screen reisistor value

[Gingertube]

Tim,
Experience with some Menno VdV VDV70/100, the 100W version with a quad of EL34 and using ECC99 (in lieu of 12AU7) for the Commmon Cathode Gain stage + Cathodyne Splitter.

The oscillation was not the rail to rail thing.
I was using the "typical" 1kHz sine wave for testing. The oscillation was an approx. 900kHz burst just after the sinewave peak (i.e. just passed the turnover), if I had to take a WAG I would guess that it was at the point where one of the push pull sides was going into or comming out of cutoff. I think this was called "squegging".

150R screen resistors (each of the 4 tubes) was fine for Triode Mode and Pentode Mode but I had to take those screen resistors up in value for Ultralinear. I did'nt try in between values but just fitted the 1K recommended on the EL34 datasheet/application notes for Ultralinear. I passed this experience to Menno who confirmed this behaviour and posted an update on his website recommending increasing the screen resistors to 1K. Counter intuitively, it seems that the better the output tranny, the more these are needed. I was using PAT4006 Plitron torroidal OTs.

Was never too impressed with those amps, they needed feedback to reduce output impedance but there was'nt enough gain to do that. I ended up running the EL34s in Triode mode for around 40 Watts - they are on the shelf waiting for me to strip them down, those OTs are going to get a quad of 300B thrown at them instead (one of those on hold/one of these days jobs) - unless someone offers me cash for them.

Cheers,
Ian

[trobbins]

Ta Ian. My interest was peaked as I have an 807 PA to restore, and am presently restoring an 807 Williamson with screen regulated supplies. It came with 150 ohm anode stoppers under the nice Bakelite anode caps, and the screen and input grids have nearby stoppers, but I think I will slip some ferrite beads/tubes over the wiring at the valve base terminals. Being pentode mode with nice short leads to the Partridge WWFB OT, I doubt they had issues, but my soundcard cro would be none-the-wiser even if I looked!

[Merlinb]

A rather spectacular example of squegging in a Vox AC50, which would only happen while warming up and with particular control settings. I can't remember what the fundamental frequency was, probably a few hundred hertz.

[trobbins]

If your cro was 1ms/div, for nearly 200Hz fundamental, then about 8-9kHz. Would that be an OT output signal measurement Merlin?

[Leo_Gnardo]

A rather spectacular example of squegging in a Vox AC50, which would only happen while warming up and with particular control settings.

Squegging? I thought they were snivets. Well, parasitic oscillation anyway. Seems I got some vocabulary to learn. They do make the audio sound awful.

[tubeswell]

I've noticed that the 5E3 Deluxe doesn't have any screen resistors at all but the div/13 CJ11 - which has (almost) the same cathodyne PI dual 6V6 power amp - has 470r screen resistors.

If I could say anything intelligent about that I would...

In your typical 5E3 the screen voltage is about 50V lower than the plate voltage, so the screens aren't as likely to over-dissipate (as they would be if the screen voltage was the same as the plate voltage).

[Alf]

What exactly do high screen resistors ( 2.7K) in my 2 x el84 amp do for the sound . Do they cut off high content?
I used these because they were used in the poweramp schem I had.

B+ is 330 Volt.

Af

[cyclone]

If unbypassed they give you local NFB and nice compression. They keep the screens from running white hot too!

[Alf]

If unbypassed they give you local NFB and nice compression.

Thanks , that's the info I was looking for.

And I read somewhere recently that it makes your bass tighter. So in my case with 2.7 K in a PP poweramp with 2x EL84 if I change them to 2.2.K ( Kevin O'Connor's advice ) the amp will loosen up a bit ( and less compression) ?

Alf

[pdf64]

The effect of NFB that occurs at the screen grids is not to make the amp more linear or 'tighter'; rather I think it may add low order harmonics (due to the compression effect rounding off the waveform) and possibly increase output impedance.
A schematic is needed to give any useful help.
However, changing a screen grid resistor from 2k7 to 2k2 is unlikely to have much noticeable effect.

[g1]

I'm a little unsure about what "unbypassed" means when speaking of screen resistors?

[Chuck H]

I'm a little unsure about what "unbypassed" means when speaking of screen resistors?

I wondered about the subject of bypassing screen grids. I started this thread. Some interesting info that illuminates cyclones misconception on the matter. It seems that, like me, he speculated it would eliminate some local NFB and result in a performance change (possibly improvement). But I never dished out my theory as fact because I didn't know. I admitted my ignorance to the higher minds here instead so I wouldn't look stupid.

http://music-electronics-forum.com/t36289/

[cyclone]

I wondered about the subject of bypassing screen grids. I started this thread. Some interesting info that illuminates cyclones misconception on the matter. It seems that, like me, he speculated it would eliminate some local NFB and result in a performance change (possibly improvement). But I never dished out my theory as fact because I didn't know. I admitted my ignorance to the higher minds here instead so I wouldn't look stupid.

http://music-electronics-forum.com/t36289/

Dang, Chuck, thanks for calling me stupid I was speaking about pentodes in general, not necessarily power tubes. It is common to bypass the screen resistor on small signal pentodes. I guess I should have been more specific. Having high resistance in the screen circuit gives you some very useful local NFB. I would never bypass a screen resistor on a power tube because it then won't limit current like it is supposed to do.

[Chuck H]

Sorry. It was early. Sometimes I poke a little too hard at the ribs. I did look up the subject before and found some examples of preamp pentodes that do it. Considering the charge time it adds I'll wager that even WRT preamp pentodes it's best used as a means for bumping very small signals. I took this discussion to be about power tubes so I thought you made the same mistake as me and failed to fact check it. I should stop assuming others are as misguided as I am about this stuff.

[cyclone]

No big whup I hear it a lot, no offense taken. OP just said pentode and I think people often overlook or are unaware of the fact that the screen controls a pentode. Thats why there's so many opinions on what the value should be for a given app. Since the KF tube amp pages in the angela catalog when he insisted that it was necessary on a Marshall with modern siemens (or NOS Mullards, crappy soft vacuum inconsistent cheap tubes in my personal experience) to use a 1k 5watt, that has become the norm. He was, of course, absolutely right in that era and application, dudes with pointy guitars, 15kohm super distortion DiMarzios and tight pants running everything at 11, in a marshall. You can use a common screen resistor with small (100 ohm) grid stoppers on G2 and get good reliability and good performance without all that screen compression. Like Hiwatt did. They never blew tubes (or OPT's) like Marshalls (in my personal experience) Its the ringing on the screen that kills tubes dead, small grid stoppers help tremendously to suppress that. In my personal experience