While I was installing the amps in this new room and with new speakers, it took me 2 days to optimize the sound properly, by only playing with how the amps are mounted (platforms and feet). Without that optimization, the amp wouldn't sound even close to what I expect from it. Maybe that's why installing filter degrades, what I already achieved by other means.
The amps are very close to main fuse box in a house. Fed AC through isolation transformer from 30A separate line, all the digital sources are run through conditioners with separate lines. It is a house with transformer on a pole outside, with no other houses in immediate area. So I could say that my CA line condition might be better than the average home setup. Also everythng is located directly on a concrete slab.
The amps are very close to main fuse box in a house. Fed AC through isolation transformer from 30A separate line, all the digital sources are run through conditioners with separate lines. It is a house with transformer on a pole outside, with no other houses in immediate area. So I could say that my CA line condition might be better than the average home setup. Also everythng is located directly on a concrete slab.
Re: Capacitor
You can see a cap in a pic. It's 1n8 Wima. I didn't go with anything more esoteric as this is an average quality cap everybody can use.
I also used Vishay resistor in a filter, so this shouldn't be degrading performance. Actually, I noticed that adding Vishay alone (1K) softens the trebles a bit (this is what Vishays tend to do) and it works better than 10K Riken by itself. But putting the cap in a circuit doesn't improve anything really. Highs are less sharp and not that well pronounced and some fuziness to the sound can be perceived. I can't see what that could improve. If you have too much highs in your amp, play with your setup and not try to fix it by adding more components. Also, my amps don't show any tendency to oscillate, so any improvement in that area is not needed either.
Ricren said:
You can see a cap in a pic. It's 1n8 Wima. I didn't go with anything more esoteric as this is an average quality cap everybody can use.
I also used Vishay resistor in a filter, so this shouldn't be degrading performance. Actually, I noticed that adding Vishay alone (1K) softens the trebles a bit (this is what Vishays tend to do) and it works better than 10K Riken by itself. But putting the cap in a circuit doesn't improve anything really. Highs are less sharp and not that well pronounced and some fuziness to the sound can be perceived. I can't see what that could improve. If you have too much highs in your amp, play with your setup and not try to fix it by adding more components. Also, my amps don't show any tendency to oscillate, so any improvement in that area is not needed either.
theory
I pretty much buy the theory that filtering the input of a somewhat limited hi-freq amp ( like the 3875) helps with the distortion.
By hi frec I mean radio frequencies that would otherwise beeing feed to the amp and mess things up.
We use RF filters in evey stage of our suites. Simply there's a lot of RF around.
cheers
Ric
I pretty much buy the theory that filtering the input of a somewhat limited hi-freq amp ( like the 3875) helps with the distortion.
By hi frec I mean radio frequencies that would otherwise beeing feed to the amp and mess things up.
We use RF filters in evey stage of our suites. Simply there's a lot of RF around.
cheers
Ric
I just had another short listen on the amps and definitely (to me) the unfiltered amps sound better by a large margin. The LPF actually makes the amp sound "tubey". The immediacy is gone and razor sharp and precisely defined (but never annoying) highs are gone. The midrange takes first plan. Everything is somwhat veiled, out of focus and somewhat "lighter" sounding.
The soundstage doesn't change.
The soundstage doesn't change.
PD
In general i agree with your assessment of the sound. You may find that decreasing the value of the cap may also decrease the tubiness. What really surprised me was a slight but worthwhile improvement in the bass. After a few days of casual listening to the filter i am back to no filters
It's probably worth mentioning that i don't listen to CD and my volume control is transformer based with very low output resistance but certainly limited bandwidth. I can well imagine that with a cheap cd player as source the filter will be appreciated.
