If you need assistance, please send an email to forum at 4hv dot org. To ensure your email is not marked as spam, please include the phrase "4hv help" in the subject line. You can also find assistance via IRC, at irc.shadowworld.net, room #hvcomm.
Support 4hv.org!
Donate:
4hv.org is hosted on a dedicated server. Unfortunately, this server costs and we rely on the help of site members to keep 4hv.org running. Please consider donating. We will place your name on the thanks list and you'll be helping to keep 4hv.org alive and free for everyone. Members whose names appear in red bold have donated recently. Green bold denotes those who have recently donated to keep the server carbon neutral.
Special Thanks To:
Aaron Holmes
Aaron Wheeler
Adam Horden
Alan Scrimgeour
Andre
Andrew Haynes
Anonymous000
asabase
Austin Weil
barney
Barry
Bert Hickman
Bill Kukowski
Blitzorn
Brandon Paradelas
Bruce Bowling
BubeeMike
Byong Park
Cesiumsponge
Chris F.
Chris Hooper
Corey Worthington
Derek Woodroffe
Dalus
Dan Strother
Daniel Davis
Daniel Uhrenholt
datasheetarchive
Dave Billington
Dave Marshall
David F.
Dennis Rogers
drelectrix
Dr. John Gudenas
Dr. Spark
E.TexasTesla
eastvoltresearch
Eirik Taylor
Erik Dyakov
Erlend^SE
Finn Hammer
Firebug24k
GalliumMan
Gary Peterson
George Slade
GhostNull
Gordon Mcknight
Graham Armitage
Grant
GreySoul
Henry H
IamSmooth
In memory of Leo Powning
Jacob Cash
James Howells
James Pawson
Jeff Greenfield
Jeff Thomas
Jesse Frost
Jim Mitchell
jlr134
Joe Mastroianni
John Forcina
John Oberg
John Willcutt
Jon Newcomb
klugesmith
Leslie Wright
Lutz Hoffman
Mads Barnkob
Martin King
Mats Karlsson
Matt Gibson
Matthew Guidry
mbd
Michael D'Angelo
Mikkel
mileswaldron
mister_rf
Neil Foster
Nick de Smith
Nick Soroka
nicklenorp
Nik
Norman Stanley
Patrick Coleman
Paul Brodie
Paul Jordan
Paul Montgomery
Ped
Peter Krogen
Peter Terren
PhilGood
Richard Feldman
Robert Bush
Royce Bailey
Scott Fusare
Scott Newman
smiffy
Stella
Steven Busic
Steve Conner
Steve Jones
Steve Ward
Sulaiman
Thomas Coyle
Thomas A. Wallace
Thomas W
Timo
Torch
Ulf Jonsson
vasil
Vaxian
vladi mazzilli
wastehl
Weston
William Kim
William N.
William Stehl
Wesley Venis
The aforementioned have contributed financially to the continuing triumph of 4hv.org. They are deserving of my most heartfelt thanks.
Registered Member #543
Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992
Avalanche wrote ...
I've just been playing around with negative feedback, what would be the pros and cons of taking the feedback from before or after the AF transformer? I've tried both, just simple resistive only feedback with a decade box, and I thought feedback direct from the pentode seemed cleaner. I would think that taking it after the AF transformer would bring in problems with phase shift, and make the frequency response a bit less linear... I can see myself getting a bit hooked on feedback networks made with various components to try and get some nice tones.
AF Output transformers colour the output with their own particular spectral responses and oddities, so taking the negative feedback from after the transformer does a good job of ironing these out - and you'll notice your triode has been designed with more than enough gain to be able to give most of it away again in a negative feedback loop.
Here is a good quality negative feedback design of a triode-pentode amplifier from 1964 using the ECL86, a double valve similar to yours, but slightly more powerful:
You'll notice that the feedback circuit is in parallel with the load, and so will draw power. For this reason, you must make sure that the resistance of the feedback circuit is at least 10 times greater than the load impedance i.e. 80R if Zload is 8R.
Registered Member #2463
Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546
[]that plus my heater center tap no longer being in the center[/b]
If this is an issue, a 100 ohm trim pot across the 6.3 volt winding with its wiper earthed will allow you to ground the midpoint center electrically. (lift the actual centertap) The other thing is the EZ81 cathode magnetically couples the 100HZ pulses of RAW Ecc to the heater bus.
