EL84 screen R´s - 1k or 100R?

Express, Liverpool, Rocket, Dirty Little Monster, etc.

Moderators: pompeiisneaks, Colossal

User avatar
hans-jörg
Posts: 463
Joined: Mon Nov 01, 2010 10:11 am
Location: Vienna/Austria

EL84 screen R´s - 1k or 100R?

Post by hans-jörg »

Hello,

I see in the schematics of Liverpool and Rocket different screen R´s althoug mainly same power amp. Can somebody explain this diff. for me?

Thank you

Hans-Jörg
User avatar
M Fowler
Posts: 14017
Joined: Wed Jan 16, 2008 2:19 am
Location: Walcott ND

Re: EL84 screen R´s - 1k or 100R?

Post by M Fowler »

Yeah but the power supplies are completely different.

The LP is diode rectified and has higher HT voltages so using at least a 1k on the screens is necessary.

100R on the Rocket makes it sound great but not much protection and when I increase the HT voltage on some Rockets I increase the screen resistance.
Mark
Posts: 2957
Joined: Sat Jan 15, 2005 8:10 am
Location: Sydney Australia

Re: EL84 screen R´s - 1k or 100R?

Post by Mark »

100R on the Rocket makes it sound great but not much protection and when I increase the HT voltage on some Rockets I increase the screen resistance.
I'm in the process of looking at the screen grid resistors. What have you found?

There is no science to these observations, but I can't say I've noticed a tonal difference (yet) but I thought the amp broke up a little earlier.

I want to take measurement as well as set up a switch to allow me to A/B between 100 and 1K resistors.
Yours Sincerely

Mark Abbott
User avatar
martin manning
Posts: 13081
Joined: Sun Jul 06, 2008 12:43 am
Location: 39°06' N 84°30' W

Re: EL84 screen R´s - 1k or 100R?

Post by martin manning »

The Rocket has 15H of inductance between the screen and the plate nodes where the Liverpool has a 1k ohm resistance. 15H has reactance of almost 1k at 10Hz, nearly 10k at 100Hz, and nearly 100k at 1kHz. I suspect this helps significantly to keep the screens in check, making the larger screen resistors unnecessary.
Bob S
Posts: 1575
Joined: Sat Feb 14, 2009 2:38 pm
Location: Up there with the Michiganders

Re: EL84 screen R´s - 1k or 100R?

Post by Bob S »

That makes sense. The Rocket also has a lower B+ at the reservoir.
331vdc for a Liverpool & original Rockets were around 275vdc apparently. If your Rocket is running 300vdc or higher (like mine) the original design is hard on the output tubes . They do sound good on the hairy edge though.
Thanks Martin.
Why Aye Man
User avatar
rp
Posts: 2528
Joined: Sun Apr 25, 2010 4:21 am
Location: Italy

Re: EL84 screen R´s - 1k or 100R?

Post by rp »

martin manning wrote:The Rocket has 15H of inductance between the screen and the plate nodes where the Liverpool has a 1k ohm resistance. 15H has reactance of almost 1k at 10Hz, nearly 10k at 100Hz, and nearly 100k at 1kHz. I suspect this helps significantly to keep the screens in check, making the larger screen resistors unnecessary.
Great stuff Martin, I learn a lot on TAG.
Mark
Posts: 2957
Joined: Sat Jan 15, 2005 8:10 am
Location: Sydney Australia

Re: EL84 screen R´s - 1k or 100R?

Post by Mark »

Bob are you sure about the rail voltage?

330VDC seems correct for a Liverpool amp, but 275VDC seems a bit too low to me, I'd typically expect about 300VDC though source here quote 290VDC the rail voltage.

Martin, I want to run a signal into the amp and see what is happening at the screens when signal is applied. As noted in another thread I found the screens dissipating over 1 watt to a bit over 1.5 watts when no signal is applied. Typically I'd expect greater screen current when a signal is applied, though I don't know how much the rail sags in this state.

I'd appreciate any advice for my testing. My thoughts were to take static voltage measurements across all screen grid resistors to determine current flow, measure the voltage on the screen and calculate the screen dissipation. Next drag out the CRO and observe the voltage at the screens with reference to the static voltage. I'd then measure the AC voltage across the screen grid resistors again with reference to the static DC value.

