Have others built versions of these amps using your preferred 6AU6? What were their conclusions?
Out of curiosity, did the RH807 you built use exactly the circuit that Alex designed? If not, could you tell us what the changes were?
Well, this is not entirely correct, and the reason is what you have written yourself: the output transformer is in the loop (global) or not in the loop (local). There are other elements to be taken into account that make it not identical.
I was trying to simplify things to show how one could get the same result in terms of distortion and taming the superfluous - but it is a better approach to use local instead of global feedback.
The diatribe regarding this issue could go on and on without end. I just want to point out what I actually wanted to state.
We are missing something here - Alex is not selling anything, and neither am I. Maybe you are selling something, and thus imagine everyone else might be selling?
Taxative reply:
1) Regulator circuits have a better regulation under load - the extent of the regulation needed is application dependent: a class A circuit has little change in current draw if any, while a class B circuit has large changes in current draw passing from idle to maximum power. SE amps are class A circuits by definition, thus regulation is not that much important.
2) Regulator circuits consist of active elements, and active elements generate noise. A choke does not generate noise (yes, maybe the heat... but that is more like metaphysics to the purpose of this discussion). If the regulator circuit is fixed at some voltage value, which it usually is, it has to have a voltage reference. All voltage references generate noise, some generate very little, some generate loads of it... and it is amplified by the corrective device or servo in the regulator circuit.
3) My interest in regulator circuits is mainly for ripple rejection. While a regulator circuit can achieve excellent ripple rejection at low cost
it will introduce noise into the circuit. On the other hand, if cost and/or voltage drop is not all that important, several chokes and large caps will reduce ripple just as well, will not introduce additional noise, and the circuit which we needed this clean power supply for (i.e. phono preamplifier) will sound much better with a classic power supply than with a regulation circuit.
When it comes to examples, I believe you can find a lot of these on the John Broskie site. He has written a lot about regulators - and I believe his style is quite readable and more than adequate for the education of DIYers.
While the issue "choke or regulator" is an old dispute, there is no need to further it. You are of course free to have your own choice and your own taste. Regulators have some advantages, classic supplies have others, and which one you choose will depend on several factors. But better noise is not a factor one would chooose the regulator over a classic supply, because that is one of the few details where it looses points to the classic supply.
That said, all this was about the g2 supply... and if you are after perfection (?) than a separate power supply beats the regulation approach. It is not the same and will not be. Now let me remind you that I do not consider neither regulation nor a separate supply necessary for the correct operation of an SE amplifier (RH amps are all SE as far as I know).
This seems to be the only argument yet worth pursuing.
1) You have not used words such as "rubbish" or "stinky", but you have written in denigration about the RH807 (and RH amps in general). That means you are implying Alex Kitic is incompetent.
2) Your authority is based on your own words, as you say yourself. We should beleve you just beause you have proclaimed yourself a reputable source. With all due respect, you have written that down yourself. Now how does that make you look like? Maybe you have some academic title that I am not aware of, and merely mentioning your title, which was given as a result of scientific research and the publication of your work, gives you some authority? Well I humbly apologize if that is the case - but you should let us know about it, since we have no crystal balls, just PCs.
3) It is a pity about the crystal ball (sorry, you have mentioned it yourself, I am not imagining anything here). In the case of Alex Kitic, his crystal ball seems to be a PC and an HDD - thus schematics, simulations, and similar are available - 24/7/365.
Therefore, I do not see that you are under attack here... but as I mentioned earlier, maybe it is customary in Anglo-Saxon countries to speak in denigration of others, and to pretend offence when you are presented with arguments that stand against your denigrative statement.
Is your join date your only argument? Or the number of posts?
Quantity does not imply quality.
Anyway, that depends on what you take into consideration.
I am looking forward to see that one, thanks for asking.
Attempts to use small signal pentodes as drivers for the RH amps, both RH84 and RH807 were mentioned - for instance, with EF184 as driver, or 6AU6, or 6688. I have not seen the simulated results, but there is mention of very low distortion figures. Still, the schematics that can be found in various threads mostly use a voltage source for the supply of the 2nd grid, which is not what will happen in practice, since in small signal pentodes the 2nd grid will have a large influence on both amplification and current drawn, and the decoupling of elements will have an influence on the frequency response.
It is thus more than likely that these attempts have mostly been unsuccessful, or not satisfactory (at least in the "sounds" department, since most people do not own FFT analyzers and stuff).
Once we see the schematics applied, there will be a nice simulated schematics for an all pentode RH84 with 6AU6 driver (complete with g2 resistors etc.). The schematic is there from the beginning, and does not change much for the RH84 - it does not drive it to higher undistorted power than the original 12AT7 driver.