Also note the series resistor in the grid of Proud Mary's ECL86 circuit. This is used to de-snivet extremely hot tubes/layout issues and leads to considering the grid circuit of class A anplifiers in terms of input admittance. Another tweak to response is to look at the screen bypass condenser and see if can be reduced, and what happens.
One thing that was said at a Syn-Aud-Con by Don Davis is that allthings done to reduce a noise dont add numerically. Sometimes the last thing you do cleans it up. On line AES tutorial recordings of Thiele & Small (124th) and Whitlock (on grounding) (126th) are great especially for valve amp builders.
Registered Member #103
Joined: Thu Feb 09 2006, 08:16PM
Location: Derby, UK
Posts: 845
Quick update... a couple of hours of fiddling around, and implementing peoples latest suggestions and I have got some quite loud sounds out of the amp. I finally drew up a schematic as it is now, I have drawn schematics before but never posted them because it keeps changing as I figure out what's going on!
It is now starting to look more and more like a cross between Proud Mary's circuit above, and this circuit which I found
The heaters are now at 6.3VAC with the center tap in the center. I have added feedback from the speaker, but noticed the volume was too low with a decent amount of feedback. It looked like the triode stage didn't have enough gain - I was using just the 2k2 cathode resistor with no AF bypass. After adding the AF bypass and extra resistor as in the schematic Proud Mary posted, I managed to get a decent amplitude with quite a lot of feedback.
I also decoupled the screen grid supply by adding the 4k7 resistor and the .47u capacitor (previously my screen was just tied to HT) I'm not sure if it's a problem, but I can now see my AF waveform on the screen grid with a (maximum) peak to peak amplitude of about 10v.
I've been testing with 2 speakers as shown - the 1950's speaker sounds the best, but only with certain types of music. It sounds fantastic with the guitar, just a shame I can't play!
Registered Member #543
Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992
Have you some particular reason for not decoupling the triode anode supply - done in the circuit I posted with the 47k decoupling resistor and its 16uF electrolytic? With the pentode section tugging on the same supply line, not decoupling seems like an invitation to instability.
Heater wires should always be twisted together. All wires and leads should be as short as possible. If hum is a problem, try using shielded wire in the triode grid circuit. Solder a copper shielding strip across the socket base to minimise interaction between triode and pentode. Use a full size screening can on the valve.
Registered Member #2463
Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546
Some 9 pin sockets have a centerpost on the terminal side in the center which if grounded acts as a bit of a shield and also a tiepoint for element parts which go to ground. Amplifiers can have either a ground tree of tie everything back to chassis lugs.
Another thing about production of amplifiers the output transformer was code blue red. Blue to the anode-red to Ecc. This makes sure all their products are phased the same.
When the transformer is working up around 15 kHz you might find that the winding terminal closest to the core has a different LC model than the outer terminal depending on the coil build. Another reason to connect this to Ecc.. Also-Bigger amps) when some clutz disconnects the speaker and then cranks the level to see why he cant hear anything, smoke results often- The valve socket flashes-(lucky) or the output transformer may ark to core if the core end is on the anode. (unlucky)
Another fine point: capacitors have an outer foil. If coupling grid to anode the anode is closer to ground, impedance wise, than is the following grid, So the outer foil to the anode would make the coupling shielded as it were.
Prowd Mary said transformers color the output. So do signal path capacitors, this is being found more and more, now that sophisticated measuring gear is common place. they have twist pecularities at certain frequencies. If in the feedback path- worse.
Just a point about the choke. A good experiment would be to replace that shunt resistor with a 1.5 ufd AC or paper cap and see if your voltage &or ripple improves.
Registered Member #30
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706
Looks cool! Is that your home workshop Avalanche, it seems quite well equipped.
The blue/red thing is only true in the USA, which had the RETMA colour code for wiring. Elsewhere in the world there was no standard colour code.
The problem of running a valve amp with the speaker disconnected is a real one. A 220 ohm resistor across the speaker terminals can help. Maybe. Sometimes.
The outer foil thing for coupling capacitors is suspect. Just think, the outer foil is connected to the inner one by... a capacitor! If it's a coupling capacitor in an audio amp, then it'll have a low impedance at 60Hz, so it doesn't matter a bit which end you connect where: it'll pick up hum just the same.
I don't really believe in "capacitor sound" either.
Registered Member #2463
Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546
Belief in "capacitor sound' --
Douglas Self went into this in the "master workshop on linear audio amplifier design" at the 124th AES and his talk snd slides can be watched and heard on their tutorial series on line.
