I've been going crazy for years with the power of amplifiers, my cabinets are active I have a 120 db coaxial compression driver, a 101 db mid-bass and a 94 db sub.
With a low power amp the compression driver is supposed to sound like a deafening dynamic single watt, but if I plug in 20w or 10 amps it sounds really loud but the sound lacks impact, everything changes when I plug 1000w amps into above. Right now I have 5000w for the sub 5000w for the midbass 3000 for the midhigh and 3000 for the treble the sound has much more dynamics and impact at any volume and the amps don't even turn on the signal lights.. .but if I reduce the power to something more logical, the dynamics, the physics and the impact of the sound disappear. Can someone explain to me why the power in abundance is noticeable even at medium or low volumes?
With a low power amp the compression driver is supposed to sound like a deafening dynamic single watt, but if I plug in 20w or 10 amps it sounds really loud but the sound lacks impact, everything changes when I plug 1000w amps into above. Right now I have 5000w for the sub 5000w for the midbass 3000 for the midhigh and 3000 for the treble the sound has much more dynamics and impact at any volume and the amps don't even turn on the signal lights.. .but if I reduce the power to something more logical, the dynamics, the physics and the impact of the sound disappear. Can someone explain to me why the power in abundance is noticeable even at medium or low volumes?
Higherr slew rate, no real flutuction in power supply voltage , more power available in class A operation with much colder transistors, less crossover distortions to be corrected by the feedback mechanism.
Great question. Would love to hear some discussion on this topic. I am using McIntosh MC402 on the Klipschorn midrange and tweeter and it sounds great. I have been wondering if it would sound even better with a smaller amp.
Dan
Dan
All amplifiers are the same ;^)
How does the amplifier peform at very low power levels? How well does it recover from clipping? What form is the hrmonic structure of the distortion prodcuts, how much information does the amplifier lose?
Ohave played with 20w amplifiers that, even on very resistive speakers, will play louder before audible clipping than a 200w amplifier.
Single Ended, single output devices, can have a “purity” at very small power levels, mW, that lose a minimum amount of the tiny bits of information.
Class A PP will trade 2 devices averaging vrs a (likely) more linear amplifier.
The speaker, and its impedance play a big role in which works better for you.
A pair of ACA is a good budget amp whwew this can be explored. They can be run SE, paralleled SE or bridged/balanced (PP).
dave
How does the amplifier peform at very low power levels? How well does it recover from clipping? What form is the hrmonic structure of the distortion prodcuts, how much information does the amplifier lose?
Ohave played with 20w amplifiers that, even on very resistive speakers, will play louder before audible clipping than a 200w amplifier.
Single Ended, single output devices, can have a “purity” at very small power levels, mW, that lose a minimum amount of the tiny bits of information.
Class A PP will trade 2 devices averaging vrs a (likely) more linear amplifier.
The speaker, and its impedance play a big role in which works better for you.
A pair of ACA is a good budget amp whwew this can be explored. They can be run SE, paralleled SE or bridged/balanced (PP).
dave
I think I figured out once that it costs about the same for a huge a/b amp vs a high bias amp with a lower rating. Power supplies are substantial in both…
I listen to classical sources, that can go 55 db above soft level (pianissimo) on LP or 70 on CD. I can get away with running 1 Vav on 8 ohms for soft level with my 98 db 1w1m speakers, but I doubt if many posters here have any speaker that efficient. That is 1/8 watt. so + 10 db is 1/4 watt, 20 1/2 w, 30 1 w 40 2 w 50 4 w, 60 8 w 70 16 w. If the dynamics on source are going to be followed you have to have that much power. If your speakers are 80 db 1w1m quadruple all the wattages. No peaks, the music is boring and polite. My amps, when the rail or B+ capacitors are bad, sound that way. Time for re-e-cap. So 64 watt minimum required to follow the source on normal acoustic suspension speakers.
On the same speakers at indoor music room volumes I hear no difference between my 70 w/ch amp (dynaco ST120 with djoffe bias mod), 120 w/ch amp (Peavey PV-4c), 260 w/ch amp (Peavey CS800s). My 35 w/ch amp (dynaco ST70) is noticeably inferior, more for the 1% harmonic distortion and the higher noise level of a 1960 design than the peak voltage. All the others are .1% hd down to .03% hd.
I have a 400 w/ch amp but have been afraid to use it. One cable tripped over and pulled out could make a very high level pop. My speakers are rated 300 w rms so whatever the 260 w/ch amp does should not hurt them.
Because big amplifiers look cool!!😉
I'm being the devils advocate here, on one hand you are likely experiencing some confirmation bias because you know what amplifier you are listening to, if you were to do a blind comparison you may have a harder time picking out the small amp.
But on the other hand all amplifiers are not created equal, and in this case you're not even comparing a bunch of different equally sized amps you're comparing a doorbell to an arc welder so everything about the way they are designed is different. What topology are the amps in question? Class D amps often have a high frequency response that changes with load impedance while the other types don't behave this way. I'm not suggesting class D amp all sound bad, and it's not just about having big power reserves on hand, there are high powered amps that still sound dull and restrained compared to others. Everything matters.
3000 watt amplifier would have a very healthy power supply and more output devices.
Sounds like a live sound amplifier. Not sure what the THD rating would be.
Don't understand I've had numerous compression drivers at home.
I ran them in studios and warehouses on high power.
And at home with incredible average 100 watt per channel home receivers.
At low volume sound the same.
Some had simple crossover networks, some with basic lamp limiters.
I would question the overall health or topology of the 20 watt amplifier being used.
A 50 to 75 watt compression driver doesnt need 3000 watt amplifier.
I would question more the health of amplifier or crossover/L pad
Within reason high power amplifier at 3000 watts isn't going to have very
good high frequency distortion rating , so it makes no sense.
Sounds like a live sound amplifier. Not sure what the THD rating would be.
Don't understand I've had numerous compression drivers at home.
I ran them in studios and warehouses on high power.
And at home with incredible average 100 watt per channel home receivers.
At low volume sound the same.
Some had simple crossover networks, some with basic lamp limiters.
I would question the overall health or topology of the 20 watt amplifier being used.
A 50 to 75 watt compression driver doesnt need 3000 watt amplifier.
I would question more the health of amplifier or crossover/L pad
Within reason high power amplifier at 3000 watts isn't going to have very
good high frequency distortion rating , so it makes no sense.
If you hear a difference in dynamics between two amplifiers, that would suggest that one of them is being stressed to its limit and the other one less so. Do not make the mistake that the peak power requirements drop as fast over frequency as the AVERAGE power requirements do - peak voltages going into tweeters can be surprisingly high even if it is seeing single-digit average watts (and not catching on fire). If you don’t believe it, put an oscilloscope on the tweeter and turn the volume up to “uncomfortable”.
Hello,
One other possibility. Amps with lots of parallel output transistors (Class AB) tend to have less crossover distortion. Described well in Douglas Self's book on amps.
All amplifiers are NOT the same. No way. Many are getting very good, but far from the same.
I believe one of the major differences is the capability of the power supply to hold up the rails. Smaller amps tend to have far less capacitance. Larger amps more. So you don't get the dynamic compression of the rails collapsing. This is still below clipping. Go listen to a Rotel integrated vs the competition and you will hear what a lifeless amp with no power supply sounds like.
Some people just look at the uF of the caps but forget to calculate the coulombs as the voltage is different. I know when I tripled the bank in my 60W amp, I could tell the difference in very dynamic recordings when listened to enthusiastically. Lower levels, not so much.
A well designed amplifier, regardless of the number of pairs, will have insignificant crossover distortion. But, for a given power, 2 pairs vs one pair will have less current per pair and so work in a more linear range and with less thermal drift. But, twice the idle current of course.
One of the hallmarks of an amp that can deliver the dynamic power is how close to double the power into 4 Ohms as opposed to 8. Mostly a combination of more output pairs as well as the power supply capability. I find this to one of the indicators of an amp that will sound ( or not sound as the goal is) for all music.
Another tid-bit. Some find very clean amplifiers to be "lifeless" compared to their old amp. What they are hearing is a lack of all the harmonic distortion the old amp was producing. All those harmonics can bring up your perception of more "detail" up high. Some harmonics make it feel "warmer" . Silly audiofool terms, but most know what they mean. I do not know the amps the OP is mentioning, but they sound like PA amps which may have more than .1% distortion, some over 1.
These defects can be easily measured. They are physical. They can be corrected by better design which of course means more money.
I believe one of the major differences is the capability of the power supply to hold up the rails. Smaller amps tend to have far less capacitance. Larger amps more. So you don't get the dynamic compression of the rails collapsing. This is still below clipping. Go listen to a Rotel integrated vs the competition and you will hear what a lifeless amp with no power supply sounds like.
Some people just look at the uF of the caps but forget to calculate the coulombs as the voltage is different. I know when I tripled the bank in my 60W amp, I could tell the difference in very dynamic recordings when listened to enthusiastically. Lower levels, not so much.
A well designed amplifier, regardless of the number of pairs, will have insignificant crossover distortion. But, for a given power, 2 pairs vs one pair will have less current per pair and so work in a more linear range and with less thermal drift. But, twice the idle current of course.
One of the hallmarks of an amp that can deliver the dynamic power is how close to double the power into 4 Ohms as opposed to 8. Mostly a combination of more output pairs as well as the power supply capability. I find this to one of the indicators of an amp that will sound ( or not sound as the goal is) for all music.