I think measuring alone with the CRO should be enough, but it never hurts to have another source of data for comparison.
Yours Sincerely

Mark Abbott
User avatar
hans-jörg
Posts: 463
Joined: Mon Nov 01, 2010 10:11 am
Location: Vienna/Austria

Re: EL84 screen R´s - 1k or 100R?

Post by hans-jörg »

Thank you for discussion,

what I have in mind is a LP with 2xEL84 and EZ81 recti. Simple said it`s a 18W Marshall which I rewired to a LP. I want Keep the tube rectification. The PT is a 290-0-290 one. Between PT secondary and tube recti is a fuse (on both sides) 2x20 Ohm Lim R and 2x protection diodes .
The coupling C between 2nd and third stage is 5n (for less peak) and a 47k R with 100k trim (on place of the 150k to ground). The rest is puro LP. The Cathode R for Output is 2x330R10W, ergo 165 R 20W for both EL84, bypassed with 220µF.
Did I mentioned - the rest is puro LP? :)

So which screen Rs would you use (in this case)?

Best R. and thanks in advance

Hans-Jörg

BTW. this amp should act as the rhythm guitar amp for my daughters (garage) band. She`s the lead g. for shure with a ( RJ-second tube half booster boosted Rocket) :D
I´m so proud!
User avatar
rp
Posts: 2528
Joined: Sun Apr 25, 2010 4:21 am
Location: Italy

Re: EL84 screen R´s - 1k or 100R?

Post by rp »

I think Martin laid this one to rest, if no choke I'd go with 1K, w/ a choke I'd relax with 100R. I was always curious why KF used 100R in the rocket and 1k in the LP, looks like he knew his stuff too. :) As did the Dick Denny, as I always wondered why the used such a tiny value on amps known for burning up, whatever problem Voxes suffered it likely wasn't from too small 100r screen grid resistors.

Why the EZ81? Keeping a good part of the 'puro LP' seems important to you. From what I gather part of the LPs mojo is that it's a hard rocker like Express, fast and punchy, I'd keep the SS rectifier. Otherwise the EZ81 isn't a bad choice as you can use up to 50mf for the first cap. I'm not sure how much it drops, if it's a spongy tube like 5Y3 or fairly stiff like a 5AR4. I'm curious too, maybe someone knows?
User avatar
roberto
Posts: 1841
Joined: Tue Nov 27, 2007 4:45 pm
Location: Italy

Re: EL84 screen R´s - 1k or 100R?

Post by roberto »

We are around 15V @100 mA for the EZ81, while we are approximately at 45V for a 5Y3.
Bob S
Posts: 1575
Joined: Sat Feb 14, 2009 2:38 pm
Location: Up there with the Michiganders

Re: EL84 screen R´s - 1k or 100R?

Post by Bob S »

Hey Mark - I trolled around a little & found this posted back by Allyn.

PostPosted: Wed Jul 02, 2008 9:47 am Post subject: Reply with quote
Hi Guys, the choke is a Hammond 158L 75mA 15H like the originals. I will try to have someone do some clips. I don't have a way to do them here. Voltages are dead on for the Rocket I copied. B+ is very low (like the originals) around 270VDC-275VDC. 100V +/- 5 on V1 (like the original. I'll do a full voltage chart when I get a chance this weekend.

I remember thinking "what?" Seemed awful low to me too. But Allyn has been inside a few originals.
Why Aye Man
Mark
Posts: 2957
Joined: Sat Jan 15, 2005 8:10 am
Location: Sydney Australia

Re: EL84 screen R´s - 1k or 100R?

Post by Mark »

RP, check out this post.

Statically there is little difference between 100 ohms and 1K ohm screen grid resistors. You'll only know what is happening when a signal is present. For my own peace of mind I will have to take measurements and verify what is actually happening.

http://ampgarage.com/forum/viewtopic.ph ... creen+grid

Bob, I stand corrected, Allyn does say 275VDC, I'm wondering how the rail voltage gets so low with the stand power transformer?

I've tried two different rectifier tubes and three different sets of output tubes and the voltage has always been consistent.