The main reason for that, of course, is the fact that a 12W dissipation tube can only produce as much power as possible, there is no perpetuum mobile here which will be triggered by the adoption of a pentode driver, which is what one faction of the RH amp critics would like you to believe.
The other faction of RH amp critics would like you to believe that those amps are "no good" because they are designed in a different way than they would do it.
Either criticism would be interesting if it were corroborated with at least some simulation results, if not FFT analysis, screenshots... and of course, comparable designs showing the same data, but better.
That would, of course, be a good technical comparison of merit. When it comes to sound, well, that would probably need some comparison in public, and would probably come down to matters of taste (or would not, everyone likes good solid bass, silky highs, etc. so tastes actually tend to converge).
But the problem is that there is criticism, while there are no:
1) shown data (simulation, analysis, etc.);
2) comparable designs (complete with the above data).
No, correct me if I am wrong here: while criticism has driven the advance of civilization (in a way, since competition has, and it is a form of criticism), criticism without arguments is more lake denigration, envy, or hatred.
I built mine as a parafeed version using alternately a pentode plate load and a choke plate load.
I abandoned SE after that in preference for Class A fully differential PP using Plate to Plate feedback.
My first design was a clone of Gary Pimms Tabor amp, which I consider shows the correct application of a Schade feedback, and comes from an expert electrical engineer.
I will build my first new SE amp in a few months using the single Trioded E55L with an input transformer to allow for Schade feedback ala the original Schade article.
All my designs (apart from an ECL82 headphone amp which used Plate to Plate feedback and a triode driver - but runs at such low swings the issues are minimised) have used pentodes as the driver in plate to plate PP amplifiers.
Shoog
I abandoned SE after that in preference for Class A fully differential PP using Plate to Plate feedback.
My first design was a clone of Gary Pimms Tabor amp, which I consider shows the correct application of a Schade feedback, and comes from an expert electrical engineer.
I will build my first new SE amp in a few months using the single Trioded E55L with an input transformer to allow for Schade feedback ala the original Schade article.
All my designs (apart from an ECL82 headphone amp which used Plate to Plate feedback and a triode driver - but runs at such low swings the issues are minimised) have used pentodes as the driver in plate to plate PP amplifiers.
Shoog
I see this as "no schematics".
The RH807 is neither a parafeed amp nor pentode plate load amp (at this point it is not clear whether we are speaking of the output tube having a pentode load, or the driver tube?).
Also no mention of driver type: 12at7 triode as per original schematics, or some pentode?
This is a case of "I built it as I wanted, putting whatever I had in mind wherever I saw fit. It didn't work out, must be a fault of the guy who designed the darn thing in the first place!"
What you built was definitely not an RH807. If you weren't happy with it, it's not Alex Kitic's fault. And at that point you jumped in on the haters' bandwagon...
BTW, someone should explain what is a clone of someone's amplifier? I believe that should be called "building an amp according to the schematics designed by someone".
What is a fully differential push-pull? Isn't every PP amp differential?
Does it mean that it has no phase splitter and accepts only balanced, i.e. differential signals?
The original amp which I built based upon the RH807 was exactly as shown in the origional schematic but with a pentode loaded output stage. You will see that this should have no effect on the overall performance of the design since the dominant output load is still the OT and not the pentode - which simply disappears since the transformer is running at 8K in parallel with the pentode at over 100K.
My designs since have been Input transformer - driving a LTP direct coupled to Output pentodes in a LTP configuration. True differential requires the Output to be a LTP and as such you will realise that most PP amps are not in fact differential front end to back end.
Shoog
My designs since have been Input transformer - driving a LTP direct coupled to Output pentodes in a LTP configuration. True differential requires the Output to be a LTP and as such you will realise that most PP amps are not in fact differential front end to back end.
Shoog
Classic parafeed is not the same as output transformer coupled. There are several potential problems, including the cap that couples the output stage loaded with a choke and the output transformer. The same is true, of active device loading. The dissapearance act is more a verbal figure than reality. Btw, the transformer was not meant to be 8k.
The circuitry you are describing has little to do with the original RH807 which requires an output transformer capable of 50mA DC current. No loading devices were taken into account and no coupling cap was specified.
It seems however that others were happy with their implementation of parafeed in RH amps, like Planet10.
Anyway, the parafeed arrangement, CCS and other active device loading strategies are not associated with RH amps, but rather with Gary Pimm and Bottlehead.
Plate to plate feedback is a feature of RH amps. Thus your amp was a heavily modified RH807, and it is objectionable whether you were able to do that mod correctly. Once you followed the schematics closely, like in the Tabor case, you were obviously more happy with the results.
