Precisely, Nico. This is the sort of thing I hear in nearly all other audio systems that I have a listen to, at a certain volume they turn into conventional hifi systems, they no longer sound "real". People often put this down to speaker distortion, but in my experience this is rubbish; I have never yet had to "fix" the sound of a system by changing to better speakers in any fashion. Yes, weaknesses inside a speaker need to be addressed, but the real killer technique is to make sure the electronics are working cleanly ...
Frank
simple fix to all these problems is to know the limits of operation of your amp.....just like DF96 said, treat your 100watt amp as if it were a 60 watt amp...better if you treat it as a 25 watter....
this is also the reason i liked the Leach Super amp, at 250 watts per channel, running them at say 10 watts, none of the psu rubbish can show its ugly head.....
this is also the reason i liked the Leach Super amp, at 250 watts per channel, running them at say 10 watts, none of the psu rubbish can show its ugly head.....
But what's that saying about the design methodology of these beasts? Would you buy a car that's advertised as being capable of going 150mph just so you're comfortable at 45mph? And if you dared to try to go anywhere over 100mph it would be as dangerous as hell? Sorry, that's just nonsense, a failure of the industry - talk about poor value for money ...
.I have a unit rated at 20 watts, on a good day
, and I run it close to clipping a lot of the time, with very good results. And that's how it should be ...Frank
But what's that saying about the design methodology of these beasts? Would you buy a car that's advertised as being capable of going 150mph just so you're comfortable at 45mph? And if you dared to try to go anywhere over 100mph it would be as dangerous as hell? Sorry, that's just nonsense, a failure of the industry - talk about poor value for money ...
I have a unit rated at 20 watts, on a good day
Frank
yours must be classA amp.....
it's about preference.....who is to tell me i can not have them?
Diy'ers can be crazy at times because they can choose to.....
i make high powered amps because i can....
Actually, cars are often designed and advertised like that. Look at truck commercials. They often show them being loaded with (for example) a bed full of bricks that easily exceed the truck's weight limit. Then you see a small disclaimer, "Professional driver, do not attempt...".
I know of more than one amp that is designed that way.
No, just chip amps, TDAs. But, they don't have to worry below about 180Hz, a separately powered woofer does that. And, they're part of a HT receiver, so the transformer, etc, has to be sized to handle all channels going full bore; so in stereo mode the unit's power supply is not as stressed as some would be ...yours must be classA amp.....
it's about preference.....who is to tell me i can not have them?
Diy'ers can be crazy at times because they can choose to.....
i make high powered amps because i can....
The other part is that a lot of tweaking has been done to make the supply as clean as possible, no glitching gets through to the voltage rails. If this work wasn't done the sound would just be slightly better mid-fi ...
Frank
Do you have a clue what forward voltage drop is, Daniel?
Yeah "let's not lose sight of the target"!!!
Last edited:
Treble dims when you crank it up? It could happen when the output devices have faltered a bit and next, the drivers become burdened, so what you get is super loud and without some of the treble.
It is possible to fix that by installing a compressor, actually a soft clip circuit. In that link, it is seen installed between in+ and in- of the LTP. The feedback shunt resistor (not the feedback resistor but rather its groundside partner) has been replaced with a voltage divider for fine tuning. It is possible to bring the amp up short before the problem starts; however, it is also possible to set it to engage when the problem starts, so you get neither more nor less output power.
That 40 watt amp has 4,400u per rail at the power supply and 100u amp decoupling caps. Before installing the soft clip circuit, the treble would go dim as you crank up the volume. After installing the soft clip circuit, the frequency response remains level.
I had no idea that the problem could have been power circuit, so, oops, I probably fixed it at the wrong spot.
I am talking about a relatively subtle effect, that probably not many people are aware of. Essentially, detail in the treble is lost, and for many listeners this is what they expect hifi, audio, to always sound like, which is why it's also typically easy to pick when a sound, whose source is unknown, is in fact coming from an audio system. Quality sound at decent volumes is very intense, it completely fills the space in which you're listening, but it doesn't sound loud or aggressive, quite often you are completely unaware of how high the volume is until you try and talk to somebody.
