I have both PP and PSE amp and they are both good for different type of music.
They can be run using EL34, 6550, KT88 but to bias differently for different tube.
In stock form the PP amp run in UL but after I changed it to triode (which I prefer sonically) I never change it back to UL.
For the PSE, at first I run it in triode and I like it very much, since the builder of the amp provided the switch for changing triode and UL mode, I switched to UL mode and I found the sound is better, with wider stage and better bass.
What I would like to suggest is, do provide the switch for UL and triode, you'll know later how useful it is.
If you're using 90+ dB speaker, you dont have to worry to run your PSE in triode mode, mine is more than enough
Bob
They can be run using EL34, 6550, KT88 but to bias differently for different tube.
In stock form the PP amp run in UL but after I changed it to triode (which I prefer sonically) I never change it back to UL.
For the PSE, at first I run it in triode and I like it very much, since the builder of the amp provided the switch for changing triode and UL mode, I switched to UL mode and I found the sound is better, with wider stage and better bass.
What I would like to suggest is, do provide the switch for UL and triode, you'll know later how useful it is.
If you're using 90+ dB speaker, you dont have to worry to run your PSE in triode mode, mine is more than enough
Bob
Hi Moose,
I built this se EL34, it gives 8 watts per ch and has a lovely sound. It has a unique adjustable feedback circuit which I have not seen before in diy projects. The bass is outstanding for a small amp. I use very inefficient speakers, 86db but it plenty loud enough. I have tried it in triode mode also but it sounded just the same. The author gives an excellent account of the build and even if you don't build it it is worth a read. It is also very easy to build.
http://www.lundahl.se/claus_b_se.html
You can buy the transformers from here: http://www.kandkaudio.com/
I built this se EL34, it gives 8 watts per ch and has a lovely sound. It has a unique adjustable feedback circuit which I have not seen before in diy projects. The bass is outstanding for a small amp. I use very inefficient speakers, 86db but it plenty loud enough. I have tried it in triode mode also but it sounded just the same. The author gives an excellent account of the build and even if you don't build it it is worth a read. It is also very easy to build.
http://www.lundahl.se/claus_b_se.html
You can buy the transformers from here: http://www.kandkaudio.com/
alexmoose said:
There is a whole bunch of audiophool folk "wisdom" out there, much of it just plain wrong. Here is a good case in point. There is nothing the least bit "scarey" about operating in pentode mode. As with anything else, there are design trade-offs. Pentode mode offers the highest efficiency and power output. Pure pentode mode has the disadvantages of poor bass response, due to the lack of speaker damping. It also tends to make the highs more "strident", and that can be most annoying to listen to, depending on program material (I found that Ozzy was the worst in this regard).
On the plus side, pentode mode greatly eases the burden on the front end. Operating this way requires much less control grid swing than would a comparable low u power triode. This allows for more "headroom", and voltage gain stages have increasing distortion with increased output voltage swing. Pentodes also have a greatly reduced Ci + Cmiller so that the drivers aren't so burdened with having to source current into that input capacitance. This improves the slew rate at the upper end.
The problems with pentode mode are easily correctable by applying NFB locally to the finals. This can be via parallel, cathode, or screen (i.e. Ultralinear) feedback. This NFB will reduce the effective r(p) of the pentode finals, improve woofer damping, linearize the stage so that you won't be getting all that higher order harmonic distortion, and will make the dynamic characteristics of the finals more triode-like. If you're going Class AB(1) or AB(2), be sure to include decent screen voltage regulation, preferrably with an active regulator, or at the very least, with gas discharge voltage regulators or well bypassed Zener strings. For single-ended Class A(1) operation, a stiff voltage divider will suffice. Add some corrective gNFB, and there is absolutely no reason why pentode mode amps can't sound every bit as good as amps that use triode finals.
I agree with Miles but many would not. Why? Because, as Miles correctly points out, pentode mode implies significant NFB and there is a significant body of people that abhors NFB, saying it deadens the sound. Menno Venderveen is so anti-feedback that one of his EL34 PP amp designs, which offers a choice of the usual three modes, pentode, UL and triode, includes no added NFB loop at all. He admits that it might be difficult to find speakers that would sound good with the amp in pentode mode (and I'm sure he's right about that). However, he doesn't mention the distortion problems of pentode mode, which, as Miles says, is normally addressed by NFB too.
