Sounds fine. So you're using a single 4P1L in the output. That sounded pretty good to me, though my speakers are 89db sensitive and I needed a pair in PSE. Probably that took the edge off the sound a bit. I found filament bias was the way to go to eliminate the cathode bypass cap. For the best sound I used a choke input supply to a Rod Coleman reg. Chokes were Hammond 159Y for series filaments.
It's true I didn't like 4P1L into 4P1L. The sound was rather thin and a little edgy. To be honest I prefer a 300b in the output these days. But with sensitive speakers I could see a single 4P1L working. I see the input is resistor loaded. You could try a plate choke there as well. Have you tried a 26 or 2P29L input? Otherwise I would use an EL33 or a PL33 which is cheaper and 19v filaments. I did a big shootout of input tubes with a gain of 15-20 including triode strapped EL41, E80L, KT61, EL84, EL33 and EL11. The 2 best to my ears were EL33 and EL11. I'd also mention triodes like 2C22/7193 and 6P5 and double triodes like E80CC. But I'd first try EL33 which has good body and warmth - ideal for the thinner sounding 4P1L. It's a very nice tube indeed - quite like a 26 in character.
My usual recommendation is to replace all cathode bypass caps with DC Link caps. I did a shootout of cathode bypass caps and they were the most transparent and had the best instrumental timbre. I used Vishay and Kemet.
It's true I didn't like 4P1L into 4P1L. The sound was rather thin and a little edgy. To be honest I prefer a 300b in the output these days. But with sensitive speakers I could see a single 4P1L working. I see the input is resistor loaded. You could try a plate choke there as well. Have you tried a 26 or 2P29L input? Otherwise I would use an EL33 or a PL33 which is cheaper and 19v filaments. I did a big shootout of input tubes with a gain of 15-20 including triode strapped EL41, E80L, KT61, EL84, EL33 and EL11. The 2 best to my ears were EL33 and EL11. I'd also mention triodes like 2C22/7193 and 6P5 and double triodes like E80CC. But I'd first try EL33 which has good body and warmth - ideal for the thinner sounding 4P1L. It's a very nice tube indeed - quite like a 26 in character.
My usual recommendation is to replace all cathode bypass caps with DC Link caps. I did a shootout of cathode bypass caps and they were the most transparent and had the best instrumental timbre. I used Vishay and Kemet.
Andy,
Thank you for your suggestions. Yes, it's true that, with a sensitivity of 97 dBs, I had no reason to try paralleling two 4p1Ls - I wouldn't expect them to sound better than a single one. But, with just 89 dBs, you need a little more power in order to get realistic levels of loudness. To tell you the truth, I estimated that, at the loudest level, the used power with my speakers should not exceed 200 mW.
There are so many tubes out there to try as an input driver, that makes this scope (I mean, searching for the best possible driver) an endless experiment.
I don't have any EL33 (nor PL33) tubes available. On the other hand, since I have seen that this tube is similar to 6V6, I could try this, as I have a couple of them. I don't know if they sound similar. I'm also tempted to try a DH pentode that I have in hand, namely the VT-47 (military version of the 47).
I have read the whole thread and I'm aware of the raw DC supply scheme for the Coleman regulators and that's exactly what I'm going to do.
Regarding the loading of the driver stage, it is in my intentions to try both choke and CCS (and/or hybrid mu follower). But time is always too little!
Can you elaborate a bit more on the DC Link caps? I have never heard of them.
You know, I'm a musician (amateur) too and my perception of reproduced music is always compared to my experience of live (acoustic) music. So, that's why I always trust people with such an experience (not necessarily musicians) of listening to live music and not the usual audiophiles.
With 40 degrees in Athens, it is not a good time to switch on the soldering iron, but I will keep you informed on my future findings.
Evangelos
Which driver you use in you 300B? In mine, I currently use a pentode wired EL84, which I have found to sound best of everything else I have tried so far (haven't tried so many, though).
The EL33 and EL11 sound better than 6V6 in triode. Not the same thing. The EL33 or PL33 is cheap - buy a couple. Contact me - I have plenty I'm selling on eBay, and some EL11 as well. If a 4P1L gives you enough volume, yes - try 46, 47 and 49. All excellent. And 2P29L which is cheap.
