resistance would actually alter the sonics. Beyond that, you're giving me a broader view of how filter caps and resistors work together; I appreciate it.

[mresistor]

I like to think that running old Fender amps off of a variac where the voltage can be set where it was in the 50's and 60's is the best idea. If the amp is a stay at home noise maker then it's not a big deal, if you are carting it around from gig to gig then a variac can be a bit combersome but not at all unheard of with folks like Robben Ford and EVH having done it.


just my $0.02 out of a plugged nickel 

[dude]

I have a 1967 VC, tried many of the changes listed here. I found using dropping R's was much better in reducing the screens voltage and skip the screen grid resistor. Dissipation dropped, screens dropped and all did was up the 1st and 2nd node R's, now my screens are 50V less than the plates. I'm happy. 
We should call this the HBP mod for the Fender Champ. 

[hcorneli]

Yes, thanks. In researching this I've seen a boatload of different suggestions. Some folks go right to a big bias resistor -- not something I like in these amps. A bigger first dropper or even putting the screens on the second node are also suggested, but that makes me realize I didn't mention a third goal; I want good sound and safe operation with an amp *and sound* that are as reasonably close to vintage as possible.


So vintage voltage makes a lot of sense. I've found I don't need to drop below 115V for B+, bias, or heater voltage. Right now there's a decent chance I'll just go back on the bucker and call it good.


But first I plan to listen harder doing more careful clip-testing with and without a 470Ω screen resistor and on both 115 and 122 VAC.  If I somehow conclude 122V (122% MPD) does sound better, then maybe I'll feel better with the screen resistor there.

[sluckey]

Temporarily connecting a switch across the 470Ω will make it easy-peesy to compare sounds with versus without.

[dude]

Try a bigger cathode cap on the bias too, I think stock was something like 50uf...? Putting a higher cap there, like even 500 or better 1000uf at around (50v) will tighten up the bass considerably. Over at the old 18 watt forum that was the craze on the 18 watt cathode biased amps. It works great if you like a tighter bass on that little champ. That tiny OT, like 5 watts, IMO, is too small, saturates too soon but tone is in the eyes of the beholder. Have fun.

[hcorneli]

Temporarily connecting a switch across the 470Ω will make it easy-peesy to compare sounds with versus without.

Great idea, thanks, near-instantaneous comparison would really help. I can figure out switch ratings. But can I hot-switch that? do I expect a 'pop'? and do I need to account for moving that V2 jumper to Pin 6 during the test, or can that wait until I after/if decide to keep the screen resistor?