The common destroyer/colorer of audio is the speaker, but in the measurements that be done today, small things can be hunted down and capacitors often are the cause.
In disgustingly aged audio gear, if it works, the practice of re-capping it just to be sure, can alter the sound, because the capacitors were lousy when they were made, but that's what they used, particularly paper. Throwing in a slew of mylars with no leakage could shift the operating point of valves.
Registered Member #103
Joined: Thu Feb 09 2006, 08:16PM
Location: Derby, UK
Posts: 845
Steve : yes it's my home workshop you see in the pics, I got fed up with lashing up things like power supplies, batteries and 555 timers every time I wanted to do anything, so I splashed out for some half decent equipment.
This feedback thing has got me a bit hooked, and I've been playing around with it all day. To start with, I cleared up a lot of the noise just by tidying up the wiring - totally underestimated the importance of that, what an amazing difference, no whistle anymore. Decoupled the triode supply with a 22k resistor and 100uF, no mains hum anymore. Totally silent... Yes!
Anyway I couldn't for the life of me get it sounding right with resistive feedback from the speaker to the triode. Many times I thought I'd cracked it, and I was down to just 147ohms at one point, but then I would put another song on and it would sound awful, or the resistor value would change just for that song. I settled for 147ohms in the end, and then wired in my volume control. I soon noticed that there was a lot of 'noise' on the anode waveform of the triode, that disappeared when at full volume. I zoomed into this noise and it was oscillation at about 24Khz. A 220pF cap from the pentode's anode to the triode cured this, but I still couldn't get it sounding right between different songs, and it was driving me mad!
I decided to get a bit more scientific and look at the phase shift between the input and output (after the AF transformer). I removed all the feedback, and ran the amp open-loop.
I think this explains the oscillation - if we say there is about 90 degrees of shift at 13Khz, then oscillation at 26Khz (speaker only feedback) sounds about right. It looks like I'll need more than a simple resistive feedback from speaker to get it sounding right across the entire audio range...
The next pic is open loop frequency response, I guess the falling off towards the end is mostly caused by the AF transformer. Not sure how well this is supposed to 'self compensate' for the above phase shift with resistive only feedback?
Next picture is with the feedback reconnected as it was before I gave up messing about with it - 220pF from pentode anode to triode, and 147ohms from speaker to triode. It peaks at 9.7Khz and sounds awful with some songs (it's clean except when there's 'a lot going on' then it goes crunchy).
Phew... I think I need to learn more about feedback. Didn't expect I would get this hooked. If I can get it reasonably hi-fi I'll be happy, but I don't know how much I'm limiting myself with my choice of components.
I have accidentally run it without the speakers a couple of times, but fortunately no smoke! Yet!
Registered Member #543
Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992
If you measure R and L of the O/P transformer primary, then you will be able to calculate its impedance at 3kHz.
The datasheet figure of 5.2V RMS on the pentode g1 for 2.5W output depends on Ra being 7k5 at 200V supply.
As E = sqrt(2.5 x 7500), in normal operation 137V should be developed across the primary, giving an amplification factor A of 137/5.2 = 26.
As the Miller effect shunt capacitance of the pentode depends on the value of A, we cannot determine the frequency response until you have calculated the impedance of the primary at a nominal 3kHz.
Registered Member #2463
Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546
The whole point of a pentode is to eliminate the the miller effect of the amplification in a triode changing the plate -grid degeneration due to capacitance. Thats what the screen grid does. a good paper on the miller effect on the triode is here, as applied to music amps.
Your open loop power curve looks fairly decent- if you changed the power levels to db- its with 3db from ~~100-8000 Hz which in reality is all those little single ended stages ever do in practice-what can your speaker do? (asssuming its true RMS which you cannot judge on a scope) Why it sould crappy is likely IM distortion the screen voltage tweeks that a bit There should be lots of old SMPTE analysers kicking about- (IM 60 hz modulating 8000). Have you had those old ECF83s tested?
I think Langford-Smith is scanned on the net somewhere-at any rate here's why it doesn't sound right. note the dissonant parts mid range.
just a tweak- try about 1500 pF across the primary of the output transformer and see how it sounds.
This site is powered by e107, which is released under the GNU GPL License. All work on this site, except where otherwise noted, is licensed under a Creative Commons Attribution-ShareAlike 2.5 License. By submitting any information to this site, you agree that anything submitted will be so licensed. Please read our Disclaimer and Policies page for information on your rights and responsibilities regarding this site.
Thinking about building a valve amp...
