Rerurn to Romy the Cat's Site

Audio Discussions
Topic: RH Amplifiers and superfluous acrobatics.

Page 1 of 2 (31 items) 1 2 »


Posted by noviygera on 12-06-2014
fiogf49gjkf0d
Hello,

I've never had an SET amplifier in my system. Having high efficiency horns (106 db on average down to 100Hz ) I have tried a number of solid state and tube amplifiers, and I am aware of their sound in my system. I can clearly describe what I like and don't from each of those amps. So far I am not happy with any I tried.

What is so special about the SET sound? Is it possible to generalize how it compares to Class A solid state or Triode mode tube amp? I think SET for me is worth trying out but since I am using active crossovers, to put three SET amps for trial, without any sense of direction into my system is a mistake. So I would like to gather some fundamental information about these SET species. Maybe it would help to narrow down on which one (which kind) to try first in my system.

How would you approach this choosing an SET for your horns? (Lets say you know nothing about their sound and  how much SET power you need).When Romy mentions DSET, this becomes another option but it is unclear to me: how is SET converted into DSET?

Gera

Posted by Paul S on 12-06-2014
fiogf49gjkf0d
Gera, there is already a LOT of discussion on this topic salted around this site. I suppose you do know, but a couple of things you said made me wonder; anyway, at the risk of offending you (and not meaning to): "SET" means "single-ended triode" (as opposed to push-pull triode). "DSET" means "dedicated single-ended triode", and around here it includes built-in, line level/per-channel X/Os. For me, the Big Deal with SET sound is unrivaled clarity,, tactile immediacy and dynamics, along with outstanding textural and tonal capabilities - within the amp's comfort zone. This latter consideration is the reason for DSET, also for HE speakers, to try to keep the SET comfortable. Be warned that most SETs do not like complex X/Os, or back EMF, and in the "real world" these considerations may, for "practical" reasons, swamp SET benefits.

Best regards,
Paul S

Posted by rowuk on 12-06-2014
fiogf49gjkf0d
Gera,
I think that it is impossible to "generalize" what a specific amplifier topology is. There are too many variables to really pin it down.

When we have any amplifier without feedback, it is VERY sensitive to the speakers connected. This means a good SET amp with the wrong but good speaker sounds bad.

There is an often voiced oxymoron: Full Range SET. This term is used by those that have no real low frequency reproduction.

Assume however that you get a perfect match between your SET amplifiers and speakers. Will the fine output stage now highlight bad things further up the playback chain? I am concerned because you said that you were not happy with anything that you have tried. That is normally a sign that it is NOT the hardware as any playback will be OK in some aspects and the real warts become obvious first AFTER you have had a chance to live with "better" for a while. If you are using DSP crossovers for instance, you may never be happy......

I am not using SET amplifiers rather SEP (Single Ended Pentodes) with feedback. I can also describe things that I think could be better and how I would get there, but I am happy with the current result and feel no need to change anything soon in the amplification.

Posted by noviygera on 12-06-2014
fiogf49gjkf0d
Considering all "cautions" mentioned by Paul and rawuk regarding SET capriciousness to crossover and speakers I am confident that I will NOT have any problems related to the above cautions of crossovers or speaker loads.  I use an active line level crossover in front of the amps and all my horn channels are above 10 ohms.  However, there may be other problems...
1. So DSET is just a SET with line lever crossover (active or passive)? 2. For now, I can try to put HIGHs, MIDs, MIDBASS (lower limit 100Hz) on SET. I believe I do not need three 40 watt SET to heat my room.  
So, on to other problems or peculiarities to consider when choosing the first SET amp: Again is possible to generalize here to narrow down: For example, there are a few of EL84 based SET out there, 1-2 watts of power. Yes I have efficient speakers, but still, 1-2 watts. Can this be enough even for multi-way horns?
For sure someone that has horns and tried a few different SET can give me a sense of direction.
regards, Herman

Posted by Paul S on 12-06-2014
fiogf49gjkf0d
Romy seems happy enough with his flea-powered DSET. But, again, each channel is well-matched to its task, and Romy gave a lot of thought and development time before deciding on a driver/output tube, and there was more to his decision than power requirements. I don't know the EL84 at all, so I can't address it. In any case, how that tube is heated and loaded, overall PS and OPT, etc., will also factor in the sound you get from it. I ass-u-me your active X/O is buffered in and out, with a very low output R.