Still, keep an open mind about the cathode follower. Everyone here seems to be crazy about it although this is one valve topology i really dislike (sorry Frank). Every time i've liberated a valve circuit from a follower the sound has improved in terms of immediacy. It's worth noting i haven't tried Allen Wright's followers which may be really good.
cheers
peter
In general i agree with your assessment of the sound. You may find that decreasing the value of the cap may also decrease the tubiness. What really surprised me was a slight but worthwhile improvement in the bass. After a few days of casual listening to the filter i am back to no filters
It's probably worth mentioning that i don't listen to CD and my volume control is transformer based with very low output resistance but certainly limited bandwidth. I can well imagine that with a cheap cd player as source the filter will be appreciated.Still, keep an open mind about the cathode follower. Everyone here seems to be crazy about it although this is one valve topology i really dislike (sorry Frank). Every time i've liberated a valve circuit from a follower the sound has improved in terms of immediacy. It's worth noting i haven't tried Allen Wright's followers which may be really good.
cheers
peter
It does seem that the filter takes care of some of the RF problems and that some lik ethe use of the filter for that purpose. But can the RF problem not be dealt with in a better way, like 4 separate steel enclosures (2 channels and 2 power supplies) and shielded mains, low power cables and interconnects?
I at least am planning on building my amp that way to ensure absence of RF problems as much as possible with a "mechanical" solution.
I at least am planning on building my amp that way to ensure absence of RF problems as much as possible with a "mechanical" solution.
analog_sa said:
Noop, Stainless steel, no magnetics and the wall thickness of the housing is larger than 5 mm so probably not too much resonances. That is on the "well designed" version I'm planning for later on. Probably are going to play with the chip way before that because I'm just tooooo curious
Peter,Peter Daniel said:
What you literally said, and that was directed to all of us who had noticed improvements applying low pass filter was that we are using “systems, that are build without proper attention to construction details and parts choice”, though just after that you said that “but in properly tuned amp it simply doesn't work and creates less than desired effect” telling us we actually have wrongly built basic GCs (so we maybe have acceptable rest of the system but have difficulties in building the GCs?).
Since I remember you did announced your “opinion” about adding any component to the minimized IGC before you tried anything of lpf/buffer stuff, and when you tried it you came across the conclusion that 100% matches to your “opinion”, I personally, just as PeteM’s, don’t expect that your findings about the buffer will be any interesting.
I’ve never heard such an opinion. But I did heard it improves the sound of the clone. By my own ears. Some other people have reported the same. And among the people that dealt with this none was so immature to clam something like that you “quoted”.
Should we start to count ourselves now? FYI, I tried many lpf configurations, including different roll-off points and differently made filters with the same roll-off point, as well as the 2nd order lpfs, often switching back. As far as remember Joe included in experiment another people, mentioned some with balanced system as a base to come across the correct conclusions… don’t push me to dig back…
__________________________
Of topic, but I’d like to know: as you did not use a pot, how did you control the volume during listening?
Pedja
PS: Looking at your not insulated foil cables… you really like dangerous games.
Pedja said:
In general, that's what I actually meant. It's either your GC, or your system, or your whole setup that needs better tuning, if you claim that LPF improves anything (by that I also mean all outside interferences, like mains and RF problems and system matching). I had my GC in few different places and not everybody was equally impressed.
Pedja said:
Personally, I don't see anything wrong if my initial opinion is confirmed by subsequent experiments and findings. If I would have to oblige to your way of reasoning, the only thing left to me, would be claiming that: "yes, LPF works surprisingly well", even if I didn't like it in reality. Don't you think it doesn't make any sense?
Isn't it also coincidental that one other member (who so far showed very good judgement in some other listening related matters) also, in general, confirmed my findings regarding the LPF. Actually, it made me feel pretty good, after expresing my initial dissatisfaction with the famed LPF circuit. If I'm not mistaken, I am actually the first person doing that. It almost feels like entering a lions cage
You might also say, OTOH, that my intuition is pretty good (if my findings match my initial "opinion"), since that's about the only thing I'm using as I don't employ simulations and my EE knowledge is practically not existant.
Pedja said:
Well, just today you heard at least two people, who expressed the opinion that they prefer the amp without LPF. I guess you will hear more people sharing that opinion as time goes by. If you really want to talk about immature claims, this is a pretty good one:
You have to find someone who built a pretty good IGC amp and then connected tube buffer with a LPF to its input to really claim that. So far I didn't hear about that sort of an experiment yet. And after today, I'm having even harder time to imagine that. Sorry guysJoe Rasmussen said:
Pedja said:
And what exactly is correct conclusion? Aren't you already predetermining the outcome, no matter what? I admit, I didn't try so many different configurations, but I also don't have the reason to use LPF. My amp's sound is pretty satisfying (I won't say fantastic) without filter and after today's one "sample" of the LPF, I already know that this is not something that can improve MY amp. I'd rather look somwhere else.