Btw, check for the meaning of fully differential amplifier on Wikipedia if nowhere else, since it strays slightly from what you would like it to be (it has a ring to the name, sellable indeed).
Alex, the parafeed cap if correctly sized is transparent to the performance other the fact that it is a cap. The load is still the same - the performance is still the same - but bass and treble extension will be considerably better than most of the transformers used to implement the RH807. You will have to forgive my mistake on the 8K, its so long ago I forgot the precise detail of the transformer specified - but rest assured I used the correct load as shown.
I am firm in my belief that what people like about the RH807 is its relatively constrained frequency response (due to the small transformers generally used) and lack of deep bass control coupled to a healthy injection of 2nd harmonic distortion. That is what most people seem to want from a valve amp so all power to their elbow, but it ain't HIFI as such and I moved on taking the basic good idea and getting it to do what it does best. most DIYers run sensitive speakers so they rarely push the amp into the range where it starts to measurably degrade the signal (as measured in the link to "Why a pentode ?"). What repeatedly comes out of discussions of the RH amps is the choice of a ECC81 might not be so bad because of distortion cancellation - but why design using a sub optimal part when its so easy to use the correct pentode.
Shoog
I am firm in my belief that what people like about the RH807 is its relatively constrained frequency response (due to the small transformers generally used) and lack of deep bass control coupled to a healthy injection of 2nd harmonic distortion. That is what most people seem to want from a valve amp so all power to their elbow, but it ain't HIFI as such and I moved on taking the basic good idea and getting it to do what it does best. most DIYers run sensitive speakers so they rarely push the amp into the range where it starts to measurably degrade the signal (as measured in the link to "Why a pentode ?"). What repeatedly comes out of discussions of the RH amps is the choice of a ECC81 might not be so bad because of distortion cancellation - but why design using a sub optimal part when its so easy to use the correct pentode.
Shoog
Last edited:
I build amps for the purpose of listening to music. What matters most is the sound. A sound design will lead to good sonic results. After some theory and schematics and insight, comes components tweaking.
The result is a grat listening experience. That's what it's all about, for me.
It is no different for me and that is why I use coupling transformers - eliminate cathode bypass caps and coupling caps. I can guarantee that the result of my E55L implementation will be blissfully transparent and detailed. It will be easy to build and relatively cheap and produce about the same power as an RH84. There will be no compromises.
Shoog
The parafeed promise depends on the following:
1) the plate choke used is superior in inductivity to the primary of the output transformer that would be used;
2) the output transformer is superior in quality to the gapped SE transformer that would be used, possibly nickel plate;
3) absolutely high quality coupling cap (usually oil due to the high voltages and large voltage swings involved.
The presence of a cap is enough to imply a compromise. While I am not all in favor of DC coupling, or transformer coupling, the "sound" or rather influence of caps on the sound has been widely shown, discussed, etc.
If the above 3 points hold true, you could basically build an SE amp of better sonic quality in comparison to the same amp in classic gapped OPT arrangement. The same is true of an RH amp as well. BUT, if any of the above falls out of the equation, you have achieved nothing. Furthermore, economically, the cost of high quality iron and caps actually outweighs the cost of a good gapped OPT.
If instead of a plate choke an active device is used, well, there are no guarantees. A CCS is not necessarily the best solution. While it might have merits in voltage gain applications for small signals, where such application is sound or logical, why not... but as a load for an output device?
Point is, a choke lets you swing that plate above the B+. The same is true of the primary of the output transformer. An active device does not allow you to swing above the B+. Thus the whole game play gets a different perspective - besides needing higher B+, the active device used as load will undoubtedly have its influence on the functioning of the circuit and the sound of such amplifier. Whether one will like it or not might come even down to personal preferences... but it is definitely not the same amplifier... not even similar, to me (unlike the choke loaded case which depends on the quality of the components).
I would say that it is obviously what you have experienced - constrained frequency response and lack of deep bass (control). I would not like that either. Quite the opposite, all my RH amps (I have got several) and all those I have heard, as well as the written accounts of those who seem to like RH amps, have excellent bass control and an extended frequency response. What you are describing is what I dislike in many (mostly triode) SE amps, and is due to poor dumping factor and probably a poor choice of components and/or power supply.
Nevertheless, your personal preference for PP amps, parafeed, active parafeed, CCS, etc - does not justify "foul" language and denigration in tone and attitude.
Just as I said, if you were not happy with an amplifier you built, it is not necessarily the fault of the designer. What you built was not an RH amp. Some people build those amps and make mistakes in soldering the components - after finding out what was the problem, and having the amplifier function properly (like, no more hum or similar) they remain happy with the results. If they overgrow the amplifier, they might change it for another, but still no bad words or criticism can be heard. Except from "haters". There is really no need nor justification for the "hater" behavior you have shown over the years on this forum whenever an RH amp was mentioned.