Paul S

Posted by noviygera on 12-06-2014
fiogf49gjkf0d
anyone familiar with the sound of this amplifier RH-307a?

http://rh-amps.blogspot.com/2013/09/rh-307a-v2-super.htm

Posted by Romy the Cat on 12-07-2014
fiogf49gjkf0d
Well, Gera, a good question in context of your playback. How a person who has multichannel 106db playback with SS amplification and active crossover to explore SET domain? It is not so simple and if you do then probably it would require bringing a single set and speaker level crossover to try. It is possible to have more or less objective subjective judgment of the result but it requires a hell of experience to equate and factor in all changes in the system and to know what exactly to listen for.  To have those proper evolution doe it would take time, money and committing many mistakes and taking many wrong routes in the way. If you content with the result you have then stay what you are with exception that If I were you I would get run of active crossover and go line-level with your SS amps.

Now regarding the SETs. Generally what people said above is somehow correct but mostly not. Many things I can’t explain to you as you did not deal with SET. Let me give you just very brief and very shallow idea. You have 106db playback, it means that your system works with very limited currents. The SS amps generally and at minimum currents in particularly have high order harmonics slightly higher then SET would do. SET in class A push VERY dominant second order harmonics that makes sound “pleasurable”, kind, soft, musical and so on. With playback of limiting currents you for sure would like to have no feedback amps but it means high output impedance. With SS you do not care about output impedance and it always very low and the amp in a way do not even “see” the MF drivers. SET however has one extra tool – you can mitigate the loading of the tube (how much impedance the tube see) and by doing it you can write your own harmonic balance of your playback.

Let say you have a bass driver , MF and HF drivers and you drive it from one of three SS amps. The amps for most of the part do not care about drivers are as they push enough current. If you happened to have driver that in context of your amps and your enclosure produces a fine harmonic result (perfectly possible but rare) then you are fine. However what option you have if the harmonic, dynamic, or transient result on one of the driver for instance not good?  Change the driver? For you can but we do not build our own drivers, we get what we get… So, here is where SET comes to help.

SET is not only has much lower amount of odd and higher harmonics but SET also you to surgically tune your harmonic, dynamic, or transient to what you need and in case of DCET you can do it for each channel (which is the greatest DSET advantage). You load SET’s output a bit more and you have “richer” and “slower” sound and if you idle it more then you have “faster” and “punchier” sound. In most instanced SS amps to my ears sound like over idled SET…

There are many other interesting things in SETs. Like there are great varieties of tubes with own slightly idiosyncratic sound and those tubes change sound type in response to how they used or driven. SETs are kind of tweaky and they can do anything you want practically with limits. It does not mean that you can’t not accomplish the same with SS amps but I do not have myself and I do not have among my advisers anybody who would have the command of SS design as the level I am interested. It is not to mention that even with SS you can mitigate output impedance and change the interface how the amp “see” the driver.

His is a bit topic and I might go about it for hours but I think you got the initial taste. The biggest question to you is how much itch you have in your ass to try something different. I always advise to use the following as a rule:

http://www.goodsoundclub.com/Forums/ShowPost.aspx?PostID=432

So, do not touch anything in your playback. Listen your playback and try to formulate what you do not like. Name it. Write it down. Then listen inhalations with SET amplification. Name what you like, write it down and listen many more times. Correlate your first and second lists and run it with many different people who have more exposure and more listening experience then you do. Update your first list as your listening objective might change then. Based upon your own guts and the advice of others create an imaginary solution how your new amplification would address the frustration and objectives you have. Now is important part: test the solution. Laterally, crate a test plan with well define definitions and metrics of success. Then bring home a SET of your choose and do not listed the overall sound but rather during your listening execute your test plan trying the targeted listening techniques I’ve described somewhere at my site. If you do hit the definitions and metrics of success during your targeted listening then it is most likely that after investing into upgrade you will not be disappointed. However, you might have many wrong or superficial conclusions during your experiments, no one preserved from that. What you however will have and will develop is the sense of actions and I think it is more important then the result itself, at least to me.