Pedja said:
I didn't control the volume. Since you are using similar DAC (at least that's what on your site) you know that passive output from TDA1543 is pretty low and with my speakers and the gain of the amp, it's just about right listening volume level on most CDs. I had previously 5 switched positions on my Greyhill and always used only the last two (depending on a disc). Without volume control I can't be accused of not precisely matching the settings by scope heads.
Pedja said:
If you played enough with GC, you should know that you can short the output for 15min (I didn't go more than that) and nothing really happens. I shorted the output many times, but it doesn't affect the amp. This is still a temporary setup and I'll be using separate amps for highs and mids. I have not much choice, but will be using LPF to cross the mids (probably)
Peter, in the 'mother of all flames' that seems to have occurred, I am still not sure if you tried the LPF with a tube buffer!
Although Joe R kindly supplied instructions for using an LPF without a buffer, there is no doubt that one will work better with a buffer.
You asked if anybody had built a decent sounding IGC and then added an LPF. Well that's the route I took, although I added a buffer with my LPF.
Without wishing to add to the animosity level of this thread, may I say that you are not exactly doing yourself any favours with remarks like
and
That first quote seems a bit at odds with the review of your amp on the 6moons site. Are you saying that they grossly over-rated your amp?
We could assume that your intuition is pretty good but we could also assume that your initial opinions about anything influence your final judgement!
I think what has annoyed many people is your inferance that nobody else can get a Gainclone or system correctly set up. Believe me, I do know how to do that. In fact, I know enough to tell you that the design of your amp (as wonderful and well-made as it is) is compromised! I can't tell you why as that information is currently the subject of a patent (not mine either).
Now I'm not trained in EE either and for testing purposes rely on my ears. That's no bad thing but then I don't tell the likes of Joe R, and Pedja that they are making 'immature claims'.
The Gainclone, when well-built is an astonishing piece of hi-fi kit. Let's not degrade it by this sort of opionated argument. By all means, have opinions, but remember that's what they are and none of us should bring any ulterior motives into the debate.
Although Joe R kindly supplied instructions for using an LPF without a buffer, there is no doubt that one will work better with a buffer.
You asked if anybody had built a decent sounding IGC and then added an LPF. Well that's the route I took, although I added a buffer with my LPF.
Without wishing to add to the animosity level of this thread, may I say that you are not exactly doing yourself any favours with remarks like
and
That first quote seems a bit at odds with the review of your amp on the 6moons site. Are you saying that they grossly over-rated your amp?
We could assume that your intuition is pretty good but we could also assume that your initial opinions about anything influence your final judgement!
I think what has annoyed many people is your inferance that nobody else can get a Gainclone or system correctly set up. Believe me, I do know how to do that. In fact, I know enough to tell you that the design of your amp (as wonderful and well-made as it is) is compromised! I can't tell you why as that information is currently the subject of a patent (not mine either).
Now I'm not trained in EE either and for testing purposes rely on my ears. That's no bad thing but then I don't tell the likes of Joe R, and Pedja that they are making 'immature claims'.
The Gainclone, when well-built is an astonishing piece of hi-fi kit. Let's not degrade it by this sort of opionated argument. By all means, have opinions, but remember that's what they are and none of us should bring any ulterior motives into the debate.
You've probably seen my site and will know that I freely share any information/experience that I have with pleasure.
However, this is somebody else's discovery and they have asked me not to divulge it at present so I must repect the promise that I made not to.
I was not trying to wind anybody up, just pointing out that Peter is not necassarily the only person who can set up a Gainclone properly.
And yes, if and when I am allowed to pass on this sort of information, it will be on Decibel Dungeon as soon as I can write it up!
From my position, standing outside the "LM3875 is god" camp, I find this debate quite fascinating.
From my listening exerience I can see why a lot of people like the LM chips, they do sound very nice, but they are lacking in the ultimate lightness and crispness that I get from my OPAs. I can also see how this would put off some listeners, because this makes the OPAs a little hard to listen to on less than perfect material. When I tried the LPF on my 548s, these qualities were lost, making it a much more friendly amp.
If the LPF has the same effect on the 3875, ( I can't tell, cos my LM amps are packed for a house move), then I can see why it would appeal to many. Please note, this is not a criticism of people's golden ears, or accuative facilities, merely a question of taste. My tastes may well be strange!
Finally, even though I disagree with Peter Daniels in just about every technical aspect of GC construction,(chip selection, vibration sensitivity, amount and type of capacitance, diodes, etc) it's only fair to point out that without his groundbreaking thread, there would probably only be about a dozen of us here on diyAudio who had built and played with this concept, and none of the interesting and informative developments that have come out of it would have occurred.
From my listening exerience I can see why a lot of people like the LM chips, they do sound very nice, but they are lacking in the ultimate lightness and crispness that I get from my OPAs. I can also see how this would put off some listeners, because this makes the OPAs a little hard to listen to on less than perfect material. When I tried the LPF on my 548s, these qualities were lost, making it a much more friendly amp.
If the LPF has the same effect on the 3875, ( I can't tell, cos my LM amps are packed for a house move), then I can see why it would appeal to many. Please note, this is not a criticism of people's golden ears, or accuative facilities, merely a question of taste. My tastes may well be strange!
Finally, even though I disagree with Peter Daniels in just about every technical aspect of GC construction,(chip selection, vibration sensitivity, amount and type of capacitance, diodes, etc) it's only fair to point out that without his groundbreaking thread, there would probably only be about a dozen of us here on diyAudio who had built and played with this concept, and none of the interesting and informative developments that have come out of it would have occurred.
Peter Daniel said:
Well, seems like you wasn’t sure till now, and now you decided to talk openly this way. These are not the light weight words at all. And you did not “just expressed your opinion and that's all”. You made very irritating statement.
Your initial opinion was a “strong opinion” making doubtful any further findings.
I should remember you that Peter from SA, using transformer volume control, in fact has lpf regardless of the RC network and also said that he doesn’t listen to CD.
Who was the second?
As I previously said, I’d consider immature all the claims like “this or that is the only way to build the proper GC”. Nothing similar here.
I have noted a few posts ago that you are trying to involve Joe in this thread. You are on the wrong track since the man generally avoids situations like this. And trying to do that, you are trying to interpret him in a completely wrong way. He doesn’t even confront Thorsten’s recommend circuit to that of his own. So no immature claims from him (I feel a bit strange
telling the obvious things). From Joe’s site: “Now this is not meant to be a critique of Thorsten’s circuit perse’. Rather it is a logical starting point as it was originally for me.”If you ask me, he has a rare combination of competent and relaxed approach to the audio. Try to see that. You can learn a lot.
As I said, pretty irritating statement (but again, you might be joking).
Conclusion is that the lpf works. You asked how many of us have bothered with it to come across the trustworthy conclusion. The answer is I don’t know how many, but certainly we did do that, and not only in a way to conclude yes or no, but also in a way to conclude how many and which way, and I am neither the only one, nor the first one who did that.
Mine too.
It is a pity for you.
Sure, LM3875 has all the protections you can think off. Maybe I’m just too conservative, but I feel better when conductors are not naked. They can touch anything...
Pedja
Nuuk said:
First of all, I don't know if review was overrated or not, but the amp requires at least a month of constant usage to sound best. My personal amps don't use coupling caps, the one supplied for review had 4.7 BG N. As good as they are, they are not perfect and degrade the sound a bit. Also in my personal amps, I,m using BG N 1000/50 caps the commercial amp has a standard version of the caps. I never claimed that the amp is perfect (it's some of you try to force that conclusion, however in a wrong way). The aluminum chassis tends to vibrate and proper setup (mostly choice of platforms) is very important to proper sound. The pot is a standard Nobel (although the older style, which supposed to sound better than current versions). I try not to use a pot in my home amp. So if you compare all the better choices I'm using in the amp I have for my personal use, the one which is a commercial version is seriously compromised. Although the review was good, the amp could be still bettered. Even then I'm not claiming that it sounds "fantastic" like the guy who's "gone" tries to tell us. I know it can be seriously improved. So please don't put words in my mouth and say it's "the best", it's simply not.
From what I understand from Joe's writing, the buffer is mostly used for a reason to isolate the filter from the pot as with a changing of a pot the filter's frequency changes as well. In a case when source can drive the filter circuit, the buffer shouldn't be necessary. And as stupid I might seem, I will never buy a fact that additional circuit there ( be it tube or SS) will always improve the sound. You have to understand that it will always depend on the rest of the associated gear.
Again, I never tried to say that "nobody else can get a Gainclone or system correctly set up". I just said, that if for some of you who prefer filter and buffer this might be the case. Please note the word "might" and "some". Not everybody who built GC is using filters and buffers (at least not yet). Why I think like that? Because I tried the filter, and it's not working for me. I can't imagine, I would be alone in my opinion.
And it is only my opinion, but I'm also very confused why Joe with some faithful group of followers tries to imply that the only way to have a "proper" sounding GC is using his tube buffered and filtered circuit. This is simply not true. There are many ways to a top of a mountain and his (or mine) is only one of the choices, but definitely not the only and the best ones. However his favourite statement is "don't forget the filter". It almost became the motto to the whole campain. At present time I have to say that it is misleading. For some people the GC WILL sound better without a LPF.
Try it and let us know
PS: I really don't want to go into argument on all that and it spoils the spirit of GCloning. I'm very impressed with the overall attitude on ChipAmp forum and helpful and friendly approach. Let's continue like that on this forum. And I also have respect for everybody elses opinion. And I'm always looking for ways to improve the amp, and you know I will share it here. So let's try to be reasonable.
Pedja said:
Well, seems like you wasn’t sure till now, and now you decided to talk openly this way. These are not the light weight words at all. And you did not “just expressed your opinion and that's all”. You made very irritating statement.
Your initial opinion was a “strong opinion” making doubtful any further findings.
I should remember you that Peter from SA, using transformer volume control, in fact has lpf regardless of the RC network and also said that he doesn’t listen to CD.
Who was the second?
As I previously said, I’d consider immature all the claims like “this or that is the only way to build the proper GC”. Nothing similar here.
I have noted a few posts ago that you are trying to involve Joe in this thread. You are on the wrong track since the man generally avoids situations like this. And trying to do that, you are trying to interpret him in a completely wrong way. He doesn’t even confront Thorsten’s recommend circuit to that of his own. So no immature claims from him (I feel a bit strange
If you ask me, he has a rare combination of competent and relaxed approach to the audio. Try to see that. You can learn a lot.
As I said, pretty irritating statement (but again, you might be joking).
Conclusion is that the lpf works. You asked how many of us have bothered with it to come across the trustworthy conclusion. The answer is I don’t know how many, but certainly we did do that, and not only in a way to conclude yes or no, but also in a way to conclude how many and which way, and I am neither the only one, nor the first one who did that.
Mine too.
It is a pity for you.
Sure, LM3875 has all the protections you can think off. Maybe I’m just too conservative, but I feel better when conductors are not naked. They can touch anything...
Pedja
Pedja,
Somebody already noted previously that you sound like talented politician, and your post quite frankly reflects this. You seem very much prefer to talk about nuances of my statements and claims. It almost looks like you are a full time crew in the campain for the "spread the LPF news", started by Joe and quite succesfuly carried out so far. If you claim that I made very irritated statement, accept the news that this is even more irritating statement:
Joe Rasmussen said:
and frankly not true. While you are claiming that I try to make statements like “this or that is the only way to build the proper GC" I can't agree with that. I never tried to force any of my opinion and choices. I simply suggest them as useful advice. Yet somebody who makes statements like the above, is indeed saying “this or that is the only way to build the proper GC". Sorry Joe, I didn't want to involve you in all that crap and immature arguments, but Pedja just drags me more and more into that.
What I'd like to say at the end is that LPF works indeed. A simple capacitor shunting to ground and cutting off frequencies above 20k is very influencial on the sound (suprisingly influencial). I'm not really saying if it's good or bad, because this will vary with personal taste and associated equipment. What I'm saying is try it for yourself and see if you like it or not. I just did that and you know my opinion.
But please don't try to force the other opinion, that it's the ultimate cure for all LM3875 difficiencies, it's simply not.
And BTW, to make it very clear, I think that so far I attained no more than 65% of performance capacity from my amp. So, it's not the best out there and don't put words in my mouth.
__________________________
PS: You asked me who was the second person. Initially I was thinking about myself, but it seems like pinkmouse could be an alternative
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.