Another tid-bit. Some find very clean amplifiers to be "lifeless" compared to their old amp. What they are hearing is a lack of all the harmonic distortion the old amp was producing. All those harmonics can bring up your perception of more "detail" up high. Some harmonics make it feel "warmer" . Silly audiofool terms, but most know what they mean. I do not know the amps the OP is mentioning, but they sound like PA amps which may have more than .1% distortion, some over 1.
These defects can be easily measured. They are physical. They can be corrected by better design which of course means more money.
I have measured the signal on the mid/tweeter section and the woofers. At normal listening, I am using 200 to 300 milliwatts (yes milliwatts) for satellites, and about 1 watt for the woofers. This at average 80dB SPL.
When I crank it up with friends during party, I measure ~1-2 watts at mid/tweeter section, and ~4-6 watts at woofer section. Measured with oscilloscope and pink noise. Levels reaching 100dB.
Normal size living room. Dynamic as the recording allows, since most are compressed.
Not sure what you are talking about, unless you have super low resistance super low efficiency speakers, like apogee scintilla.
Imho the subjective dynamic ease of big amps at even modest power levels is a combination of several factors
The more interesting question is whether a big amp can also have the subjective speed and agility of a small amp. This appears to be a rarity.
- bigger transformers
- bigger caps
- higher voltage
- multiple outputs
- heavy duty wiring down to the pcb tracks
The more interesting question is whether a big amp can also have the subjective speed and agility of a small amp. This appears to be a rarity.
This is a really interesting thread. I have been thinking about loudspeaker and amplifier dynamics for a few years. Is there a way to measure dynamics? If the speaker is fed a steady signal (to set input volume) while measuring loudspeaker output, and then fed a known impulse (or known louder signal), can we measure the response to determine if the correct peak was produced by the loudspeaker. Anybody know how to calculate what the correct output is suppose to be.
I do not have the smarts to figure out how to do this, but is there a way using REW (or something else) to calculate how many db loudspeaker increase there should be for a given db input change? Seems like someone could write a program to test loudspeakers at many different frequencies and many different input spikes (not sure what to call it) while measuring the speaker output to get a handle on dynamics. Of course, the dynamics is a function of preamp, amp and loudspeaker (and probably other things).
Any thoughts on this?
Dan
I do not have the smarts to figure out how to do this, but is there a way using REW (or something else) to calculate how many db loudspeaker increase there should be for a given db input change? Seems like someone could write a program to test loudspeakers at many different frequencies and many different input spikes (not sure what to call it) while measuring the speaker output to get a handle on dynamics. Of course, the dynamics is a function of preamp, amp and loudspeaker (and probably other things).
Any thoughts on this?
Dan
Rew already has impulse and step response options, not to mention decay waterfall plots and so on.
One issue may be that when you do measurements with high power amps, then you swith to lower power amps, some adjustments may be necassary.
How do you know you are measuring under exactly the same conditions.
Let see what thread starter will come up with. He better provide some data, otherwise its just imagination.
One issue may be that when you do measurements with high power amps, then you swith to lower power amps, some adjustments may be necassary.
How do you know you are measuring under exactly the same conditions.
Let see what thread starter will come up with. He better provide some data, otherwise its just imagination.
I used to be soundman for amateur music group at work. I often brought 7500 watts behringers home for fun. Sound quality was not there in comparison to well executed classA amp.
Dan, imagine these two scenarious.
You have relatively highly efficient speakers, and are using few 60 watts amps. You are using less than watt, and you perform your dynamics test.
Then you switch to 5000 watts amps, still using less then watt, and perform your dynamics test.
Do you expect to find significant difference?
I doubt that.
(You may find different harmonics profile, depending on amps construction, but that is totaly different issue)
"speed and agility" have nothing to do with big or small amps. That is a design factor independent of power.
Higher voltage is irreverent if you never use it. Maintaining the voltage is.
Do not mistake this with any engineering related quantity. It is obviously an entirely subjective observation, based on years of cohabiting with large Krell, Threshold, Jeff Rowland amps. From what i have heard at shows, more recent amps, and especially ones that don't suffer loop feedback seem to be improving but there is always a compromise.
This compromise, which at least in part is due to the massively parallel outputs, has produced some very interesting attempts to combine dynamics with speed and clarity. Gamut and the related Reimyo and Misiklab Ultimate Audio come to mind. Despite being high power these amps use a single pair of 500W, 400A peak mosfets in order to avoid paralleling. I have heard the Reimyo a few times but it was inconclusive. Perhaps someone can share impressions about the Gamut.
Then there is the power supply. I am not entirely convinced that a bunch of monstrously sized capacitors can help produce fast and agile sound. Perhaps an LCLC filter would be an improvement but such designs are very seldom seen in commercial equipment.