Amplifier specifications

[msfeinstein]

Looking for a new amplifier, I'm utterly baffled by the sheer quantity of specifications available. Power output is always the one bandied about, but there are about a dozen I've seen commonly used (THD, input/oytput impedance, seisitivity, gain, damping factor, etc.). Has anyone ever seen a guide to these values to which they can point me? It's not always obvious (to me, at least) whether a higher or lower value is better for a given specification. It's also not obvious what a "good" or even an "acceptable" value would be for each measurement.
Ultimately, it boils down to choosing the one that sounds good to me. But, I'd like to have the specifications to back it up in case I encounter a previously un-auditioned situation with my equipment. Thanks.

Edit - I've read the lengthy "ALL CHANNELS DRIVEN FALLACY" thread, and while this was extremely helpful (thanks to all who contributed), I'm still left wondering which (if any) specifications I should be focusing on.

 

[mtrycrafts]

Edit - I've read the lengthy "ALL CHANNELS DRIVEN FALLACY" thread, and while this was extremely helpful (thanks to all who contributed), I'm still left wondering which (if any) specifications I should be focusing on.

Concentrate on RMS power 20Hz-20kHz into what loads ( the lower loads rated the more difficult speakers it will drive properly), frequency response, THD +N, are the major ones you should compare. Damping factor is a non issue in SS amps.

 

[tbewick]

According to an old book I have (1980, Practical Hi-fi Sound, Roger Driscoll - MSc, PhD, lecturer in acoustics, was a reviewer for Gramophone) the amplifier (not tube) is "the strongest link in the reproduction chain being, as it is, a wholly electronic device". He grouped possible signal degradations into: noise, distortion, frequency response, and crosstalk.

In most modern decent quality amps/receivers, noise shouldn't be a problem. Check the signal-to-noise ratio of the pre-amp, power-amp, and DAC (if using a digital receiver) components. Preferably the S/N should be greater than 100 dB for all.

Distortion should also not be a problem with modern amplifiers. Look for THD ratings for the same components. Ideally it should be measured across 20 Hz to 20 kHz. It should be below 0.5 %. I'm under the understanding that other types of distortion like intermodulation distortion, transient intermodulation distortion, and slewing induced distortion shouldn't be an issue in modern designs.

Frequency response should be a non-issue. Look for 20 Hz - 20 kHz, +/- 1 dB for the same components. Crosstalk should also be a non-issue.

You can find more info on the Rane website:
http://www.rane.com/digi-dic.html

40 watts power should be sufficient in most cases, but ideally you'd want nearer to 100 watts. I've found this link from Audioholics useful in relating power to sound output:

http://www.myhometheater.homestead.com/splcalculator.html

 

[tbewick]

Oh yes, if you're getting a digital receiver, then I think it will have an analogue-to-digital converter (ADC). I think this is used for sending bass to the subwoofer and is required for digital signal processing effects. I'm guessing that the performance of the digital filter used in setting the subwoofer crossover frequency will be considerably better than the performance of the passive crossovers in the speakers.

You could look for the specifications for this (the ADC) as well. I've never seen specifications published for this component, but maybe you'll be more fortunate.

 

[highfihoney]

Damping factor is a non issue in SS amps.

this is not a true statement & not based on experience,beware of solid state amp's with damping factors under 100.

 

[mtrycrafts]

this is not a true statement & not based on experience,beware of solid state amp's with damping factors under 100.

I need better evidence than most likely a biased perception.

http://www.audioholics.com/techtips/audioprinciples/amplifiers/dampingfactor.php
But many of the claims made, especially for the need for triple-digit damping factors, are not based in any reality, be it theoretical, engineering, or acoustical.

 

[highfihoney]

I need better evidence than most likely a biased perception.

the evidence can be had by going out & buying 2 amplifier's with specs that are the same except damping factor then driving both amp's to their limit,it's very obvious at that point.

 

[mtrycrafts]

the evidence can be had by going out & buying 2 amplifier's with specs that are the same except damping factor then driving both amp's to their limit,it's very obvious at that point.

So, you don't have evidence then, just speculations. What does DF have to do with driving an amp to the limits? How did you do your experiment? What controls? Protocols? Amps?

However, if your assertion was correct, there would be a lot of not all DBT would turn out to be a positive difference, no? Why not?

Did you read the link? Are they wrong? That need even more evidence to show them wrong.

