• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Why use a pentode?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Seeing as how Christmas is round the corner, desire for me to thank all posters here for loads of info that was new, to me at least - and others of confirmation.

I am particularly grateful for Nigel Pearson (round about post #71) for the notes on Blumlein and the events surrounding him.

Perhaps a further extra - by now much belated. I had a graph somewhere (mislaid, I believe the nice term is for sloppy filing) indicating some 3x variation in distortion with the ratio of plate and screen voltages in a pentode RC application.

Then back to the original question for using pentodes as input stages. I don't recall the (lack of) Miller effect being mentioned. As a power amplifier input stage with an undefined feeding impedance - one often simply sees a 1M volume control there - I have found on accasion that a triode input can have an influence on NFB stability with the feeding impedance varying between say 200K and a few K.ohm. Again depending, Miller effect time constant can approximate that of some phase-correcting RC over the plate, making the h.f. stability plot less than desired. Perhaps not of prime concern, but another reason why I prefer using pentode inputs: Independence of feeding impedances.
 
^
|

Can't say I have. Those kinds of things were designed to be cheap, cheap, cheap, and not for any sort of sonic excellence. The OPTs for SE 50C5s are tiny with horrid frequency response. That seemed to be a means of hum suppression: roll off the lows below 200Hz or so. (That's about the only application as the spec sheet mentions nothing of PP designs at all.) The odd thing about the specs for the 50C5 is two sets of plate characteristics: one for V22= 110V (most common usage) and another for V22= 90V that's a good deal more linear, though you sacrifice some output power. Like the manufacturer had better sonic performance in mind?

You can find good PP, Class A, loadlines for the type that look promising. Of course, you'd need better OPTs, and these might have to be custom jobs to use these right.

now where is the like button?:D
 
I think why people prefer triodes is the variables are few ( still too many ) . One can say quickly if the device is going to meet the requirements . That might be an illusion as the interaction with the next stage is important . Doubly important when a zero loop feedback design . Often people say 5 % THD is OK because the advantages of zero loop are obvious . Perhaps . Certainly I can live with it . To say I can't hear it is not true . Reduce this to 1% and have good bandwidth then is is less clear cut . Often what people would say is distortion is lack of LF and HF output or damping factor .

The pentode often offers a free lunch . A way of breaking the all triode distortion chain . The simple rule seems to be even stages pentode + triode is possible . Uneven as in 3 stages perhaps all triode . An SE amp I made was perfect as pentode + triode . When I converted it to PP ( using 2 x EL 34 in LTP with valve two grid grounded ) it worked best as all triode . I did it only to see if my driver valve could raise 56 Vrms ( for 211 ) which is did . This meant a gain penalty . The PP amp looked SE in distortion which I liked and had low numbers . The transformer a very cheap Danbury 6K6 impedance design . Hopefully I get to make a proper one when time and funds permit . It wasn't very powerful, . When pushed it looked like a cheap amp in distortion spectrum so I let it be 8 watts ( I had UL taps which if I tried hard enough might have worked ) . It was more cost effective than SE and would have better bass . An extra EL34 is cheaper than the extra cost of a good SE transformer over a cheap PP .

Someone told me harmonic cancellation between valves is unreliable . I had a bit of beginners luck and found that not to be true . It probably is true if repeating circuits of others . All I can say is don't assume that . Use worn out pentodes to confirm it is all I will say . When I did I got 1.5 % THD and when new 1% . The spectrum mostly identical old or new if allowing for the THD level .
 
Alan Blumlein . My friend John is an electronics engineer , mostly magnetometers these days . More importantly a sound engineer ( in Oxfordshire , very good if needing to know , huge collection of microphones and still has a high grade rig ) . John could not believe the photo . He lived and worked on that road and knew the house well . Never realizing one of his hero's had lived there .
 
I think why people prefer triodes is the variables are few ( still too many ) . One can say quickly if the device is going to meet the requirements . That might be an illusion as the interaction with the next stage is important . Doubly important when a zero loop feedback design . Often people say 5 % THD is OK because the advantages of zero loop are obvious . Perhaps . Certainly I can live with it . To say I can't hear it is not true . Reduce this to 1% and have good bandwidth then is is less clear cut . Often what people would say is distortion is lack of LF and HF output or damping factor .