Rgs, Romy the Cat

Posted by noviygera on 12-09-2014
fiogf49gjkf0d
Hello Romy,
Your responce is valuable and I have some questions: You say: >>You load SET’s output a bit more and you have “richer” and “slower” sound and if you idle it more then you have “faster” and “punchier” sound.
This is an interesting concept but What does it mean to "load" and "idle" SET and how is this accomplished without changing the speaker it's driving?
I have a pretty clear formulation of what I like and don't like about each amplifiers sound that I tried in my system. Also, not just the first impression of sound but long term settled down feeling that this amp established in the room. Like how it sets the mind.
And now I am very curious to try something different -- SET, and how SET sets my mind. I do not know what to exactly expect. So the question is what SET should I bring home. Maybe Single ended pentode, but in either case it makes no difference because I do not know both. And with this in mind, there has to be way to narrow down to one amplifier to try. I don't have a friend with SET to let me try. I thought to start by setting the lower power output limit.
I do not like to limit myself to only quiet listening. With 1-2 watts amps I don't know what to expect even with 106db speakers. There is a reason you probably do not use 2 watt amplifiers?
regards, Gera

Posted by Romy the Cat on 12-09-2014
fiogf49gjkf0d
 noviygera wrote:
This is an interesting concept but What does it mean to "load" and "idle" SET and how is this accomplished without changing the speaker it's driving? 
   
The "loading" and "idling" in SET is amount if impedance the out tube plate is loaded upon. Mostly it managed by the ratio of output transformer, that is why many output transformer have multiple taps, or even better the re-mapable section, that you can change the ratio of the transformer and effectively how much load the tube "sees". There are many other ways to do it of couse.  
 noviygera wrote:
And now I am very curious to try something different -- SET, and how SET sets my mind. I do not know what to exactly expect. So the question is what SET should I bring home. Maybe Single ended pentode, but in either case it makes no difference because I do not know both. And with this in mind, there has to be way to narrow down to one amplifier to try. I don't have a friend with SET to let me try. I thought to start by setting the lower power output limit. I do not like to limit myself to only quiet listening. With 1-2 watts amps I don't know what to expect even with 106db speakers. There is a reason you probably do not use 2 watt amplifiers?  
   
I do not know what ask and I would anyhow stay away from any specific recommendations.  You live in Chicago and it is not an Antarctica – you will find someone. I doubt that you need to make those SET experiments. I do not what SS amps you use but if they are good amps then getting rid of active crossover and go line-level shall give you very worthy and very predictable benefit then chasing a murky SET ideas.

The Cat

Posted by rowuk on 12-12-2014
fiogf49gjkf0d
I have been using this amp for almost a year. In the context of my system, it works very well.
 noviygera wrote:
anyone familiar with the sound of this amplifier RH-307a?

http://rh-amps.blogspot.com/2013/09/rh-307a-v2-super.html

Posted by noviygera on 12-12-2014
fiogf49gjkf0d
This is the amp I chose and just started building! Would it be possible to contact you directly, rowuk? I want to consult about the build and your experience, without flooding this post with specific questions. Or you can contact me.
thanks,Gera

Posted by noviygera on 12-12-2014
fiogf49gjkf0d
Romy,
>> getting rid of active crossover and go line-level shall give you very worthy and very predictable benefit then chasing a murky SET ideas.
This is exactly what I plan to try by moving the current active filter to line level (in the amp). I will start with tweeter section. I believe you do that in your system? 
regards,Gera

Posted by rowuk on 12-12-2014
fiogf49gjkf0d
 noviygera wrote:
This is the amp I chose and just started building! Would it be possible to contact you directly, rowuk? I want to consult about the build and your experience, without flooding this post with specific questions. Or you can contact me.
thanks,Gera


The only problem is that I will not publish my email address on an open forum. Contact Alex Kitic the developer. He has my address. I couldn't find a PM possibility here at GSC.