 

[highfihoney]

So, you don't have evidence then, just speculations. What does DF have to do with driving an amp to the limits? How did you do your experiment? What controls? Protocols? Amps?

haha,your like a pit bull i swear

the evidence is in performance in many rig's that ive built & listened to over the last 25 years,how damping factor relates to an amp being driven to it's limits is plainly heard as a loss of bass response in any true full range speaker system that can produce a full bass response.

control? protocol? dont be silly when i can go to my closet & pull out 2 different 200 watt solid state amplifiers each with identical spec's except damping factor & 1 will give a great bass response at all levels while the other amp clearly loose's control of the bass what other spec is there to look at to determine why?

you asked"what does df have to do with driving an amp to it's limits",the answer is simple,people do actually listen to loud music on their system's & they also have the tendency to push their amplifier's & speaker's to their limit while listening,i know the cool answer from many people is that they never drive the crap out of their gear but that is not based on truth most of the time.

ever see a subwoofer amplifier with an extremely low damping factor? i never have.

 

[tbewick]

Highfihoney, surely you realise that objective tests are needed alongside subjective ones. This is because subjective tests are less reliable than objective ones. I have an example.

I'm quite fond of this observation and I've mentioned it in a couple of posts, but What Hi-Fi? magazine claim that DAC's in CD players need a break-in period. This is probably a purely subjective observation. Do you honestly think that this is true? Without objective tests, it is difficult to disprove.

 

[highfihoney]

Highfihoney, surely you realise that objective tests are needed alongside subjective ones. This is because subjective tests are less reliable than objective ones. I have an example.

I'm quite fond of this observation and I've mentioned it in a couple of posts, but What Hi-Fi? magazine claim that DAC's in CD players need a break-in period. This is probably a purely subjective observation. Do you honestly think that this is true? Without objective tests, it is difficult to disprove.

very good point i dont believe in any break in period in solid state gear but your post does touch on a very good subject which is ridiculous claims made about audio,i think part of the problem for many people is being able to tell the difference between whats nonsense & what is fact based .

i have no beef with exposing these mostly idiotic claims for what they are,i also agree that objective tests hold alot of value but none the less these tests are not the holy grail of hifi & can only go so far when building a system or when giving advice on building a system.

what i dont agree with is the people who think just because they can read a spec or google some info that qualifies them to make recomendations to other people as to what gear to buy,as with any other published info manufacturers specs can be misleading in many ways,one recomendation for an amplifier in another thread was based on specs & hearsay from a member who didnt even know that a khorn had a woofer,is that good advice?

i also believe that dbt testing is a useless & mute point that is not relevant at all when it comes to making recomendations or discussing performance issues.

i dont want anybody to get me wrong & think that i believe objective tests are not usefull or that specs dont count but there is a whole lot more to building a great sounding rig than just reading specs.

have a great evening.

 

[tbewick]

I sort of agree with your point of double blind tests. I prefer to have other physical tests (ie. measurements made without human involvement) to reinforce DBT's. I do think though that they are a useful way of easily testing certain observations. For speaker cable, if in a DBT you can't statistically tell the difference between cable costing £10 and a cable costing £5000, then it at least it does tell you that the audible difference between them must be subtle. Strictly as a scientist you would dismiss there being any audible difference. I don't really like statistics but I do accept it is useful. For testing whether a type of distortion is audible or not, then you probably have to use statistics. This is in some ways analogous to drug testing in medicine, where the exact physical mechanism for how a drug works is so complex that you have to rely very heavily on statistics as a measure of effectiveness.

 

[PENG]

I sort of agree with your point of double blind tests. I prefer to have other physical tests (ie. measurements made without human involvement) to reinforce DBT's. I do think though that they are a useful way of easily testing certain observations. For speaker cable, if in a DBT you can't statistically tell the difference between cable costing £10 and a cable costing £5000, then it at least it does tell you that the audible difference between them must be subtle. Strictly as a scientist you would dismiss there being any audible difference. I don't really like statistics but I do accept it is useful. For testing whether a type of distortion is audible or not, then you probably have to use statistics. This is in some ways analogous to drug testing in medicine, where the exact physical mechanism for how a drug works is so complex that you have to rely very heavily on statistics as a measure of effectiveness.

Excellent point! DBT speaker and interconnect cables do not require much controls such as level matching etc. If you cannot tell the difference, either difference does not exist, or subtle.