The pentode often offers a free lunch . A way of breaking the all triode distortion chain . The simple rule seems to be even stages pentode + triode is possible . Uneven as in 3 stages perhaps all triode . An SE amp I made was perfect as pentode + triode . When I converted it to PP ( using 2 x EL 34 in LTP with valve two grid grounded ) it worked best as all triode . I did it only to see if my driver valve could raise 56 Vrms ( for 211 ) which is did . This meant a gain penalty . The PP amp looked SE in distortion which I liked and had low numbers . The transformer a very cheap Danbury 6K6 impedance design . Hopefully I get to make a proper one when time and funds permit . It wasn't very powerful, . When pushed it looked like a cheap amp in distortion spectrum so I let it be 8 watts ( I had UL taps which if I tried hard enough might have worked ) . It was more cost effective than SE and would have better bass . An extra EL34 is cheaper than the extra cost of a good SE transformer over a cheap PP .

Someone told me harmonic cancellation between valves is unreliable . I had a bit of beginners luck and found that not to be true . It probably is true if repeating circuits of others . All I can say is don't assume that . Use worn out pentodes to confirm it is all I will say . When I did I got 1.5 % THD and when new 1% . The spectrum mostly identical old or new if allowing for the THD level .

I agree with most of that, especially harmonic cancellation that I have used successfully in several SE amps to cancel mainly those large amounts of 2nd harmonic...:D People could not believe how the bass was so good with a damping factor barely equal to 3.
Also, the Loftin-White 2 stages SE amp with pentode driver is a lot better than that with triode driver, IMHO.
I do not agree completely on the "perhaps" regarding the 5% THD limit. It depends on how the amplifier reaches that limit. So in the end is not exactly a general rule. Yes, the nominal power should that at 1% (difficult or inconvenient for most low power zero fb SE...:D) but with triode PP or higher power SE can be a different story in comparison pentodes + fb.

If you have PP pentode with feedback the max effective power you can use will not be much more than the nominal at 1% because it is usually close to clipping that will negate the effect of feedback. Distortion will simply be too much and more importantly the rate at which distortion increases too fast. With my class A 45 PP amp I have got just over 0.1% THD at 1W, 1% at 6W and, thanks to proper driver, I get 5% at 15W (class AB2) without any feedback! 6W is not a lot but I can fully use it without worrying about peaks, recovery (also thanks to DC coupled driver), damping etc...it is enough for me. If needed more power I could get the same behaviour from bigger triodes. The bandwidth (-3dB) is 2Hz-75KHz at 1W and 12Hz-75KHz at 6W. HF response with triodes doesn't need feedback to get even higher at 100KHz with perfect textbook behaviour because of their low/moderate internal impedance (once you get an optimal value for SQRT(Ld/Cs) from the OT). That is a lot more difficult with pentodes, you need at least local fb.
I can get 15W at less than 1% from an EL84 PP (in AB) but is not in the same league for me, although I like a lot the EL84 as well. Of course the advantage of the EL84 PP is that it is simpler and cheaper.
 
45 . That's very interesting . My friend admitted my SE amp had better bass compared with his 300 B ( better everything if cost comes into it ) . We listened to many 300 B's , I wasn't sure any were especially good . My amp using either KT88 or EL 34 ( 34 if me ) . It never occurred to me that lower distortion was the factor as much as damping factor . I think you must be right . My transformer cheap and by measurement OK ( 20 Hz to 47 kHz - 3dB at 5 W ) The 300 B a Sowter design costing 3 times more ( I suspect more as it is a bespoke version ) . I think the 300 B amp was a clone of typical designs . The attention on " brand name parts " . My parts were high grade for other purpose devices sometimes 5 % the cost . I still feel on all things mine sounded better . Faster yet open . One device was a 0.1 uF 400 V polyester by BC components . Not a choice , it was what I had . On reflection perfect for the job . Far better than 99% of what transistor amps get from big companies ( Hypex , looks like ceramic to me , input to NE 5532 ) .

One of the best Hi Fi systems I ever heard was using Garrard 301 and stacked Quad ESL 57's . I was astonished to see it was driven by Heath EL 84 amplifiers . I must conclude their transformers must have been top notch . When asked how I liked the sound I said I didn't as it was too real . The guy next to me said he knew exactly what I meant . I liked it very much truth be told , even on CD . Nigel hates CD so that's high praise . Maybe if being nit picking upper mid-band prominent . With microphones we seek that so not a major criticism . More odd as ESL 57 are not ones to do that .