Search for Kemet C4AQ DC Link caps from an EU supplier.
https://www.digikey.de/product-detail/en/kemet/C4AQLBW5700A3LK/399-17006-ND/8345969
An update regarding my trials and my findings so far:
Regarding the driver stage, after trying various types of anode load on the 6J9P-E, in particular, resistor, CCS (Garry Pimm's self-bias CCS) and choke, I decided that I liked the anode choke load most, with the CCS load coming close.
Having happily ended to a certain configuration with my 4P1L and being currently in an extensive quest for the best power amplifier sound I could get from what types of tubes I have in hand, I tried also a 2A3 as an output stage. Well, it beat the 4P1L hands down. Comparing to the 2A3, 4P1L was more harsh and edgy, with less separation. Extension, both to the low and to the high end was similar, but the 2A3 was so much clearer and descriptive that made me feel it was magical. I must note that I have a couple of original RCA 2A3 tubes that I had found over a decade ago in a ham fest (bought for peanuts) and never had a chance to try. So, for the moment, my favorite amp seems to be this one.
For the record, all other parts being identical, I also tried a 300B as an output tube to see if there were any differences in relation the 2A3. Well, although not large, the differences were there, with the 2A3 being more favorable for me.
Regarding the driver stage, after trying various types of anode load on the 6J9P-E, in particular, resistor, CCS (Garry Pimm's self-bias CCS) and choke, I decided that I liked the anode choke load most, with the CCS load coming close.
Having happily ended to a certain configuration with my 4P1L and being currently in an extensive quest for the best power amplifier sound I could get from what types of tubes I have in hand, I tried also a 2A3 as an output stage. Well, it beat the 4P1L hands down. Comparing to the 2A3, 4P1L was more harsh and edgy, with less separation. Extension, both to the low and to the high end was similar, but the 2A3 was so much clearer and descriptive that made me feel it was magical. I must note that I have a couple of original RCA 2A3 tubes that I had found over a decade ago in a ham fest (bought for peanuts) and never had a chance to try. So, for the moment, my favorite amp seems to be this one.
For the record, all other parts being identical, I also tried a 300B as an output tube to see if there were any differences in relation the 2A3. Well, although not large, the differences were there, with the 2A3 being more favorable for me.
Hi Kenev. Your findings are pretty much the same as mine. My favourite driver is a triode with anode choke. I have a nice big amorphous pair of NP Acoustics ones. The challenge is finding the right triode, and for me it should be unbypassed if at all possible.
And also, I previously used PSE 4P1L as outputs in filament bias. When I changed that to 300b I got exactly the same impression - the 300b was more musical, detailed but smooth and without any tiring harshness. In comparison the 4P1L was thinner with a slight edge to the sound. I haven't properly tried 2a3 in comparison with 300b. I do have a pair of EH ones.
And also, I previously used PSE 4P1L as outputs in filament bias. When I changed that to 300b I got exactly the same impression - the 300b was more musical, detailed but smooth and without any tiring harshness. In comparison the 4P1L was thinner with a slight edge to the sound. I haven't properly tried 2a3 in comparison with 300b. I do have a pair of EH ones.
Hi Andy,
I'm glad that you confirm my findings. It seems that we have similar listening criteria and, therefore, I can trust your opinions.
Yes, 4P1L, although impressive, has not the fine detail achieved by a DHT. Listening to a track with an organ and a chorus, is a piece of music that I use a lot in critical listening tests: I found out that, properly amplified, I can hear each voice on its place, rather than a confused ensemble of voices.
Differences between 300B and 2A3 are minor, but distinguishable. Of course, 2A3s are NOS RCA and my best sounding 300B is a TJ FullMusic, if that makes any difference.
Regarding the driver stage, since I need that extra gain, a DHT cannot serve me well and, from a variety of triode connected pentodes or tetrodes that I have tried so far, I have concluded that I like the 6J9P-E most of all. For an anode choke I don't use anything fancy, just a choke I have wound myself on a EI75 core, with 4 series connected windings and an inductance of ca. 120 Hy at 20 mA.
