screen reisistor value

[Alan0354]

I can't speak for Ian but my 2 x 6V6 cathode biased amp just sounds better at 330V than it does at 380V.

Thanks, do you actually use an OT with slightly lower primary impedance or you just lower the +B and adjust the bias for the right current?

eg, if I lower the +B from 400 to 330, then I use a 3.4K OT instead of 4K for the 6L6.

[teemuk]

I did experiment with different value screen resistor on 6L6 all the way to 10K!!!! It just get looser....

Higher resistance = more voltage drop at the screen grid when the screen draws current.

Basically, when the tube saturates the screen begins to draw significant amounts of more current and at that point the screen voltage drops and consequently the tube's gain drops too. This effect primarily takes place on only one halfwave (because the other won't saturate the tube but drives it into cutoff where no current is drawn at all).

Since screen grid voltage is practically at least trying to follow the signal voltage the DC offset tries to find a midpoint between extremes of signal swing. If the signal swing is asymmetric, being higher towards one halfwave, (this happens when the screens begin to draw more current) the DC offset point shifts and sustained effect of this will gradually decrease the entire screen voltage. The effect has greater magnitude with higher screen resistor values due to higher voltage drops they generate.

If I remember correctly, up to about 2K, the difference in sound is minimal.

Yep, the differences between values like 470R - 2K, even 5K, are indeed quite small. At certain point in screen grid resistance increase the effect becomes more pronounced. This usually mandates larger than "usual" screen grid resistor values.

Someone can likely explain it better but the screen current draw seems to have a somewhat "logarithmic" behaviour. If the voltage sags so and so amount with screen R of 1K it won't sag linearly four times more with screen R of 4K. It sags less. Therefore differences with screen R values like 1K - 3K are subtle and you need much higher differences in resistances to develop pronounced differences in the magnitude of voltage sag at screens.

BTW, this only affects open loop gain of the tube and things like global negative feedback will more or less try to compensate the variation in open loop gain. Just to stress the importance of this all being very design-dependent behaviour / characteristic.

Novice question, I was led to believe since EL34 is pentodes, it draw a lot more current when tube saturates.

I'm not sure of exact details but true pentode tubes like EL34 will - due to differences in internal tube architecture - draw more screen current than beam power tetrodes (e.g. 6V6, 6L6...) even when they operate "normally".

[daz]

My schematics show 100 ohm screen resistors...

You're correct, but the reason i thought 47 ohm is because i was looking at a couple different schematics, and long story short, one calls them R35 and 36 but didn't have a parts list. I opened another schematic and searched the PARTS LIST not the schematic and 35,36 are listed as 47 ohm. But i also loed this time at the schematic in that PDF and unlike the one w/o a parts list the screes are labeled 24,25. Different versions.

[Enzo]

So what year is stamped on your circuit boards? That will tell us which schematic matches the amp.

[g1]

Because it was never intended to be overdriven.

The Blues Junior? I can't believe that. If so, it certainly demonstrates the disconnect between engineering and the real world. Do the designers have any knowledge of modern music whatsoever?

[Alan0354]

Judging from the channel switching circuit of the Super Sonic, engineers in Fender really don't know modern electronics.......modern....I mean from 80s or so modern!!! They use opamp to drive relays, really funky in the muting that at best mute one transition only. Still pop on the other transition.

I just looked at the datasheet of Mulard EL84, it's pentodes and draw a lot of current when saturates. I would use higher screen resistor.

[Merlinb]

The Blues Junior? I can't believe that.

Well, OK, then it was derived from previous Fender designs which weren't designed to be overdriven!

[Justin Thomas]

I can in no way argue the theory & facts of valve operation, but I can say that as far as we know, the only model of Fender with EL84s EVER was one short-lived model of Tremolux - 6G9. And the screen resistors are... nonexistent! The Blues Jr. came around in the 90s, and I'm pretty sure they intended it to be overdriven... I vaguely remember the ad copy. The Pro Jr. & Blus Jr. were for cranking. So in this case, I lean towards g-one's theory of "crap tubes necessitates protection of SOME kind..." They did have to survive the warranty period somehow. Otherwise, I think they're cheap enough. Make em sound good and in one instance where I agree with [NAME REDACTED], make them scream for holy mercy. I wouldn't do it with more expensive tubes. I myself slap in 100, but I'm just copying someone else... :P

Justin

[J M Fahey]

If so, it certainly demonstrates the disconnect between engineering and the real world. Do the designers have any knowledge of modern music whatsoever?

"Demonstrates" .... if it's true and proven ... which so far is not.

Until now it's just a personal opinion pulled from thin air.