Cool, man, can't wait to see that!
Is it a competition? Mind you, I heard the RH84 has had an official revision
RH Amplifiers: February 2013
and it seems that new versions are soon to come...
Alex, I think you need to calm down again.
You have consistently responded poorly to the many genuine critiques of your design approach and so I am not that inclined to give you the benefit of the doubt when you seem so closed to the advice of others who have shown considerable design talent in their own right.
Maybe you should stop been so defensive. I believe your approach has got you banned before and hence the new sock puppet incarnation to reenter the debate.
The original point of all this is that someone produced a pentode input design and then referred it back to your RH designs, in my experience its superior and you don't deserve the reflected glory. Can I suggest that this is an example of a cheap plate to plate feedback amp which is better designed;
http://www.pmillett.com/file_downloads/midget_sch.pdf
Shoog
You have consistently responded poorly to the many genuine critiques of your design approach and so I am not that inclined to give you the benefit of the doubt when you seem so closed to the advice of others who have shown considerable design talent in their own right.
Maybe you should stop been so defensive. I believe your approach has got you banned before and hence the new sock puppet incarnation to reenter the debate.
The original point of all this is that someone produced a pentode input design and then referred it back to your RH designs, in my experience its superior and you don't deserve the reflected glory. Can I suggest that this is an example of a cheap plate to plate feedback amp which is better designed;
http://www.pmillett.com/file_downloads/midget_sch.pdf
Shoog
Last edited:
I don't understand what are you trying to say? What do you mean by poor response? And, why should I accept the advice of others? I do not recall any advice being asked for nor given. Further, I usually take good care whose advice I take...
Your insistence is to say the least, boring. It seems that someone has asked for your advice on something? I have not.
Just like I do not recall asking for or being given advice, I do recall harsh words. For some reason, it seems that people like you or other similar minded characters would like me to be either quiet, or away - not interfering with their turf? Am I such a treath to your interests or what?
Regarding the credits, I do not think that the amp that started the thread has anything to do with RH amps, apart from the feedback resistor between the plates. Actually, the author wrote himself that he just wanted to appeal to the DIYers who find the RH amps popular.
The posted link leads to an interesting amp by someone who obviously knows how to design with pentodes. Interestingly, P. Millett does not consider the 307A particularly interesting in pentode mode - while my experience is quite the opposite, and find that it is a great sounding pentode. Anyway, the 6T10 amp seems fine. I cannot say which level of optimization the schematics has reached. I would prefer vacuum rectifiers (in this case hybrid graetz, I use it often) but it would not yield enough B+ with the specified transformer.
It should be taken into account that this is a combined tube, which is maybe easy to source, but definitely not present as current production and thus less universally applicable than ECC81 and EL84. The power is slightly lower, and you get what you get, in the sense that you cannot try a different make or type of driver, or output tube (no tube rolling).
While similar amps can be executed with ECL82 and ECL86, I was never interested due to lower output power and no tube rolling possibility.
On that account, I believe that by now it is widely understood that an RH amp can be made with many drivers in mind. Just take a look at the RH Universal ver2.
Last edited:
Per forum policy previously banned members who set up new accounts for the purpose of getting around moderation are automatically banned.
Just couldn't help myself, as the Kitic-gate is one of the more colorful episodes in tubeland in recent years, so I will add my two bits here as well - if only for purely amusement purpose.
Mr. Kitic has repeatedly claimed that he was not aware of any known design that used the plate-to-plate feedback exactly as he has done it with the RH amplifiers. Perhaps not with exactly the same parts, but in the Hugo Gernsback's 1947 Amplifier Builder's Guide, there were several designs using the exact plate-to-plate feedback as the RH amplifiers. Anyway, Mr. Kitic should not be held accountable for this oversight, who can say that he/she has seen or is aware of all the previous designs that exist? No one can, so it is certain that he came to the RH amplifier designs on his own.
But after many suggestions to Mr. Kitic that there was nothing really novel about the plate-to-plate feedback, that was when things went haywire and the grudge matches began. Mr. Kitic has went so far to suggest that the plate-to-plate feedback be called a Kitic or RH amplifier feedback, who the heck is this Schade guy anyway?
In the Gernsback text, the subject of feedback was treated as a non-event, it simply stated that the output tube's plate signal was fed back to the GRID, that's all. Which was what many have tried to convey to Mr. Kitic over the years, but of course, most of them were simply "not knowledgeable" in electronics, so really had no clue as to what they were saying...
Now why don't we all just go and build the RH Universal and shut the @#$% up... 
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.