well I'm amazed that there would be amplifiers designed in such a way.
I've been involved in Hi Fidelity sound for over 40 years, and have not read, seen or heard of such a hi-Fi amplifier, and I was a regular reader of the Hi-Fi and audio magazines.
I think you must be referring to something other than Hi-Fi/Audiophile, or just good, amplifiers (where this thread belongs), and is an inappropriate comment here.
I was under the impression that Diy Audio members were interested in building good quality components, not such inferior amplifiers as you say exist.
I know you are very technically competant, but I think you may have not been thinking clearly on this one.
billabong.
I've been involved in Hi Fidelity sound for over 40 years, and have not read, seen or heard of such a hi-Fi amplifier, and I was a regular reader of the Hi-Fi and audio magazines.
I think you must be referring to something other than Hi-Fi/Audiophile, or just good, amplifiers (where this thread belongs), and is an inappropriate comment here.
I was under the impression that Diy Audio members were interested in building good quality components, not such inferior amplifiers as you say exist.
I know you are very technically competant, but I think you may have not been thinking clearly on this one.
billabong.
I am thinking very clearly on this one.
don't **** the amp because your speakers are not designed to match the room.
don't **** the amp because your speakers are not designed to match the room.
What are you sniffing/inhling/smoking, as you are not making sense.
If you are not sniffing/inhaling/smoking, then explain how your comments are relevant to this thread.
If you are not sniffing/inhaling/smoking, then explain how your comments are relevant to this thread.
posts 1195 & 1196 seem to be suggesting that altering the amplifier away from good, to correct errors elsewhere in the system is an OK philosophy.
I am pointing out that there is another way to improve the sound quality. Make each component good and find the weakest link and improve that.
In this case, fix the Q/room interaction and don't reduce the Q effect of the amplifier to correct for a faulty speaker.
I am pointing out that there is another way to improve the sound quality. Make each component good and find the weakest link and improve that.
In this case, fix the Q/room interaction and don't reduce the Q effect of the amplifier to correct for a faulty speaker.
Come on now- fair go!
Post 1195 only expanded the topic of input capacitor selection (in which you also took part), but mentioned additionally that the capacitor could be sized to roll off the low frequencies even as high as 30 Hz, to suit the listening environment and speakers.
Other than that low frequency roll off, the amplifier remained the same.
You, Andrew, in post # 1157 said you like to select an input capacitor to roll off -3db at 2HZ, as you preferred it to a capacitor sized to roll of at 8 Hz, using your small speakers, which roll off at 60Hz.
The only difference between you and the rest of us is that someone mentioned a slightly higher frequency.
Anyway, since an input cap has to be selected anyway, why not select it to roll off at 30 Hz, if it sounds better in an un-ideal room with the only small speakers that the poor chap could afford at the time.
Easy enough to replace the input cap when the listening environment and speakers improve.
Regards,
billabong.
Post 1195 only expanded the topic of input capacitor selection (in which you also took part), but mentioned additionally that the capacitor could be sized to roll off the low frequencies even as high as 30 Hz, to suit the listening environment and speakers.
Other than that low frequency roll off, the amplifier remained the same.
You, Andrew, in post # 1157 said you like to select an input capacitor to roll off -3db at 2HZ, as you preferred it to a capacitor sized to roll of at 8 Hz, using your small speakers, which roll off at 60Hz.
The only difference between you and the rest of us is that someone mentioned a slightly higher frequency.
Anyway, since an input cap has to be selected anyway, why not select it to roll off at 30 Hz, if it sounds better in an un-ideal room with the only small speakers that the poor chap could afford at the time.
Easy enough to replace the input cap when the listening environment and speakers improve.
Regards,
billabong.
Andrew,
I am still not quite sure if I understood your point. You may not have explained it adequately.
You keep mentioning small speakers. Is it because you often see small speakers in the photos that Carlos attaches, and think that he designs his amplifiers around them?
I have wondered about that myself, but Carlos has said he uses headphones for critical listening. Carlos is after all an intelligent man.
billabong.
I am still not quite sure if I understood your point. You may not have explained it adequately.
You keep mentioning small speakers. Is it because you often see small speakers in the photos that Carlos attaches, and think that he designs his amplifiers around them?
I have wondered about that myself, but Carlos has said he uses headphones for critical listening. Carlos is after all an intelligent man.
billabong.