1) it has a Master control .
I wonder why would they put one in an amp "not meant to be overdriven "

2) the Manual claims

adding warm, “Fender America Tube” distortion

3)

over 50 years of dedication and
experience that our engineers and countless
musicians have gained both in recording
studios and performing on stages around the
world.

4)

H. FAT SWITCH - When this button is
depressed, the amount of preamp gain
i n c reases providing extra distort i o n
and sustain.

5)

I. VOLUME - Adjusts the amount of
preamp gain and distortion/sustain.

etc. etc. etc.

Judging from the channel switching circuit of the Super Sonic, engineers in Fender really don't know modern electronics.......modern....I mean from 80s or so modern!!! They use opamp to drive relays,

Maybe they should use CD40xx or similar modern switches to handle tube signals?

Meaning up to 200/240V PP signals?
(Hint, max signal handling for standard Mos gates, "80's on technology" is 15V PP , with latest ones such as DG212 handling astonishing 30V PP)
Relays can do that all day long with both hands tied to their backs

Yet noobs dare to laugh at one of the most successful Amp Companies in the World

[Enzo]

So easy to get bogged down in these discussions. Overdriving an amp is one thing, overdriving its power tubes is another. The former is largely preamp tube related.

They have to survive warranty. The amp has a 5 year warranty, but tubes are only warranted 90 days.


As a Fender warranty center, my experience with Blues Juniors is the most common fix was replacing the reverb pan. After that it was replacing a bunch of loose tube sockets. I have surely replaced a bunch of tubes in them, but that is any model. I can't say I have noted any excessive screen related issues.

[Leo_Gnardo]

I can say that as far as we know, the only model of Fender with EL84s EVER was one short-lived model of Tremolux - 6G9. And the screen resistors are... nonexistent!

Screen grid R's nonexistent in just about ALL the early Fenders and other amps besides. Not to mention SG's run at full B+, what were they thinking? 1) save construction costs for bigger profits and better market competitiveness. So what, your output tubes went on the fritz again, skip lunch and buy a new pair at the mom n pop radio shop, there's one on every block. and B) our customers, they're just guitarists, what do they know about tone. As long as it's LOUD, that's enough. Back then, louder than a clock radio, was LOUD.

Wonder why there isn't a run on those EL84 6G9's, and/or why aren't they offered as a kit. Hmmm, don't mind the smoke, I'm getting an idea ...

[trobbins]

The Screen Stop function:
Tubes exhibit gain from G2 and so they can oscillate due to inadvertent feedback to G2 (from stray capacitances). 47 to 100 Ohms minimum is required for screen grid stop function. This is particularly important if running Ultralinear Mode. In a quad of EL34 Ultralinear Amp (HiFI Power Amp) I found that 150 Ohms screen grid stops were fine for Triode and Pentode Mode but in Ultralinear connection I had to increase these to 1K to supress parasitic oscillations.
Ian

Has anyone noticed what the frequency of the parasitic oscillation was on amps they've come across with that problem?

[teemuk]

Well, OK, then it was derived from previous Fender designs which weren't designed to be overdriven!

Plausible.

Most likely the particular Fender designs with
- very underatted power handling for screen resistors (they would heat up or toast from slight overdriving of the power tubes)
- non flameproof screen resistors (yep, they definitely failed spectacularly)
- located right beneath all kinds of wiring (yep, now the wiring was toasted too)

Just goes to show that guys at Fender weren't always almighty engineers.

On that note, isn't Fender still using standby switches rated for somewhat low current and AC voltage? (Insert facepalm) They have been "disconnected" on that regard for years and years. Moreso for even including such features. Sometimes it seems modern guitar amp "designing" is not so much about actual designing but copying without a second thought stuff that was conceived decades ago.

[Dave H]

Thanks, do you actually use an OT with slightly lower primary impedance or you just lower the +B and adjust the bias for the right current?

eg, if I lower the +B from 400 to 330, then I use a 3.4K OT instead of 4K for the 6L6.

I lowered the B+ to 330V with a 220 ohm resistor but I still used an 8k OT (data sheet value for 2 cathode biased 6V6s). It started out with a ‘guitar amp’ plate voltage of 380V but at 330V it’s still higher than the data sheet voltage. Interestingly the 6V6GTA data sheet has a load of 8K at 285V but at 250V it’s 10k!

I just used a 4k OT I already had in my 6L6/EL34 amp thinking it was close enough for rock and roll. It runs fine at 430V with either tube. It would probably produce more measured power into an 8 ohm resistor with EL34s and a 3.4k OT but I doubt it would be noticeable into the varying impedance of a real speaker and I don’t need more power. It’s too loud as it is. 15W is enough for me. I gigged for years with a 2 x EL84 amp.

[Enzo]

Yep, Fender, and most of the rest of the industry are using way underrated switches for standby. Fortunately the switches have not found out about this and have been working well the last 70 years.