Also, I have to note that I use a parafeed OPT with a primary reactance of 5K, since that was what I had in hand from a previous project. That means that output power is lower than usual, but who needs that extra watt?
Evangelos
I'm glad that you confirm my findings. It seems that we have similar listening criteria and, therefore, I can trust your opinions.
Yes, 4P1L, although impressive, has not the fine detail achieved by a DHT. Listening to a track with an organ and a chorus, is a piece of music that I use a lot in critical listening tests: I found out that, properly amplified, I can hear each voice on its place, rather than a confused ensemble of voices.
Differences between 300B and 2A3 are minor, but distinguishable. Of course, 2A3s are NOS RCA and my best sounding 300B is a TJ FullMusic, if that makes any difference.
Regarding the driver stage, since I need that extra gain, a DHT cannot serve me well and, from a variety of triode connected pentodes or tetrodes that I have tried so far, I have concluded that I like the 6J9P-E most of all. For an anode choke I don't use anything fancy, just a choke I have wound myself on a EI75 core, with 4 series connected windings and an inductance of ca. 120 Hy at 20 mA.
Also, I have to note that I use a parafeed OPT with a primary reactance of 5K, since that was what I had in hand from a previous project. That means that output power is lower than usual, but who needs that extra watt?
Evangelos
Well; it is the wrong way to compare. Each tube needs own optimal regimes to sound the best. Anode voltage, load resistance, bias, driver. If "all other parts identical", one of 3 tubes that you compared was in proper regimes, for other 2 regimes were sub-optimal. Even for 300B, since it can do more when properly used.
Saying "parts", I wanted to note that passive components were not different. All other parameters, except operating conditions of the output tube, were identical. That is: driver stage operating with the exact same voltage, current and anode load, coupling capacitors being identical, anode load chokes and OPT being identical. Even the power supply source was the same (of course, with a different voltage) - that is, I used my laboratory PS. Only thing that changed was the voltage and bias of the output tube. How could this be the wrong way to compare with all other conditions (the output stage) being just the same?
Last edited:
Perhaps it could be as simple as the driver stage being incapable of supplying the current required to drive one of these output tubes or that the power supply refreshes differently with heavier loads. It'd be wrong to conclude that output tube is "worse" than the others, but you might conclude that it wasn't as compatible as the others within its overall "test rig".
What is slightly worrying is the continuous subjective mix and match of components rather than studying and understanding the overall design and how it ought to work. This is a little akin to taking a car made by Ford, fitting a more powerful turbo from a Bentley to it, then upgrading the transmission to one from a Corvette (to handle the extra torque). Next someone points out the GM cars had gold plated connectors on them, so the original wiring loom is stripped out and one from a GM is shoehorned into the car, which now sports low profile torroidal Ferrari wheels and amorphous air cored brakes from a Jaguar. Surely this must be better?
The most worrying aspect of this is that in another thread, Andy has told us he in his seventies and his system has small bookshelf speakers. He may be offering very good advice, but this is still largely untested at both the higher and lower ends of the frequency spectrum.
kind regards
Marek
What is slightly worrying is the continuous subjective mix and match of components rather than studying and understanding the overall design and how it ought to work. This is a little akin to taking a car made by Ford, fitting a more powerful turbo from a Bentley to it, then upgrading the transmission to one from a Corvette (to handle the extra torque). Next someone points out the GM cars had gold plated connectors on them, so the original wiring loom is stripped out and one from a GM is shoehorned into the car, which now sports low profile torroidal Ferrari wheels and amorphous air cored brakes from a Jaguar. Surely this must be better?
The most worrying aspect of this is that in another thread, Andy has told us he in his seventies and his system has small bookshelf speakers. He may be offering very good advice, but this is still largely untested at both the higher and lower ends of the frequency spectrum.
kind regards
Marek
Why are you using the word "worrying"? What has emotive language to do with testing components?