I believe Andrew's point is that we should not "mess up" a well designed amplifier with flat response to accommodate speakers/rooms that are not flat.
Francois
Francois
Francois G said:
Hello Bilabong
That is what I got too... not only that, but using speakers (even good ones) that are not suitable for the listening room will not help either.
Therefore a better option (rather then changing the amp) would be finding speakers that (1) are appropriate for the listering environment and (2) match that well designed amplifier.
I have to agree with that... I do it professionally on a daily basis for really critical listening environments.
(Aftertought acoustics cannot be fixed in the mix... it cannot be fixed by changing a cap or two either.
I am very happy with my DX amps (both Classic and HR-II) and have not tried too many changes as I am not an amp designer.
I tried them with several speakers that match my test room and some combinations are "better" than others, but all sound well.
(Yes, I have them connected using a rather "massive" PSU. Next there will be some testing using a SMPS I am working on.)
Cheers,
Francois G said:
That's my understanding too. And it's entirely relevant. I have never seen or heard of an amp with a modified response, but I don't doubt they exist in consumer (non-hi-fi) world. Especially, I suspect, those all-in-one Home Theatre kits they make.
However, deciding where to cut off the bandwidth (within reason) isn't what I'd consider altering the response away from flat, and that is the subject of this capacitor talk. But 30 Hz is too high for my comfort zone, and probably does hit a threshold of reducing the bass response.
And Carlos' point that flat amplifiers sound different is also perfectly valid, and not in disagreement with Andrew's point.
The only smoke going on here seems to be the puffs coming from your ears, Billabong. Count to ten. Cool off. Everyone is agreeing with each other.
..Todd
I agree.
Design an amp for a speaker is a bad idea.
Amp sholud be "universal", with flat (20Hz...20KHz or 10Hz...40kHz -1dB) response.
Any "messing up" is a bad idea then.
That's why I don't prefer any treble or bass boost in the power amplifier. That's why I never choose 30Hz/-3dB rolloff in the input in my amp designs...
Andrew, that's a good point. Very good!
I agree 100% - as always 😀
Design an amp for a speaker is a bad idea.
Amp sholud be "universal", with flat (20Hz...20KHz or 10Hz...40kHz -1dB) response.
Any "messing up" is a bad idea then.
That's why I don't prefer any treble or bass boost in the power amplifier. That's why I never choose 30Hz/-3dB rolloff in the input in my amp designs...
Andrew, that's a good point. Very good!
I agree 100% - as always 😀
taj, maybe the smoke coming from my ears is due to the flames coming from your mouth. 
So you are all in agreement that Andrew's point was only that amplifier response should not be altered from flat for any reason.
The discussion that Andrew commented on was about selecting a low frequency -3db roll-off, and the required capacitor value to achieve this.
Any input cap. will (horror) alter the amplifier low frequency response, and therefore alter the amplifier response from flat. It is just a matter of selecting the frequency at which to do the altering.
Only omitting the input cap. will prevent it from altering the response. Any responce deviation would then be attributed to the source components.
If you still wish to use an input cap., the question then is at what low frequency.
AndrewT selected -3db at 2 Hz, finding it sounded better with his small speakers, than when using an 8 Hz roll-off. ( did he design his amplifier to suit his speakers?)
taj was still undecided on which frequency he would select, suggesting he may go with -3db at 5 Hz.
Using Andrews selection of 2Hz, it would cause the least deviation, and since it meets his approval when used with his small speakers, maybe it will sound good with yours as well, whatever their size.
billabong.
So you are all in agreement that Andrew's point was only that amplifier response should not be altered from flat for any reason.
The discussion that Andrew commented on was about selecting a low frequency -3db roll-off, and the required capacitor value to achieve this.
Any input cap. will (horror) alter the amplifier low frequency response, and therefore alter the amplifier response from flat. It is just a matter of selecting the frequency at which to do the altering.
Only omitting the input cap. will prevent it from altering the response. Any responce deviation would then be attributed to the source components.
If you still wish to use an input cap., the question then is at what low frequency.
AndrewT selected -3db at 2 Hz, finding it sounded better with his small speakers, than when using an 8 Hz roll-off. ( did he design his amplifier to suit his speakers?)
taj was still undecided on which frequency he would select, suggesting he may go with -3db at 5 Hz.
