Go Back   Home > Forums > > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Tubelab Discussion and support of Tubelab products, prototypes and experiments

Reply
 
Thread Tools Search this Thread
Old 15th April 2019, 01:57 PM   #1
OldHector is offline OldHector
diyAudio Member
 
Join Date: Jun 2018
Location: Täby, Sweden
Default Tubelab SPP with EL86/6CW5

I have just finished my 2nd SPP, this one using the EL86 variant.

After one teething problem - due to the difficulty reading the colours on those tiny 1/4 watt resistors (I am going to start to use 0.6 watt to make it easier) - I had a living, breathing amp, producing sweet music. The sound was comparable to my 6P14P (Russian EL84) one.

However after 20 minutes or so I noticed the tubes were starting to take on a orange-ish hue, the dreaded red plate.

Checking voltages, I see (with some variation between the tubes)
V(k) - 17v
V(a) - 210v
V(g1) - 0.02v
V(g2) - 211v

Voltage across the screen resistor (100 ohms) is 0.3v, so I(g2) = 3ma
Voltage across the cathode bias resistor (270 ohms) is 17v, so I(k) = 63ma

Plate voltage is (210 - 17) = 193v. Therefore dissipation is 193 * (63 - 3), which is 11.6W.

Assuming my maths is OK, and my rookie knowledge is not too lacking, then I suppose that dissipation figure is not so high, but I have read that cathode current should be restricted to 50ma.

Should I contemplate a 390 ohm bias resistor?

Out of interest - why is the bias resistor 5W? Is there a spike in voltage or current at some point?

And also, what is the benefit of having seperate resistors? Lots of PP designs using a common resistor, and a lot of the tube data assumes a single resistor in a PP setup.


I have two pairs of tubes, one is Philips NOS and the other is HQ, which looks like an old brand but I had not heard of before I bought them.

The HQ tubes are not as well balanced as the Philips ones (19.6 and 15.6 V(k)), and I suppose that is down to them not being a quality brand.

Does anyone ever fit a balancing trimmer from each of the bias resistors with the wiper going to ground?
  Reply With Quote
Old 18th April 2019, 09:25 PM   #2
Tubelab_com is offline Tubelab_com  United States
diyAudio Member
 
Tubelab_com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
Quote:
Out of interest - why is the bias resistor 5W?
In your example with the mismatched tubes you see 1.4 watts dissipated in the cathode resistor. Common engineering practice calls for a 2X safety factor on all resistors with respect to power rating, so technically a 3 watt resistor could suffice. This board is also used with 6BQ5 / EL84 tubes which need a 5 watt resistor, so the board must accommodate 5 watt resistors. The common 5 watt white ceramic and sand resistors are cheaper than a 3 watt resistor, so the choice is easy......put a 3 watt resistor there if you want.

Quote:
And also, what is the benefit of having seperate resistors? Lots of PP designs using a common resistor
one resistor and one cap for a push pull amp WAS common. That was done in the days when tubes were cheap, and caps were expensive, but it is a compromise. Much of the AC currents cancel, so a smaller value cap can be used. You won't find too many modern HiFi designs with a common resistor.

Quote:
Should I contemplate a 390 ohm bias resistor?
If the tubes are showing plate color you should reduce the current. Increasing the cathode resistor is the easiest way to do this. My guess is that a 330 ohm may be OK, but you might try both 330 and 390 ohm to find the best compromise between sound quality and tube life.

There are lots of different flavors of "6CW5's, and quite a bit of variability between them. I have used mostly GE 6CW5's in my SPP amps and the 270 ohm works OK. I tested some UL84's that only state "made in France" and they seem to want less current in a SPP. They were actually used in a guitar amp where they worked well with 340 volts on the plates and 170 on the screen grids. I was getting over 20 watts from them without any glow at all.
__________________
Tubelab, it's 5 year mission. To explore strange new tubes, to seek out new circuits and topologies, to boldly go where no tube has gone before......
  Reply With Quote
Old 18th April 2019, 10:09 PM   #3
OldHector is offline OldHector
diyAudio Member
 
Join Date: Jun 2018
Location: Täby, Sweden
Hi George, thanks for the explanations, which make perfect sense now I can see the design rationale.

I’ll experiment with the cathode bias resistor. Hoping i’ll be lucky enough to have some candidates knocking about, otherwise it will be a long wait until the spares shop opens again on Tuesday ;-/
  Reply With Quote
Old Today, 09:05 AM   #4
OldHector is offline OldHector
diyAudio Member
 
Join Date: Jun 2018
Location: Täby, Sweden
I have suspended this build because I don't think it is viable.

I have a power supply that provides a B+ of 211v, and I have OPTs that have ultralinear taps. The goal was to limit screen = plate at 200v, and run the amp in ultralinear config. However the tubes soon start to glow red in that setup, and now I think it is the screen that is overheating, and that the tube is partially running away and partially overheating due to a design limitation (screen voltage) being overstepped.

So I think there are two options from here if I retain the EL86.

1) Modify the power supply to look like The Red Light District supply, with a dedicated regulated screen supply. Then configure the setup to be pentode.

2) Reduce B+ to max 187v, which would also impact the driver and phase splitter supplies, so might necessitate some re-jigging there.

The alternative is to just put EL84's in the board, and source another power supply transformer with higher volts.

Another solution would be to find a tube that has an operating point of Vg2 = Vp = 200V (I.e a special EL86) - does anyone know a tube that has a normal specification like that?
  Reply With Quote
Old Today, 12:49 PM   #5
Tubelab_com is offline Tubelab_com  United States
diyAudio Member
 
Tubelab_com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
A true GE 6CW5 will run at 200 volts in fixed bias. The spec sheet actually shows 250 volts plate, and 200 volts screen with a 3000 ohm load for 25 watts power output in fixed bias push pull class AB1.

They show an idle current of 91 mA for two tubes at 250 volts which is 11.25 watts per tube in their data sheet. These tubes eat that without issue, but not all tubes are made that tough.

I was running GE 6CW5's in my board at about 230 volts plate and screen in cathode bias mode. That puts about 210 volts across the tube. All my tubes were pulled out of dead HP audio oscillators, so they were well used when I got them.
__________________
Tubelab, it's 5 year mission. To explore strange new tubes, to seek out new circuits and topologies, to boldly go where no tube has gone before......
  Reply With Quote

Reply


Tubelab SPP with EL86/6CW5Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
anyone ever use el86/6cw5? mctavish Tubes / Valves 43 23rd April 2019 02:03 PM
does anyone know of a el86/6cw5 quad amp ? rifraf50 Instruments and Amps 16 6th June 2016 09:54 PM
Subbing EL86 (6CW5) for EL84 andyjevans Tubes / Valves 7 15th April 2013 03:45 PM
EL86 / 6CW5 availability for Simple P-P? hartze11 Tubelab 1 27th May 2010 03:16 AM
Spice model for EL86/6CW5 genome Tubes / Valves 0 17th October 2005 02:49 PM


New To Site? Need Help?

All times are GMT. The time now is 04:52 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.79%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki