How much power ?... Peaks, man. I've seen 100 v p-p transients on my big amp. I don't care if it can do 250w all day , I just want that 500w "burst" when I need it. Lots of caps (100kuf) and high rails (76V). I have no idea what clipping is , don't want to , either. 😀
OS
OS
I used to have a pair of Altec 9440's in bridge mono delivering 800 watts per channel to a pair of Klipsch Heresy speakers. How much power do you really need? Well lets just say that everything sounds better having an excess amount of power coupled to it. In the case of the Heresy speakers the bass became extremely fast and accurate with no effort. Did I need this amount of power? No, I didn't. I have found in my many years of home addiction that there is no such thing as a maximum amount of power. I have done far more damage to speakers by running them with amplifiers not having enough brute power than I have by having amplifiers exceeding the manufactures recommended power ratings.
In the case of my Altec model 9 speakers purchased in 1975 I had damaged them a number of times with a 60 watt Pioneer SA8500 amplifier. I believe the efficiency was 93db 1w 1meter. Having had them repaired before warranty ran out I finally purchased an amplifier in excess of 250 watts per channel. They sounded better as a result and I never damaged them after that.
About 10 years ago roughly I became a tube head. My tastes changed and I don't play the music like I used to either. Tube amps sound differenty than SS amps do. I currently run 2 modified Dynaco Stereo 70's each driving a pair of custom Klipsch Cornwall clone speakers. These seem to fair well when compared to a pair of good 100 watt SS amplifiers driving the same load. When I want to push them harder I hook up a pair on Quicksilver mono block 60 watt per channel amps and these seem equal to a pair of 200 watts SS amps.
There is no answer to this thread in my opinion its basically a how much power do you have to your what brand of speakers you have. There is no cut and dried answer because it depends on room size, speaker efficiency, room treatment, and personal taste not to mention the scope of music listened to. What are you happy with? Thats your answer then.
In the case of my Altec model 9 speakers purchased in 1975 I had damaged them a number of times with a 60 watt Pioneer SA8500 amplifier. I believe the efficiency was 93db 1w 1meter. Having had them repaired before warranty ran out I finally purchased an amplifier in excess of 250 watts per channel. They sounded better as a result and I never damaged them after that.
About 10 years ago roughly I became a tube head. My tastes changed and I don't play the music like I used to either. Tube amps sound differenty than SS amps do. I currently run 2 modified Dynaco Stereo 70's each driving a pair of custom Klipsch Cornwall clone speakers. These seem to fair well when compared to a pair of good 100 watt SS amplifiers driving the same load. When I want to push them harder I hook up a pair on Quicksilver mono block 60 watt per channel amps and these seem equal to a pair of 200 watts SS amps.
There is no answer to this thread in my opinion its basically a how much power do you have to your what brand of speakers you have. There is no cut and dried answer because it depends on room size, speaker efficiency, room treatment, and personal taste not to mention the scope of music listened to. What are you happy with? Thats your answer then.
I'll take a guess that the OP is actually suggesting a little more attention from DIY folks to smaller amps. This may be for design ideas, technical requirements or just an observation that 99% of our listening time, sub-10W levels are more than adequate.
With reasonably sensitive domestic speakers and limited range digital sources,that seems like a fairly reasonable statement. It's also something that has been a nagging concern in my mind for many years with respect to efficiency and design appropriateness.
Others who perhaps cruise around with megawatt car systems at suicidally high SPLs probably feel it's too quiet back home with only a couple of kilowatts and a wall of speakers in the bedroom. I can't live with that myself and set my goal at where I can discern all I want to hear at min. SPL and calculate, test and mostly just listen to what is sufficient to deliver it with some 50% power reserve.
Gross overkill may annihilate any concerns of overload and give great bragging rights but it also annihilates your cash and the chance try out some other DIY things. Here is the chance to also build such an amp for those 1% occasions where you do need 10-50 times the power to really blast your neighbours for half an hour or traumatise the wildlife.
I think that this lower end of the power spectrum is worthy of closer attention and care to define what is necessary to retain hi-fi in domestic circumstances with lower 8 ohm power. Maybe there is more data and good ideas out there?
With reasonably sensitive domestic speakers and limited range digital sources,that seems like a fairly reasonable statement. It's also something that has been a nagging concern in my mind for many years with respect to efficiency and design appropriateness.
Others who perhaps cruise around with megawatt car systems at suicidally high SPLs probably feel it's too quiet back home with only a couple of kilowatts and a wall of speakers in the bedroom. I can't live with that myself and set my goal at where I can discern all I want to hear at min. SPL and calculate, test and mostly just listen to what is sufficient to deliver it with some 50% power reserve.
Gross overkill may annihilate any concerns of overload and give great bragging rights but it also annihilates your cash and the chance try out some other DIY things. Here is the chance to also build such an amp for those 1% occasions where you do need 10-50 times the power to really blast your neighbours for half an hour or traumatise the wildlife.
I think that this lower end of the power spectrum is worthy of closer attention and care to define what is necessary to retain hi-fi in domestic circumstances with lower 8 ohm power. Maybe there is more data and good ideas out there?
Member
Joined 2009
Paid Member
my 2c...
It seems to me that unless you have access to unlimited heatsink (like Nelson Pass) high power invariably means SS with gnf, large power supplies and a road that leads perhaps to Class D. It takes great skill to design a very high power amp and the challenges are a great interest for DIY.
But for beginners like me the attractions of low power are lower voltages, potentially less cost, design options that include tubes, no-feedback and Class A. Gives us a chance to learn about different things.
Hi Bigun
I recall you are looking still at great PC desk sound. This is spot-on for low power application but as some have said here in this thread, small amps per se, just don't cut it.
I know there are lots of design approaches but I'm confident ss can do it without need to go to class A or tubes. Just don't know how precisely (Lol)
I recall you are looking still at great PC desk sound. This is spot-on for low power application but as some have said here in this thread, small amps per se, just don't cut it.
I know there are lots of design approaches but I'm confident ss can do it without need to go to class A or tubes. Just don't know how precisely (Lol)
Member
Joined 2009
Paid Member
Howdy Ian,
'per se' means 'in itself' - but you have to consider the speaker. Horn speakers are widely known to provide high SPLs with good bass 'kick' but require only low power amplifiers.
desktop PC applications are a strange category - small speakers operating at low volume levels - probably suffer a lot from Fletcher-Munsen effects.
Just wondering how is it possible to damage speakers with not enough power? By making the amp start clipping then this damages the driver?
In power terms the bass and mid take about equal punishment.
The treble driver takes a tiny portion of the average power, possibly around 5%.
If you feed a mixture of sinewaves to the speakers, the crossover will split the signals up and deliver the power proportioned between the drivers.
Since these are all sinewave in the audio band the treble sees the simple sinewave power of the few frequencies that the crossover sends to it. That 5% again.
Now change that mixture of sinewave signals to a mixture of squarewave signals. The total power delivered is doubled.
The sinewave portion of the power has not changed. all the extra power is in the higher harmonics of each of the original sinewave signals starting with the 3rd harmonic and going to way beyond the audio band.
The proportion of HF power in the squarewave signal that is sent as 3rd harmonic and higher is the same as the power in the original signal.
However crossovers send much of this extra 100% of power to the Mid and to the treble.
As a result the heavily clipped near squarewave form has a very high proportion of HF power content. Not only 5% doubled but possibly 5%doubled + 40% of harmonics of the fundamentals fed to the mid range.
Total power to the treble could be 50% instead of 5%. This is a tenfold increase in power being dissipated in the treble driver. Expect it to live how long?
As you turn the volume up from no clipping to slight clipping you can hear the increase in treble content. Turn to moderate clipping and all the treble content becomes quite unbearable and less recognisable.
Go to severe clipping and the treble driver heats up very quickly.
No matter how many papers I read that claims the contrary, I have never found any argument to convince me other wise.
The treble driver takes a tiny portion of the average power, possibly around 5%.
If you feed a mixture of sinewaves to the speakers, the crossover will split the signals up and deliver the power proportioned between the drivers.
Since these are all sinewave in the audio band the treble sees the simple sinewave power of the few frequencies that the crossover sends to it. That 5% again.
Now change that mixture of sinewave signals to a mixture of squarewave signals. The total power delivered is doubled.
The sinewave portion of the power has not changed. all the extra power is in the higher harmonics of each of the original sinewave signals starting with the 3rd harmonic and going to way beyond the audio band.
The proportion of HF power in the squarewave signal that is sent as 3rd harmonic and higher is the same as the power in the original signal.
However crossovers send much of this extra 100% of power to the Mid and to the treble.
As a result the heavily clipped near squarewave form has a very high proportion of HF power content. Not only 5% doubled but possibly 5%doubled + 40% of harmonics of the fundamentals fed to the mid range.
Total power to the treble could be 50% instead of 5%. This is a tenfold increase in power being dissipated in the treble driver. Expect it to live how long?
As you turn the volume up from no clipping to slight clipping you can hear the increase in treble content. Turn to moderate clipping and all the treble content becomes quite unbearable and less recognisable.
Go to severe clipping and the treble driver heats up very quickly.
No matter how many papers I read that claims the contrary, I have never found any argument to convince me other wise.
Last edited:
It's usual for the tweeter to get damaged by clipping as it cannot withstand sustained rail voltage (flat part of the clipped wave is at full rail voltage). Though this looks like DC, the cap for high pass filtering on the tweeter will not block it, as it has harmonics and it passes through like AC.
Hard to say - in everyday use F5 awithout preamp and CD payer with line out level down -20db still many CD is playing too loud with 96db speakers.
For other speakers set (89db and 90db) i use Leach and gianclone, but the output is not more then 15-30W or so for everyday listening.
I will soon finish the construction of Le monstre 8W amp, then i will compare it with F5.
For me looks -the optimum power 5-10W with the 93-96db speakers.
For other speakers set (89db and 90db) i use Leach and gianclone, but the output is not more then 15-30W or so for everyday listening.
I will soon finish the construction of Le monstre 8W amp, then i will compare it with F5.
For me looks -the optimum power 5-10W with the 93-96db speakers.
let's take the upper of both figures: 10W and 96dB/W @ 1m
10W adds 10dB to the 96 giving a sinewave rms SPL @ 1m of 106dB.
The peak SPL on a transient can be +3dB more giving 109dBpk @ 1m.
A pair of speakers give a further +3dB arriving at 112dBpk @ 1m.
Subtract 8dB for a listening distance of 2.5m and the listening position peak SPL is ~104dBpk.
This is high enough for many listening who require 20dB of headroom for transients. Your 106dBpk SPL allows for average listening level of ~83dB
Many will be happy with this.
Some will settle for a bit less, others demand more.
However if you take the lower of both figures: 5W into 93dB/W @ 1m then the maximum average level is now only 77dB at the listening position. Most will not accept this as loud enough.
That would be me 🙂
It is of no use whatsoever if I can't turn it up once in a while, regardless of any 'normal' listening level.
Hello,Remember the human ear-brain combo in this loudness thing. If the amplifier does not fart and stick to the rails clipping produces harmonics that are interpreted as loudness more so to the ear-brain than the real thing. SS creates square top clipping besides sounding loud sounds really bad, kills tweeters too.
DT
All just for fun!
Remember the human ear-brain combo in this loudness thing. If the amplifier does not fart and stick to the rails clipping produces harmonics that are interpreted as loudness more so to the ear-brain than the real thing. SS creates square top clipping besides sounding loud sounds really bad, kills tweeters too.
DT
All just for fun!
That is a rather ill informed synopsis of solid state amplification! You need to learn a few of the basic facts before you weigh in.
Solid state amps don't have to clip and 'kill tweeters' - evidence is my tweeters are all still in excellent condition.
Funny ... This was in the comics this morn ...
An externally hosted image should be here but it was not working when we last tested it.
Remember the human ear-brain combo in this loudness thing. If the amplifier does not fart and stick to the rails clipping produces harmonics that are interpreted as loudness more so to the ear-brain than the real thing. SS creates square top clipping besides sounding loud sounds really bad, kills tweeters too.
DT
All just for fun!
I know....🙄 and tubes have that Fog before the sound but i wouldn't hold it against them ...😛
Shouldn't that be 6 dB when you include the xtra power...
I've killed 2 tweeters with a 35 W/ch tube amp. Maybe it was all the pops on those used records by the Doors & Rolling Stones. I do tend to play Stones really loud. My new tweeters are "liquid cooled".
Hello,Are we a little sensitive ey?
Although I did not take a direct approach to the need for headroom, soft clipping, limiters and the like I am informed. I do believe that I made my point.
I have a 500 watt SS amplifier driving my near field monitors, no clipping there and happy tweeters too.
Alesis RA-500 and Behringer B2031P’s, there are dual triodes in the RIAA preamplifier.
(When pushed tubes will hard clip and kill tweeters too!)
DT
All just for fun!
Last edited:
Well, you made a sweeping statement without qualifying it. The reaction should have been anticipated. 😉
- Status
- Not open for further replies.