Using Andrews selection of 2Hz, it would cause the least deviation, and since it meets his approval when used with his small speakers, maybe it will sound good with yours as well, whatever their size.
billabong.
Hi Billabong,
The point you seem to be overlooking is:- I disagree with deliberately raising the amplifier F-3dB into the audio band to hide a failing in the speaker/room interaction.
It matters not whether we choose F-3dB at some chosen sub-audio frequency (between 0.1Hz & 4Hz) on that earlier premise.
I suggest you also think about two cascaded passive filters both with a Q of 0.7
The resulting frequency response and Q will depend greatly on how close the rolloff of the two filter frequencies are.
Suddenly, we find that pushing one filter up even as little as one octave will have a measurable and audible effect up to around 20times F-3dB of the individual filters. i.e combine two filters of F-3db=20Hz with F-3dB=9Hz and move the lower frequency up one octave to 18Hz. The almost Bessel filter characteristic now shows significant effects upto around 300Hz. That will have a profound effect on what the final audio signal sounds like. But the filter we have just changed is from 9Hz to 18Hz, some would argue that such a change should be inaudible.
The point you seem to be overlooking is:- I disagree with deliberately raising the amplifier F-3dB into the audio band to hide a failing in the speaker/room interaction.
It matters not whether we choose F-3dB at some chosen sub-audio frequency (between 0.1Hz & 4Hz) on that earlier premise.
I suggest you also think about two cascaded passive filters both with a Q of 0.7
The resulting frequency response and Q will depend greatly on how close the rolloff of the two filter frequencies are.
Suddenly, we find that pushing one filter up even as little as one octave will have a measurable and audible effect up to around 20times F-3dB of the individual filters. i.e combine two filters of F-3db=20Hz with F-3dB=9Hz and move the lower frequency up one octave to 18Hz. The almost Bessel filter characteristic now shows significant effects upto around 300Hz. That will have a profound effect on what the final audio signal sounds like. But the filter we have just changed is from 9Hz to 18Hz, some would argue that such a change should be inaudible.
Oh!...you are still discussing this Billa?
Do the way you want and be happy.
The best solution will be the best to your years.
Do not believe and do not trust anybody else than yourself, listening...this is "reality" for you.
Test everything, compare all you can, conclude by yourself...this will represent increasings in know how, the rest are numbers, ideas, fashions and those things.
Always ask things to the ones you think are more experienced, but go to check what they say.... they are humans and sometimes not correct into their ideas... and even correct, you may dislike the sonic effect...those things are subjective, and depends from the "system"...no only the amplifier, it is just one piece, of course if the amplifiers is the worst one into the chain, the whole LOW quality will be the amplifier responsability.
But as you may have concluded, amplifiers are never the worst part into the whole story.
Into Precision input, technically, 1uf results flat.... bigger or lower capacitor must be your decision, listening and comparing...keep one channel standard and tweak the other.
Carlos
Do the way you want and be happy.
The best solution will be the best to your years.
Do not believe and do not trust anybody else than yourself, listening...this is "reality" for you.
Test everything, compare all you can, conclude by yourself...this will represent increasings in know how, the rest are numbers, ideas, fashions and those things.
Always ask things to the ones you think are more experienced, but go to check what they say.... they are humans and sometimes not correct into their ideas... and even correct, you may dislike the sonic effect...those things are subjective, and depends from the "system"...no only the amplifier, it is just one piece, of course if the amplifiers is the worst one into the chain, the whole LOW quality will be the amplifier responsability.
But as you may have concluded, amplifiers are never the worst part into the whole story.
Into Precision input, technically, 1uf results flat.... bigger or lower capacitor must be your decision, listening and comparing...keep one channel standard and tweak the other.
Carlos
Attachments
HELP ME UNCLE CHARLIE!
one channel of my Vanilla is go to hell. because an accident.
the output (to speaker) has shorted with -18V DC supply for Pre-amp.
I hear BIG "DUM" after that the speaker is "hmmmmmmmm" like noise but this one is veri big, even the input i dont connected to source.
I have checked all the Vbe transistor looks normal (540-580mV).
I have try to change the differential (2n5401) but still have a big noise.
HELP ME
La Ode
AndrewT,
Thanks for that. I wish you had posted something similar in the first place - it would have saved some unecessary posts.