The amplifier is not a solution, it is a tool that I have found a good UseCase for.

Posted by Romy the Cat on 12-13-2014
fiogf49gjkf0d
 noviygera wrote:
anyone familiar with the sound of this amplifier RH-307a?
The RH Amplifiers is a snippety subject. For what they meant to be they are reasonable design avoiding large cap in feedback and in particularly of you have no good OPT where grounding of secondary raises your parallel losses. The main question I would ask: why? If you use some kind 1000W plate dissipation tube then use of pentode is understood: you need gain and power. But if you at 106bB sensitivity then why would you need to worry about all those feedbacks and how to apply them if a regular triode would give all of it without all of this superfluous acrobatics? The answer might be that it has some sonic merit. It might be or not might be. I had a vey brief contact with the guy behind the RH Amplifiers at one of the forums and I got impression that he absolutely uninformed about larger picture. He sounded like a solid circuitry collaborator but absolutely clueless about Sound and what this “hobby” all about. There are plenty of them at all imaginable DIY audio websites, my attitude is that they are good because they solder instead of drinking or shooting cocaine but that is about all why I value them. Anyhow, it is hard to estimate how the RH Amplifiers might sound. They are fine technically, so any other amp in “Best Buy”. Am amp is a sonic expression and the only expression that I heard from that Alex Kitic was that he knew the Ohm Law and that he called me Ukrainian. Anybody who call me Ukrainian is an idiot in my book.

Posted by Jorge on 12-13-2014
fiogf49gjkf0d
Hello Gera,

First thing I  would do is get rid of the active Xover, in my experience they mud everything up. preamps are very important and this is like a double preamp with a lot of gain stages.  For bass you can maybe use it.  I use a separate Xover and Bass amp from 110 hz down.

About Low wattage SET, I have right now a 1.5 watt SET amp playing and am very used to 2 watt amps,  9, 12, 18  up to 700 SS watt amp.  This is running from 100hz up to 50 khz
Low wattage SET amps are very dynamic when compared to SS amps.  At first you crave this dynamics, but later you realize this is probably not very natural, an orchestra playing can handle a  lot of different moods, from piano, to pianissimo, to forte or forstissimo... and a world in between which creates the seduction of music.  I realized that these low level watt SET amps go directly from piano to fortissimo all the time,  with few gradations in between,  If you are playing Rock or Girl with guitar, you probably won´t notice, but with a full orchestra this becomes very noticeable.
As you go higher in wattage this effect gets reduced. With a 9 watt 300B things get a better balance.
My favorite amp at the moment (unluckily not mine) is an 18 watt SET amp.

I have not tried 1.5 watts on one driver directly as a Dedicated DSET amp as Romy has yet, I played around a bit only.  I would need identical amps because of phase shifts, tunning different amps as DSET is a HUGE job and time alignment becomes really very critical.  I would not recommend playing with it unless you have a lot of time on your hands for this.  Multi horn systems are already pretty complicated with just one amp.  Now the results form DSET from the times I started getting somewhere with it were incredible, really out of this world!

Good SS amps can sound great,  finding them is the problem, and bad ones are simply horrible... Bad tube amps are "musical".

Posted by Romy the Cat on 12-14-2014
fiogf49gjkf0d
In all those conversations about SET vs SS there is one important element that is allays escapes people attention. People like or do not like this or that type of amplifiers and they are willing to advocate their love. What however they always forgot that their love to a given amps or topology exist only in context of given conditions and circumstances and among those conditions the type of the speaker drivers the amps driving is one of the most critical element. 

For instance you can create as insane as you wish 3W SET and you can use some kind of super capacitors in there, using the octopus sperm as dialectic and making sure that the sperm was collected by a Buddhist monk in 4th generation who was lefty, bling on right eyes, vegetarian but eats peperoni.  Still, even that would not allow you super SET to be effective if it drives a compression diver with a ceramic magnet. Ironically the ultra-low output impedance and current drive is what doctor prescribed for ceramics. With the rare-earth magnets the situation is even more confusing but it truly confusing with estimable alnico.  There are a few version of alnico and they are slightly different but we never know what type alnico is used for a driver. It is not to mention that this specific alnico driver can be insolently charged or discharged to a given but unknown level…. 

So, loving of an Amplifier is like loving a bullet without even thinking what gun the bullet will be filing away and to which target… 

Rgs,
Romy the Cat

Posted by rowuk on 12-14-2014
fiogf49gjkf0d
I have experimented with SS/SET and now DSEP for ceramic magnet horn drivers. I will agree that ceramic drivers with SETs that I have heard being suboptimal, but after a year of DSEP, at least for my use case, I am quite content. I cannot explain it in a technical fashion, but in spite of standard Mundorf caps and standard resistors - but a remappable OPT, I am VERY happy with the sonic result. I have had no trouble with integrating the "sheen" of the violin (especially during the initial attack or the decay), piano or soprano. I never got this result with ANY SS amplifier.


I have no current desire to change my amps. I may not be at the cutting edge of what is possible, but my amplifiers are not drawing any attention to themselves. Further experiments with neodymium midrange and HF solutions may change this, but up until now, each small step is a breath of fresh air instead of a life saving situation. That makes me think that I am at least getting closer.......

 Romy the Cat wrote:
In all those conversations about SET vs SS there is one important element that is allays escapes people attention. People like or do not like this or that type of amplifiers and they are willing to advocate their love. What however they always forgot that their love to a given amps or topology exist only in context of given conditions and circumstances and among those conditions the type of the speaker drivers the amps driving is one of the most critical element. 

For instance you can create as insane as you wish 3W SET and you can use some kind of super capacitors in there, using the octopus sperm as dialectic and making sure that the sperm was collected by a Buddhist monk in 4th generation who was lefty, bling on right eyes, vegetarian but eats peperoni.  Still, even that would not allow you super SET to be effective if it drives a compression diver with a ceramic magnet. Ironically the ultra-low output impedance and current drive is what doctor prescribed for ceramics. With the rare-earth magnets the situation is even more confusing but it truly confusing with estimable alnico.  There are a few version of alnico and they are slightly different but we never know what type alnico is used for a driver. It is not to mention that this specific alnico driver can be insolently charged or discharged to a given but unknown level…. 

So, loving of an Amplifier is like loving a bullet without even thinking what gun the bullet will be filing away and to which target… 

Rgs,
Romy the Cat

Posted by Romy the Cat on 12-15-2014
fiogf49gjkf0d
 rowuk wrote:
I have experimented with SS/SET and now DSEP for ceramic magnet horn drivers. I will agree that ceramic drivers with SETs that I have heard being suboptimal, but after a year of DSEP, at least for my use case, I am quite content. I cannot explain it in a technical fashion, but in spite of standard Mundorf caps and standard resistors - but a remappable OPT, I am VERY happy with the sonic result. I have had no trouble with integrating the "sheen" of the violin (especially during the initial attack or the decay), piano or soprano. I never got this result with ANY SS amplifier.

Your experience with DSET is kind of irrelevant to the conversation of SS vs. SET with ceramic. Put in this way it is relevant only because it in fact proves the SS/ceramic problem. You need to understand that in context of this question DSET is NOT SET. A definition of SET is a single ended amplifier, presumably in A1, having output triode and one single OPT that make amp properly loaded across the whole amp's  band pass.  The properly loading implies that LF, MF and HF produce the reference volume levels more or less consistent harmonic structure. It is easy to measure with some training it is easy to recognize by hearing. 