Yes, as I said, I can't subjectively assess low bass or extreme high treble. But I'm probably not much different from the very many audiophiles in their 70s. What I can hear well is the essential frequencies where the overwhelmingly large proportion of the sound is situated. And being a musician I know what to listen for in the sounds of acoustic instruments. I have to be a bit more careful in assessing components, and so in addition I ask my wife to listen when I'm making changes. She has excellent hearing in her late 50s. Her feedback turns out to be essentially the same as mine, so I appear to be hearing what she hears. But bear in mind also that I've been building continuously for 15 years using a variety of speakers from Magnepan, Apogee, Mark audio etc. previous to my current bookshelf speakers this year, and a lot of my comments date back several years.
We've had plenty of discussions on this, and a common view is that subjective comments can be "useful" in giving directions for further study, though any builder would want to test for themselves. Subjective tests may not replicate each and every parameter, but results are improved as said above by keeping as many components as possible the same, and also comparing A,B,C... to give a rank order. Nobody is saying that subjective results are conclusive, but neither are measurements "conclusive" even though they have a different kind of validity. In the end we evaluate whatever information is available and test for ourselves in our own systems to achieve the results each of us seeks. If you want to ignore all subjective comments on audio components, that's entirely your choice.
.
Last edited:
I don't think that the driver stage has to deal with much different loads - all 3 tube I tried are biased between 40 and 70 mA, with 4P1L and 2A3 being closer at 40 and 55 mA, respectively. Also, since for my trials I always use my lab PS (both for convenience and for safety reasons), I don't think that it behaves differently, as it is well regulated (in fact, I don't quite understand the meaning of "refreshes differently").
First of all, before implementing a new design, I always study and calculate my circuit, based on loadlines, published data and others' suggestions. I don't see where this "continuous subjective mix and match of components" idea comes from. Did I give you the impression that I don't know what I'm doing? Surely I'm not an expert, I'm always learning, but, definitely, I'm not an ignorant. Of course, any opinion of how a piece of equipment sounds is absolutely subjective, based on my acoustical experience being an amateur musician and having played in a large number of gigs and having attended numerous live music concerts, mainly classical and acoustic (jazz, folk etc). However, due to the fact that my judgment might be affected by the fact that I have spent time and effort to build something, I always seek the assessment of a couple of highly trusted friends and it proves that, at times, my impression was biased. But still I cannot understand your statement about this mixing. Everything that I build is a mix of different components and stages. If something suits me well, why bother if it is a mixture of Ford and Jaguar? Besides that, I don't think I ever mentioned the kind of components I used (either gold plated connectors or amorphous core OPT - just humble nickel plated banana sockets and M6 core, self wound, anode load chokes and OPTs).
Anyway, my opinion is that if something is pleasant to my ears, that makes me happy and pushes me to go further in an endless quest for the "ultimate sound". If I can help someone here with a couple of suggestions or findings, I think this might be helpful, as I have received much help through this forum reading other people's posts.
Using both quantitive and qualitative data is banal in several disciplines. I'm a psychologist and when setting up an experiment it is fairly common to collect qualitative data (views of subjects polled) to get an idea of what to test further. Definitive results of experiments are then carefully designed for significance. This can be done in audio with double blind testing and extreme care with standardising the procedure, but it's a complex process which is rarely carried out when simply building for ones own use. So subjective listening tests with a lower level of significance are common and widespread. Builders on this forum are usually quite careful in setting up tests, but of course it's subjective and it is what it is.
It would be great if everyone on the forum had really good measuring equipment and was familiar with using it, but this isn't the case and so those that use the forum contribute what they can in good faith and share out information. We're all grateful to those who do have sophisticated equipment because this is electronics and we need the solid data.
It's a debatable point but as a musician I put a bit more weight to the subjective listening accounts of other pro musicians or experienced amateurs who gig frequently since they are highly trained in music and have a lot of first hand experience of the timbre of acoustic instruments. Playing music is all about critical listening, and a lot of that is about tone and timbre. As I say this is a debatable area and others like studio engineers are trained and experienced in critical listening so it's not exclusive to gigging musicians. Comparing listening notes with other musicians over the years seems to point to their favouring acoustic timbre and tone as a focus. Less or virtually no significance is put on soundstage and electronic instruments (rock guitar etc) even though this is frequently discussed by other listeners.