Your posts, I've noticed, are often very brief and cryptic-like, and for the less electronically knowledgable than yourself, such as myself, sometimes I find it difficult to understand your meaning.
I was not aware that frequencies 20 times higher would be effected.
Makes me wonder how noticeable to the average person a 20 or 30Hz roll-off would be, compared to a 4 or 5 Hz roll-off. Phono preamps would have a roll-off around 20 Hz I think.
Regards,
billabong.
Thanks for that. I wish you had posted something similar in the first place - it would have saved some unecessary posts.
Your posts, I've noticed, are often very brief and cryptic-like, and for the less electronically knowledgable than yourself, such as myself, sometimes I find it difficult to understand your meaning.
I was not aware that frequencies 20 times higher would be effected.
Makes me wonder how noticeable to the average person a 20 or 30Hz roll-off would be, compared to a 4 or 5 Hz roll-off. Phono preamps would have a roll-off around 20 Hz I think.
Regards,
billabong.
Carlos,
Yes, unfortunately still at it. You would think I could find something more useful to pass my time with.
I had already decided on my input cap selection.
I'll use a Black Gate N non polar 10uf input cap on the Precision1, which will give me a -3db roll-off at 2.75 Hz ( Preamp has an output cap of 3.8uf)
I may later consider removing the Precision input cap. The 3.8uf cap will then roll-off at 2.79 Hz.
Regards,
billabong.
Yes, unfortunately still at it. You would think I could find something more useful to pass my time with.
I had already decided on my input cap selection.
I'll use a Black Gate N non polar 10uf input cap on the Precision1, which will give me a -3db roll-off at 2.75 Hz ( Preamp has an output cap of 3.8uf)
I may later consider removing the Precision input cap. The 3.8uf cap will then roll-off at 2.79 Hz.
Regards,
billabong.
I usually don't need to spell out every detail. Probably because I'm here learning as well and can't explain it all.billabong said:
Most folk will go and research it themselves, others will simply argue.
A few will admit they don't know and ask.
Where are you?
billabong said:
No flames here.
I doubt there would be much audible difference between roll-offs of 20 or 30 Hz under normal circumstances, and my concern is purely one of specifications. With an F3 of 20 Hz at Q=7.07, the response starts to dip higher than 20Hz and I find that conceptually objectionable (despite it probably not making any audible difference). To maintain truly flat response down to 20 Hz, you would need an F3 of 10Hz or slightly less assuming 1st order filtering and Q=7.07. (Hence my arbitrary choice of 5 Hz).
Your scenario of series capacitance is different than my single input capacitance, and would potentially have much more impact on the audio band (considering both source device and amplifier) . However, I'm not aware of what my source device output capacitors are, if they use any, so I now tend to want an even lower F3 point to avoid any high Q response bumps if there happens to be a significant output cap on the source device.
I suppose a practical test would be to measure the response of both the source device and the amp together to ensure the response is still flat down to chosen F3 point with chosen capacitance. And alter F3 point from there (by listening.)
And to clarify (again) I expect any quality amplifier response to be flat as a prerequisite. Let the circuit variations and possibly the component types determine the 'quality' of the sound, not the F.R.
..Todd
This is all I'm babbling about. Yellow trace is F3 at 20 Hz. Pink trace is F3 at 6Hz (still flat at 20 Hz). My guess of 5 Hz was pretty close.
And Carlos may be right; it's better to preserve the power by rolling off the response higher rather than lower. However I think the P1 provides enough output to make that a moot point for normal listening levels.
..Todd
And Carlos may be right; it's better to preserve the power by rolling off the response higher rather than lower. However I think the P1 provides enough output to make that a moot point for normal listening levels.
..Todd
Attachments
Taj,
what is the filter that you have posted the graphs for?
It appears to be more than a single pole passive filter.
It also appears to have a Q much high than Butterworth.
What have you posted?
eg. a Butterworth (Q=0.7) is approximately -1dB at twice F-3dB
and approximately -0.1dB at ~7times F-3dB.
what is the filter that you have posted the graphs for?
It appears to be more than a single pole passive filter.
It also appears to have a Q much high than Butterworth.
What have you posted?
eg. a Butterworth (Q=0.7) is approximately -1dB at twice F-3dB
and approximately -0.1dB at ~7times F-3dB.
- Status
- Not open for further replies.