The defenders of NFB, e.g. Norman Koren, make counter arguments, saying that NFB has too often been applied incompetently. As a result, it has been given a bad name. There are important benefits, they say, if NFB is applied correctly.
As to which camp you support, that will probably depend on your own experience and, importantly, the type of speakers you have. If you decide you don't like NFB, then you will probably want to avoid pentode mode.
Personally, I am suspicious of extreme opinions and I regard the almost superstitious avoidance of pentodes and the attendant NFB as silly. On the other hand, the bald statements made half a century ago by HiFi manufacturers, that total harmonic distortion must be less than 0.1%, NFB should be applied to the maximum possible extent and damping factor should be as high as possible, seem silly as well. As with most things, the truth probably lies somewhere in between and the use of NFB only to the extent that it is beneficial seems sensible to me.
The defenders of NFB, e.g. Norman Koren, make counter arguments, saying that NFB has too often been applied incompetently. As a result, it has been given a bad name. There are important benefits, they say, if NFB is applied correctly.
As to which camp you support, that will probably depend on your own experience and, importantly, the type of speakers you have. If you decide you don't like NFB, then you will probably want to avoid pentode mode.
Personally, I am suspicious of extreme opinions and I regard the almost superstitious avoidance of pentodes and the attendant NFB as silly. On the other hand, the bald statements made half a century ago by HiFi manufacturers, that total harmonic distortion must be less than 0.1%, NFB should be applied to the maximum possible extent and damping factor should be as high as possible, seem silly as well. As with most things, the truth probably lies somewhere in between and the use of NFB only to the extent that it is beneficial seems sensible to me.
Pentode VS triode, NFB or no feedback, SE or push pull? There are valid points, and strong opinions on both sides of each of these. I don't think that too many people are going to change their opinion based on a few words on the internet either.
I also have both SE amps (3 currently) and push pull amps (2). If space was not a problem I would have more. They each have their characteristic strengths and weaknesses, and I currently have 3 amps (2 SE and 1 PP) set up so that they can be swapped in under a minute, so I can match the amp to the music and my mood. Most of the time I used my TubelabSE, but I recently swapped it out for a SimpleSE (very similar to your amp) with big Hammond 1628SEA transformers. My second choice is my 300B push pull amp (30 watts) these two amps handle 99% of my musical demands. The 845SE gets very little use. I have an 80 watt screen driven sweep tube P-P amd on the breadboard, that may unseat the 300B P-P.
Lets see, you already have an SE amp that probably makes about 5 watts per channel. From earlier conversations, I know that some of your musical choices are rather dynamic, so to get a significant change in sound, you are going to need more power, and an output transformer capable of better bass response. It has been stated that you must double the power to even hear a perceivable difference. This is pretty much true. My 40 watt SE amp is not 8 times louder than my 5 watt SE amp, it is more like twice as loud, especially standing back from the speakers.
You could build another SE amp, but to get a significant powr increase you need to parallel several tubes, or use specialized tubes. You will also need a much larger output transformer. This will not be cheap, and it will sound similar to the amp that you have although louder and more dynamic with better bass. This may be what you want, but I think you would be better off upgrading the transformers in that amp when you have the budget. You could also try pentode mode in your amp, but it WILL need some feedback. I tried it in my amp, and I discovered that cathode feedback (limited to about 4 db) was not enough to control the speaker. UL with CFB works very well and that is the way the amp is currently connected.
When I really want to play some dynamic music I use my P-P amp. It is a triode amp that I built a long time ago, but it rocks in a way that an SE amp can't. Very useful for Ozzy, Van Halen, Metallica, etc. This is a no feedback triode amp, not because I am against pentodes or feedback. I set out to build a 300B P-P amp, and these tubes have a low enough plate resistance and good enough linearity that feedback is not needed.