.
It would be great if everyone on the forum had really good measuring equipment and was familiar with using it, but this isn't the case and so those that use the forum contribute what they can in good faith and share out information. We're all grateful to those who do have sophisticated equipment because this is electronics and we need the solid data.
It's a debatable point but as a musician I put a bit more weight to the subjective listening accounts of other pro musicians or experienced amateurs who gig frequently since they are highly trained in music and have a lot of first hand experience of the timbre of acoustic instruments. Playing music is all about critical listening, and a lot of that is about tone and timbre. As I say this is a debatable area and others like studio engineers are trained and experienced in critical listening so it's not exclusive to gigging musicians. Comparing listening notes with other musicians over the years seems to point to their favouring acoustic timbre and tone as a focus. Less or virtually no significance is put on soundstage and electronic instruments (rock guitar etc) even though this is frequently discussed by other listeners.
.
Last edited:
I don't have a problem with opinions, but after 130 pages and 1300 posts, does this really have anything to do with post #1 and Wavebourn's original proposal, or have any insight into why or how we should value, enhance, dismiss or enjoy what was put forward?
Unless the discussion is demonstrably benchmarked against Wavebourn's work, it really ought to be in its own thread or topic.
kind regards
Marek
Unless the discussion is demonstrably benchmarked against Wavebourn's work, it really ought to be in its own thread or topic.
kind regards
Marek
This has been the go-to thread to discuss using 4P1L outputs in single ended amplifiers. That's what we have been doing throughout the thread. Comparing 4P1L outputs to 2a3 or 300b outputs would seem to me to be completely relevant. In fact Wavebourn himself started off by comparing the 4P1L to a 2a3 as an output. In my case I've been posting multiple times about SE 4P1L builds together with circuits and construction details. Kenev is talking about one of his builds. Other contributors have been talking about their own builds from post 31 onwards, and post 31 was back in 2011! I don't understand your objections here. Everything has moved on since the original post, and this has been the thread to read for 4P1L builders ever since. Many thanks to Wavebourn for starting it.
.
.
Last edited:
Andy, I started the thread to share my impressions about a "new" great tube that almost nobody knew about. The thread evolved into lots and lots of experiments and wonderful projects. Now the tube is popular, so it is normal that people argue pro and contra something they heard about. The more people know about the tube, the more opinions we have as the result.
The point is very simple. This thread was about being able to have 2-4watts of almost 300b grade performance simply for just a few dollars. The relevant direction or comparison now would be find how to use some equally linear and inexpensive (TV sweep tubes?) or components no one particularly values or learn about a simple, cheap and freely available "approach" or topology that lets you massively outperform relative what is the accepted norm.
After 300 pages, it is a great disappointment and totally missing the point to simply be led towards a "which Llundahl, Mundorf and 300b sounds best" thread. That belongs in a different topic.
kind regards
Marek
A quick update on the subject.
After some extensive listening tests, I have been back in preferring the 4P1L over the 2A3. Since I had written previously that the 2A3 beat 4P1l, I have to take it back. Of course, I have to note that the 2A3 was not "optimally" loaded, as I used the same 5K OPT that I use for the 4P1L. But, anyway, currently, I consider that 4P1L is much better.
For the moment, I'm waiting for a new pair of anode load chokes for the driver stage and a pair of OPTs. Both are self designed and are being wound by a friend of mine. Also, both will use some "exotic" amorphous double-C cores.
Soon as I try the new irons, I will keep you informed.
After some extensive listening tests, I have been back in preferring the 4P1L over the 2A3. Since I had written previously that the 2A3 beat 4P1l, I have to take it back. Of course, I have to note that the 2A3 was not "optimally" loaded, as I used the same 5K OPT that I use for the 4P1L. But, anyway, currently, I consider that 4P1L is much better.
For the moment, I'm waiting for a new pair of anode load chokes for the driver stage and a pair of OPTs. Both are self designed and are being wound by a friend of mine. Also, both will use some "exotic" amorphous double-C cores.
Soon as I try the new irons, I will keep you informed.