DSET is a different animal. DSET discards the wide bandwidth by initial functional specification and assumes that the amp will produce the desired and targeted harmonic texture ONLY in very narrow frequency range, as the result a LOT of compromises and modifications are made in SET in order to accommodate this very specific DSET objectives. For instance let presume that you optimize DSET's OPT. With a "normal" set you have the output stage that let say loaded with 18x transformer. You have fine overall sound out of your Alnico wide range driver. Now you decided to do multichannel with DSET with the same types of the drivers. Your do your DSET and use the same 18x transformer, well perhaps you go with faster core, lower DCR, restricted LF and lower capacitance transformer and you would feel that to get the same harmonic texture fro the same driver you would slightly load your output stage, let say now you chose to have 16X transformer as your amp become "faster" and you feel that you need some harmonics to deal with your more transient overtones. Then you got a good ceramic driver to replace your alnico, let presume that they are the same only magnets are different (which is never the case but this is a part of a different conversation). Now you will feel that your DSET with 18x transformer is not too expressive and that you are losing the "bobblenees" in sound.  Changing the transformer to let say X25-X28 would give you the formal speed at highest octave. You will still have compromised lover octave of your narrow band channel but you might play with upper band of your next lower frequency channel to deal with it. This is juts an illustration but this is VERY close to real life illustration. 

Let see what happened in the illustration in reality. Moving from X18 transformer to X28 you effectively drops the amp output impudence, increasing the driver damping, increasing the amount of current your amps can send to the driver and deceasing the reactance of the amplifier to the driver's coil feedback. Ironically you are making you DSET to behave like it was a …. SS amp. That was my whole point. I certainly do not say SET or DSET can't drive ceramics. They might but I would not use ceramics for any more of less MF or HF anyhow. Still, if ceramics are used then SS like driving characteristics of amplifier are very desirable in my view.

Posted by Paul S on 12-15-2014
fiogf49gjkf0d
I hope everyone appreciates the value of the foregoing post. IMO, anyone thinking of "going DSET" might read it through several times first. The post reminds me of Romy's earlier (recent) post wherein he wonders how in the hell he did it at all (which I think is a great question...). Meanwhile, I was thinking about field coil drivers, not to introduce a new topic but to re-consider the driver itself as "tune-able", at least with respect to the gap. I do not think this would actually make rote matching DSET "channels" easier; but it MIGHT be a useful tool at some point, once someone starts getting close.


Paul S

Posted by Romy the Cat on 12-15-2014
fiogf49gjkf0d
 Paul S wrote:
Meanwhile, I was thinking about field coil drivers, not to introduce a new topic but to re-consider the driver itself as "tune-able", at least with respect to the gap. 

Not willing to much to hijack the thread but I need to note a notion that field-coil is tunable configuration is very faulty. I know that tenability of field-coil is hugely popular out there but only among the people who ether never had it, or have no brain, or have no ears. For a given magnetic system there is always the optimum saturation point and this is on one way or another sniggle point. if you use low voltage field-coil and saturate the code with current then you have sluggish and unsophisticated upper end. Can you correct it with overly idle DSET with super fast core? The answer is most certainly not. I am sure if one eats drinks a diarrhea and then eats a greatly made creme-brulee then he might claim a great meal. I do not think it would be however. The key is the individual elements shall be operating on own individual best and then those individual elements all need to be engaged in a larger performing picture where individual best would be multiplied and summed but not divided and subtracted.

I know there are many people out there that consider that field-coil are "adjustable". I even saw an idiot who advised volume control with field-coil on the drivers. Did I mention that the idiot was TAS reviewer? It was almost predictable, isn't it? The reality is that if a good field-coil driver is made then it is locked in own voltage, current, saturation and any deviation from it is a compromise of the driver. If there is a way to improve the driver by the adjusting anything then the best operation parameter for the driver has not been yet discovered. Otherwise it is as stupid as to experimenting with a DSET loading knowing that you use a discharged perm magnet driver….

Page 1 of 2 (31 items) 1 2 »