So, if you want to build another amp (I told you that this hobby was addicting), why not make a push pull amp? A push pull OPT requires much less iron than an SE OPT of equal power, so it is possible to make good bass with a reasonably priced transformer. A 30 watt P-P amp will be quite a bit cheaper than a 30 watt SE amp, for this reason. If you choose to build a P-P amp, why not shoot for as many of the other possibilities as practical. It would be relatively easy to make one amp capable of operating in triode, pentode and UL modes (assuming you use pentodes and an OPT with UL taps). Switchable feedback is possible but usually requires a scope and a square wave generator to set up. I built a 30 WPC KT88 P-P amp with variable global feedback (0 to 10 db) a while back that worked very well, so well that it was sold a few months after I built it.
As mentioned before, I think that an EL84 amp would only be slightly more powerful than what you have now, so I would go for something a little bigger, like EL34's or 6L6's, maybe KT88's but they tend to get pricey. As also mentioned before a class A push pull amp will make the same power as a SE amp with two tubes in parallel. So set it up for class AB and get more power. You should be able to get 20 to 25 watts from EL34's or 6L6's in triode with minimal feedback (as mentioned before EL34's work well in triode), and close to 50 watts in pentode mode.
Yes, but a little goes a long way. I have heard the same effect with Ozzie, and some other vocalists. I think that pentode mode without feedback, or with incorectly compensated feedback, is highly affected by the speakers impedance curve. Some combinations of music and speakers just sound irritating.
you guys seem really passionate about this, I really appreciate the response I am getting here.
I actually met another 18 year old tube head who lives just down the hall from me, he is a little more expirenced, and is a guitarist, but I have been talking to him alot about what kind of a design to build. I am actually now doing some serious thought about a PP 6l6GC design, because I have heard it has a better sound than the EL34, and it yeilds close to the same power in PP. In addition the designs are fairly simple. anyone have any additional comments?
-Moose
I actually met another 18 year old tube head who lives just down the hall from me, he is a little more expirenced, and is a guitarist, but I have been talking to him alot about what kind of a design to build. I am actually now doing some serious thought about a PP 6l6GC design, because I have heard it has a better sound than the EL34, and it yeilds close to the same power in PP. In addition the designs are fairly simple. anyone have any additional comments?
-Moose
Been there, built that. Although most of my 6L6 push pull amps have been for guitar playing, the 6L6 will make a good sounding HiFi amp too (thousands were built in the 40's and 50's). It is my opinion that the EL34 sounds better in push pull triode mode.
I know that others will probably recommend against it, but I routinely swap EL34's and 6L6GC's in the same amp (including my SimpleSE). If you use adjustable bias and a transformer between 4K and 6.6K (for P-P) you won't even violate the specs. With cathode bias, the 6L6GC's will run hot (at spec or slightly over) if the amp was set up for EL34's. Either way I would include the ability to run triode or pentode mode, UL if your transformer has the tap. The Hammonds and the Edcors have the tap, so do the Transcendars available on Ebay.
Either way (6L6GC or EL34) this amp will rock. It is not too hard to get 40 watts per channel with 6L6GC's, 50 if you push it.
Hi Alex,
Sound is in the ear of the beholder, the 6L6 is preferred by a lot of guitarists over the EL34 for its behavior when overdriven amongst many other things. ("Tone, Crunch, Sustain, etc." It's kinda like wine tasting.. ) It is not necessarily the case with hifi applications, although imho the 7581 is a pretty good sounding tube and quite similar to the 6L6. The real standout in this family for hifi would be the KT66, now only available as Chinese reissues.. I'm told some of them sound pretty good, durability is another question, and I have never used them in a design so caveat emptor.
Sound is in the ear of the beholder, the 6L6 is preferred by a lot of guitarists over the EL34 for its behavior when overdriven amongst many other things. ("Tone, Crunch, Sustain, etc." It's kinda like wine tasting.. ) It is not necessarily the case with hifi applications, although imho the 7581 is a pretty good sounding tube and quite similar to the 6L6. The real standout in this family for hifi would be the KT66, now only available as Chinese reissues.. I'm told some of them sound pretty good, durability is another question, and I have never used them in a design so caveat emptor.