Frank
If you run it close to clipping a lot of the time then, unless you listen exclusively to modern pop music with no dynamic range, you will also be running it into clipping for much of the time. That means you are listening to the clipping performance of your amp. You need a bigger amp.fas42 said:
I don't know off-hand what the peak/mean ratio is for typical real recorded music but I would hazard a guess that it is somewhere in the range 20-30dB?
Last edited:
I mean close to clipping as in the real thing, never actually clipping: the nature of the system is such that the volume is bounded by the integrated design, to limit the gain of the signal being fed to the power amps. Maximum volume can't overdrive the circuit, I have several CDs where full volume only gives the subjective level of a small TV set. Yet, on some compressed pop, 3/4 volume is absolutely deafening, my ears are ringing after 5 minutes.
To get a handle on it, a setting of 36 out of a maximum 40 reproduces a piano at the right SPL: it sounds correct at the other end of the house, and outside ...
Frank
You're finding it hard to accept that the system can be used up to a point of clipping, in the correct technical sense, without any apparent audible signs. There is a peak SPL that a system can theoretically cleanly deliver, easily determined from the RMS power of the amplifier and the sensitivity of the speaker, which for most setups is somewhere in the region of 105dB, more or less, taking into account that you have 2 speakers and are listening at a couple of metres distance, with room reflections adding to the mix. This is what I estimate I'm getting from my system, and if you look up the literature this is sufficient to account for what most real, musical events present to the ear.
So there is no reason for a system not to be able to present unclipped, realistic sound. Except that a large percentage don't, in large part because of flaws in their power supply setups, which is what this thread is all about.
An obvious sign that the sound is not clipping is that I can lower the volume from maximum all the way down to near whisper quiet, and the character of the sound doesn't alter, it just gets softer, less loud, that's how it's supposed to work ...
Yes, I know most systems don't behave like this, but that's the point that I make quite often: most systems are flawed in their operation ...
Frank
So there is no reason for a system not to be able to present unclipped, realistic sound. Except that a large percentage don't, in large part because of flaws in their power supply setups, which is what this thread is all about.
An obvious sign that the sound is not clipping is that I can lower the volume from maximum all the way down to near whisper quiet, and the character of the sound doesn't alter, it just gets softer, less loud, that's how it's supposed to work ...
Yes, I know most systems don't behave like this, but that's the point that I make quite often: most systems are flawed in their operation ...
Frank
and I find the opposite effect when turning up the volume with a Spike protected chipamp.
I am well away from clipping but the signal arriving at the speaker is altered as the volume is increased. I attribute this "characteristic" to Spike interfering with valid audio signals into a valid audio load.
Your example, being the opposite, is allowing valid audio signals to pass to a valid audio load. Nothing in the system, whether it be protection or lack of current capability in the PSU should interfere with the clean passage of audio through to the listener.
That hopefully is what this thread is about.
Current capability of the PSU such that audio passes through without interference.
I am well away from clipping but the signal arriving at the speaker is altered as the volume is increased. I attribute this "characteristic" to Spike interfering with valid audio signals into a valid audio load.
Your example, being the opposite, is allowing valid audio signals to pass to a valid audio load. Nothing in the system, whether it be protection or lack of current capability in the PSU should interfere with the clean passage of audio through to the listener.
That hopefully is what this thread is about.
Current capability of the PSU such that audio passes through without interference.
Assuming you are listening to real music and not test tones, yes. Any audio system being run consistently just below the point of clipping will either frequently clip, or must have some method to prevent clipping. Music dynamics ensure this. Hence my insistence that you are either listing to clipping or whatever method you use to avoid clipping.fas42 said:
The only alternative, which most people actually use whether they realise it or not, is to run the normal level well below clipping so clipping on real music is rare. You claim not to do this, by explicitly saying that you normally run close to clipping. You can't have it both ways. Music peaks are not 3-6dB above normal level, but 20-30-40dB above normal level.
in my view this is wrong.
If the amplifier is handling signal below clipping then it is not reasonable to suggest that "must have some method to prevent clipping"
The key words are 'consistently just below the point of clipping'. Music dynamics are such that this is not possible. You can either run significantly below the point of clipping and accept an occasional clip, run just below and have frequent clipping or a mechanism to avoid this (soft clip or fast compression), or listen to 'music' with no dynamic range like many modern pop CDs. In the latter case you are not in a position to judge the sound of the equipment as the signal has already been well mangled by the recording engineer.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.