I have to say mostly valves annoy me and like steam locomotives it isn't all joy . However when distortion is lowered whilst no loop feedback used I can hear layers in the music I don't usually recognize as being there . When low in distortion I never think valves . The Williamson amp goes too far . I would rather have a JLH transistor amp . This is not to say in it's time it and the Leak TL 12 weren't miracles .
 
Nigel, valve amps that achieve low Zout by means of loop fb still are not in the same league as SS amps. In fact you can see that, for a given input, they remain costant power sources, i.e. the power into 4R is the same into 8R, in the best case! Most of the times the have less power into 4R because are usually designed for max Pout into 8R. So won't be ideal anyway for speakers with roller coaster highly reactive impedance. However, provided they remain stable, they have less distortion. Not easy to attribute a certain result to a certain characteristic in this case.
However, with low Rp triodes in PP, even more with PPP, some power gain is possible, about 10-25% increase, going from 8R to 4R impendance change just requiring pure class A into 8R which becomes AB into 4. Having no fb here you can assess better the role of low Zout and distortion. If distortion is not low you always notice some lack of damping despite low Zout.


I actually have a couple of friends that are going to build a PPP EL84 amp and and I am really thinking to make a normal 15W PP for me as well. This one could be DHT killer!:D
I have been thinking for a while about a different solution from usual including the OT and feedback.
First, the idea is to use 20% UL which, in the Mullard datasheet, seems to be the best compromise because, for exactly the same operative conditions, the max Pout is just 1 W less than the pentode but distortion is 3 times lower! Distortion doesn't change significantly going for 43% UL thorugh out the entire Pout range (from low level to Pmax) but the available max Pout drops to 10-11W. One thing that I didn't notice before is that both UL solutions, despite perfect balance, have 2nd and 3rd harmonic contributing in equal amounts. This might be due to tapping rather than separete tertiary winding at lower DC voltage? Maybe.
Second, I have designed the OT with cathode windings that should give about 9 dB feedback, enough to avoid loop fb, getting quite lower Zout than triode connection and in theroy less than 0.5% THD at 15W. To make it more practical I will put the screens at 25% which comes very handy with the sectioning I have fugured out. HF response should be already 60-70KHz(-3 dB) without any fb.
I live in the same county as you. If you want to meet sometime we might be closer than we think! Just drop me a pm.
 
Last edited:
I have played with various UL feedback . Using taps I had by chance at 82% towards the triode end I got higher gain and no obvious downgrade from triode ( better I think ) . This seemed a better compromise compared with ideas of Kitic , he seems to have 30% gain boost . Just numbers and not saying I listened carefully as I reject the Kitic designs early on ( I didn't find I could equal his numbers ) . I love Kitic's idea . However I feel it to be theoretical for a Rp different to his beloved ECC81 . EF184 would be the better ECC81 . It almost is one as a triode . That way the Rp can be higher and a better V to I converter . Kitic is to me a super triode . The two devices almost fuse .

I studied valves 40 years ago . I was frustrated , as fascinating as it was what possible use ? Now after the death of my brother I have been playing with his inheritance to me . Sadly things started which he seemed dissatisfied with . He often wrote into Television magazine and was a guru. I feel both guilty ( his love more than mine ) and pleased to have relearned how to ride this horse . One I knew by writing mostly .

I have often looked at sine-waves . Even when highly distorted they can deceive the eye and look to be of lower distortion . Equally some that have OK THD look odd . The Jean Hiraga ideal of progressive reduction of harmonics of distortion seems to agree with the visual . I totally believe excessive 2 nd harmonics to cause boominess . That it is often connected with low damping factor should be questions to be considered . We also have that a seemingly restricted frequency response which might be aided by these factors . My transformer is very OK at 20 Hz . It shots up to 10 % distortion qat15 Hz 5 W . This seems to be where fine turning will make a silk purse out of something less . Who is to say how much the 15 to 20 Hz range matters ? My speakers are 55 Hz - 3dB . Loosing anything is the question and not that 15 Hz matters .

Analogy . Often people prefer RIAA over the IEC additional 7950 uS curve . Looking at how a record is produced suggests CD to have obviously superior bass . Ears say otherwise . Possibly the 8 to 12 Hz resonance of the pickup is preferred ( 8 Hz is better than thought if asking ) ? SME marketed a damper to suppress it . I never liked it . The Shure and Stanton brush damper seems OK . It is very difficult to specify what is best . The cartridge people know .

I will drop you a line .
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.