Hi Kevin,
They sound great. Also, many 7581A's are branded KT66 as well.
Another thing about the 6L6GC's is the reduced heater current. Your will draw about 0.9A as opposed to 1.4~1.6 A for EL-34/6CA7's.
I think you can make any of these sound good or bad depending on transformers and circuitry. They both can sound great.
I am partial to the Fender Twin as a head. A quad of 6L6GC's doing damage big time. But clean enough to get almost any sound you want.
-Chris
They sound great. Also, many 7581A's are branded KT66 as well.
Another thing about the 6L6GC's is the reduced heater current. Your will draw about 0.9A as opposed to 1.4~1.6 A for EL-34/6CA7's.
I think you can make any of these sound good or bad depending on transformers and circuitry. They both can sound great.
I am partial to the Fender Twin as a head. A quad of 6L6GC's doing damage big time. But clean enough to get almost any sound you want.
-Chris
tubelab.com said:
Okay, what design did you use? I really want to build one of your designs, I need to pay my homage
Is this running in UL mode? or Pentode?
has anyone built this amp?6l6 PP Class A someone asked a question about it a while ago (I think)
The Schematic:
An externally hosted image should be here but it was not working when we last tested it.
anatech said:
Very good point, as I it is pretty hard to find a transformer with a 8-10amp 6.3v rating
do you have any ideas as to a better CCS?
This schematic seems to be the frontrunner at the moment, however will it operate outside of Class A, despite the name? and Also, should I find the means, would it be too hard to adjust this design (changes to the bias circuitry, and to the B+ Voltage) to run the Design with the EL34?
-Moose
I agree with Ray. But you can still use the LM317 to set the current. Just cascode the LM317 with a transistor like a BC549 above it (between it and the cathodes). That will extend its CCS-like behavior to higher frequencies. Bias the upper transistor with a two-resistor voltage divider. Easy and cheap.
Hope lunch was good. I'm about to have mine now.
This will be a cascode with an “o”, not a cascade with an “a” – big difference. Try this: Connect a 10K resistor to ground. Tie the other end to two 4.7 K resistors in series and tie the outer end of the second 4.7 K to the negative supply. So you have three resistors in series between ground and the negative supply (roughly -15 volts, I’m guessing). Now connect the transistor’s base to the junction of the 10K and the middle 4.7K. This sets the base at roughly 1/2 of the negative supply (roughly -7.5V). Now put a 47uF or 100uF cap from the junction of the two 4.7K resistors to ground to filter noise. Place the positive terminal of the cap to ground. The cap’s voltage rating should be 25 volts or so. Now break the connection between the LM317 and the cathodes. Connect the NPN transistor’s emitter to the input terminal of the LM317. Connect the collector to the joined cathodes. Almost any high-hfe NPN small-signal transistor should work. The reason that I don’t put a filter cap right on the base is that I want a large resistive path to ground for the nasty collector-to-base capacitance, to keep the cathodes from “feeling” it so much.
This will be a cascode with an “o”, not a cascade with an “a” – big difference. Try this: Connect a 10K resistor to ground. Tie the other end to two 4.7 K resistors in series and tie the outer end of the second 4.7 K to the negative supply. So you have three resistors in series between ground and the negative supply (roughly -15 volts, I’m guessing). Now connect the transistor’s base to the junction of the 10K and the middle 4.7K. This sets the base at roughly 1/2 of the negative supply (roughly -7.5V). Now put a 47uF or 100uF cap from the junction of the two 4.7K resistors to ground to filter noise. Place the positive terminal of the cap to ground. The cap’s voltage rating should be 25 volts or so. Now break the connection between the LM317 and the cathodes. Connect the NPN transistor’s emitter to the input terminal of the LM317. Connect the collector to the joined cathodes. Almost any high-hfe NPN small-signal transistor should work. The reason that I don’t put a filter cap right on the base is that I want a large resistive path to ground for the nasty collector-to-base capacitance, to keep the cathodes from “feeling” it so much.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.