Allyn also says: The tranny puts out around 510VAC to the rectifier (255-0-255vac). The amp runs really hot (bias) with a 50 Ohm Cathode resistor.

EDIT: I've turned the amp on and with the mains at 244vac I have 259-0-259vac which is consistent with Allyn's findings. A 40VDC difference is significant, I'm at a loss to explain this difference.

The transformer voltages aren't all that different from the specs of the transformer I'm using.

http://www.classictone.net/40-18065.pdf

I do have a 50 volt zener, I might have to try that. :shock:

http://ampgarage.com/forum/viewtopic.ph ... ght=rocket
Yours Sincerely

Mark Abbott
Gaz
Posts: 1145
Joined: Tue Aug 25, 2009 6:27 am

Re: EL84 screen R´s - 1k or 100R?

Post by Gaz »

You gotta just measure the screen dissipation to see for yourself what's safe or what sounds good. Don't just copy the Rocket, Liverpool, AC30, etc. if you are really concerned about tube life.

New tubes are so crappy, with their misaligned screen grids, that even safe voltage levels can over dissipate the screen. Higher screen grid resistance values sound subjectively worse because they low output power/headroom, increase sag and contribute to some local NFB (as shown by Andy Le Blanc in some older threads). I would argue that if you are going for headroom and punch, and 4xEL84 cathode biased amp is a bad starting place anyway, but people do seem to like the smaller (almost non-existant) 100R value because it retains a little bit of the big amp feel.

I personally like to keep the plates high (like 380vdc), then use a resistor in place of or in series with the choke to get the screens down to a safe level. I then add the 100R resistors on the screens. Of course you have to bias the tube appropriately if you do this. The high plate voltage (within reason) won't make those tubes flinch, it's really the over dissipation of the screen and under-biasing that will kill them.

JM2C.
Mark
Posts: 2957
Joined: Sat Jan 15, 2005 8:10 am
Location: Sydney Australia

Re: EL84 screen R´s - 1k or 100R?

Post by Mark »

EDIT: I've turned the amp on and with the mains at 244vac I have 259-0-259vac which is consistent with Allyn's findings. A 40VDC difference is significant, I'm at a loss to explain this difference.

EDIT 2: I wired in my 50 volt Zener, the amp currently has 1K ohm screen grid resistor and I used Fender Bassman box with Kendrick Blackframes, I do have an 18 watt amp with a Blue and a 12H30 anniversary speaker, but I find the the Kendrick Blackframes aren't as loud as the Celestions and the response is more even so I hear the amp better than the Celestions, the Celestions do have their charm, but I didn't think they were the right speaker for this exercise.

The guitar I used was my 82 1957 reissue Strat, it has a real nitro finish, Callaham bridge, and the original pickups were rewound by Slider with enamel wire. In short it is a very nice sounding guitar with plenty of vintage chime.

The mains voltage was pretty much spot on 240vac. The stock rail was 316VDC and when the Zener was engaged (yes it was on a switch) the rail voltage dropped to 266VDC. I noticed the amp had less top end and the mids were more apparent, I preferred the the stock voltage as the amp sounded more open and natural than with the lower rail voltage. I will also put the 100 ohm and the 1K ohm resistors on a switch to see their effect.

My feelings at the moment are anything from 290VDC through to 315VDC will sound good.


[img:1024:766]http://i149.photobucket.com/albums/s69/ ... 0ba08f.jpg[/img]

[IMG:1024:766]http://i149.photobucket.com/albums/s69/ ... 2c01a6.jpg[/img]
Last edited by Mark on Sun Sep 21, 2014 4:53 am, edited 1 time in total.
Yours Sincerely

Mark Abbott
Gaz
Posts: 1145
Joined: Tue Aug 25, 2009 6:27 am

Re: EL84 screen R´s - 1k or 100R?

Post by Gaz »

Yeah, you just "browned out" the amp a bit, which subdues the high end and makes the whole amp a bit more compressive and 'chewier' if you will. If you are just trying to listen for the effect of power tube voltages alone, then you would have to also put a switch on the preamp dropping string as well to keep the pre voltages in spec... What ever that is.
Post Reply