 

[Buckeyefan 1]

Looking for a new amplifier, I'm utterly baffled by the sheer quantity of specifications available. Power output is always the one bandied about, but there are about a dozen I've seen commonly used (THD, input/oytput impedance, seisitivity, gain, damping factor, etc.). Has anyone ever seen a guide to these values to which they can point me? It's not always obvious (to me, at least) whether a higher or lower value is better for a given specification. It's also not obvious what a "good" or even an "acceptable" value would be for each measurement.
Ultimately, it boils down to choosing the one that sounds good to me. But, I'd like to have the specifications to back it up in case I encounter a previously un-auditioned situation with my equipment. Thanks.

Edit - I've read the lengthy "ALL CHANNELS DRIVEN FALLACY" thread, and while this was extremely helpful (thanks to all who contributed), I'm still left wondering which (if any) specifications I should be focusing on.

Some things I would look for when purchasing a quality amp:

*Rated at full bandwidth of 20-20,000Hz, and not 1kHz.
*Rated at .05% THD or less.
*For full range speakers (as opposed to a subwoofer) - class A or A/B amplifier.
*At least 100 watts per channel, and believe it or not, a maximum of two channels. When you try driving 3 or more channels, the heat sink would need to be extremely large to handle the high temps at reference levels.
*4 ohm stable, and a bridged mono option
*For a two channel amp, I'd look for one with a weight over 40lbs. at 100wpc rms (you're talking most likely vintage, commercial, or high end here) The beefier the amp, the larger the power supply and heat sink. Most of these goliaths have massive capacitors that have huge on hand "stored up" power for those demanding peaks. Class A/B amps are somewhat efficient (class A are not), but produce heat. Heat can create distortion and damage electronics. How amplifiers deal with heat usually equates to performance. Heat damages electronics, therefore amplifiers that keep heat away from electronics will perform better than those that don't. Pretty simple.

Here's three examples:
1) Audio Source Amp 300 ~$399
150 watts x 2, 20-20,000 at .2% THD; 28.6 lbs.

2) Mackie M1400 ~$599
250 watts rms x 2, 20-20,000 at .012 THD; 36lbs.

3) this: http://cgi.ebay.com/Pioneer-Elite-M-90-Power-Amplifier_W0QQitemZ5863126083QQcategoryZ94900QQssPageNameZWDVWQQrdZ1QQcmdZViewItem

4) this: http://cgi.ebay.com/MONSTER-BOSE-POWER-AMPLIFIER-WITH-METERS_W0QQitemZ5860386630QQcategoryZ39783QQrdZ1QQcmdZViewItem

5) this:
http://cgi.ebay.com/YAMAHA-M-80-Stereo-Amplifier-Class-A_W0QQitemZ5862801137QQcategoryZ71550QQrdZ1QQcmdZViewItem

http://www2.yamaha.co.jp/manual/pdf/av/english/SepA/M-80.pdf
Check the weight on the last three. They just don't make em like they used to (for the price)! And yes, that is a Bose amplifier - one of the crown jewels of Bose in its day.

 

[highfihoney]

Ha!

buckeyefan, i allmost fell out of my chair when i saw your link to the bose 1801 amp,i used to own an 1801 & it was one of the best solid state amps i have ever ran,to this day i regret selling that amp.

boy did bose sure take a wrong turn since the 1800 & 1801 amps,alot of younger guys dont realize that bose actually used to make some world class gear.

SWEET LINK

 

[mtrycrafts]

haha,your like a pit bull i swear

the evidence is in performance in many rig's that ive built & listened to over the last 25 years,how damping factor relates to an amp being driven to it's limits is plainly heard as a loss of bass response in any true full range speaker system that can produce a full bass response.

control? protocol? dont be silly when i can go to my closet & pull out 2 different 200 watt solid state amplifiers each with identical spec's except damping factor & 1 will give a great bass response at all levels while the other amp clearly loose's control of the bass what other spec is there to look at to determine why?

you asked"what does df have to do with driving an amp to it's limits",the answer is simple,people do actually listen to loud music on their system's & they also have the tendency to push their amplifier's & speaker's to their limit while listening,i know the cool answer from many people is that they never drive the crap out of their gear but that is not based on truth most of the time.

ever see a subwoofer amplifier with an extremely low damping factor? i never have.

Well, for a moment I thought you may have something up your sleeve but I was disappointed. I ran into a plumber who claimed to be a master plumber, many years of experience and taught the classes, etc. Come to find out he was off base by a mile. All those years, he was wrong. It happens all the time.

 

[mtrycrafts]

i also believe that dbt testing is a useless & mute point that is not relevant at all when it comes to making recomendations or discussing performance issues.

Performance issues are testable and that is exactly why one must do a DBT comparison if reliable info is to be gained and useful recommendations made. If there is no audible differences, then the other features are easy to select for ones needs.