Dismiss Notice
Attention AVForums app / Tapatalk users
Sadly GDPR means that, from 25th, we can no longer offer access to AVForums via the branded app or Tapatalk.
Click here for more information.

How much power do you need?

Discussion in 'AV Receivers & Amplifiers' started by Nic Rhodes, Sep 16, 2003.

  1. Nic Rhodes

    Nic Rhodes
    Well-known Member

    Joined:
    Mar 23, 2001
    Messages:
    17,152
    Products Owned:
    0
    Products Wanted:
    1
    Trophy Points:
    133
    Location:
    Cumbria
    Ratings:
    +1,276
    How much power do you need?

    Well as many of you know I have been looking for a long time for a new amplifier. I have posted questions both here and elsewhere, as well as quizzing many shops who are trying to sell me the amplifiers. What has become apparent is how few people know how much power is needed. I am fed up with people telling me I should buy a Rotel as it has good current capability. To be brutally frank these people don't understand how amplifiers work or more to the point how the amplifier / speaker interactions work. So I decided to write on how to work out how you pick 'what you need' and then you can compare it with what you have. Many may then be surprized just how far 'from the mark' they are. This is the main reason for the differences people hear between amplifiers as distortions rise with output power as the amplifier approaches it's maximum output. I am not really talking about continuous output here as this is relatively low but more the very large peaks (that get distorted if you have too
    little power). If people want to see more info on this I can add it a later stage.

    So what matters:

    1 Amplifier output (in Watts)
    2 Speaker sensitivity (in dB for 1W input at 1 m)
    3 The room or more importantly how far you sit from the speaker?
    4 Do you use equilization?
    5 Do you use a subwoofer?
    6 How loud do you listen to music?

    Not once have I been asked 3 - 6 in the shop, and only about one time in two am I asked about point 2!

    A watt is a watt. This may sound daft but a watt from a SE Triode amplifier is EXACTLY THE SAME as a watt from a Krell or a Bryston or a Tag (substitute your favourite here). The ability to drive a low impedance may be different for valves and transistors but it not down to power per se but more the stiffness of the power supply used. Don’t forget sensitive speakers are not always easy to drive. Kef Q series have always been real buggers to drive well due to low impedance swings in the bass region.

    Speakers are just 'dumb' devices and don't know what they are doing. Give them a watt and they produce an acoustic output according to their sensitivity but it is hard to correlate amplifier power in watts with acoustic output. We
    need to specify amplifier output in dBW so look at this table for a few common ones (and the odd extreme)

    10W 10dBW
    20W 13dBW
    50W 17dBW
    100W 20dBW
    200W 23dBW
    250W 24dBW
    400W 26dBW
    1000W 30dBW

    every doubling of power gives another 3 dB of acoustic output.

    Right for the important bit. A decent system should be able to produce 100dB of sound, an ‘enthusiasts’ system really should be looking at 105dB and the anal retentive should aim for 110 dB :))). This will allow any music to be reproduced comfortably. 9 This is why nominally the same quality amplifies (say Bryston SST range) sound different (better) with the increase in power. The amplifier is actually the same quality but it is not driven into the peak clipping range. Look inside the 120w, 300w and 600w Brystons and you will see the same quality components, only the power is different.

    Right, do you remember loudspeaker sensitivity being rate at 1m distance? Well I suspect that few of you will listen at more than 1m from your speakers and you therefore have to ‘correct’ for this. In theory there is a 6 dB loss for each additional 1m from the speaker. In ‘practice’ 5 dB is a better figure to use (as we don’t live in anaechoic chambers).

    Right armed with all this information it is easy to calculate the peak capability of a system. Take the loudspeaker efficiency, add the dBW figure for the amplifier and deduce the additional loss from a more distant listening position.

    So for

    Speaker 87 dB/ 1m
    100w amplifier (20dBW)
    3m from speaker ((3m –1m ) * 5 dB =10dB

    we have 87 + 20 – 10 = 97 dB

    actually well below my minimum (WAF) 100dB specification quoted earlier and miles away from the anal retentive 110dB despite the fact we are talking some very ordinary everyday figures here.

    So in my case with 92dB speakers, at 3 m listening, wanting 110dB peak I need an amplifier of

    110 – 92 + (2 *5) = 28dBW

    This equates to just over 500w, say 600w approx. Now my speakers can just about handle this sort of power but most can’t. I also have one of the more efficient speakers available, for most people, this will be MUCH worse (ie even more power) and now you know why amplifiers sound different.

    A few things to consider.

    Using a sub woofer will greatly ease the situation as most of the ‘power’ is required at low frequencies. Equilization may or may not effect power requirements as well. The louder you listen to music, the potentially the worse it sounds :( Finally if you bi amp your speaker you may be able to gain an additional 4 dB head room so this may actually be VERY beneficial, perhaps even more so than a nominally bigger amp?

    So the bottom line is I think a 600w amp is what people really should be aiming at for normal speakers / room. This is ‘wide’ of the current mark. Why are we so wrong?

    Still so sure your amp is fine at driving your speaker?


    :lesson:
     
  2. groundy

    groundy
    Standard Member

    Joined:
    Dec 8, 2001
    Messages:
    1,554
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    39
    Location:
    Northern Ireland
    Ratings:
    +1
    Not now I'm not :(.

    Very interesting stuff though :smashin:.
     
  3. EvilMudge

    EvilMudge
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Good post Bee!
    Was this something that came out of nowhere, or was it premeditated?

    I only ask as HFN and Musical Fidelity had a look at this not long ago. Though in their case I agreed with very little of their methods. They tried to get realistic full range volume levels out of a pair of ESL-989s:suicide: :nono:
     
  4. Steve.EX

    Steve.EX
    Active Member

    Joined:
    Oct 17, 2001
    Messages:
    3,161
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    68
    Ratings:
    +69
    Ah, but this is in absolute - all things being equal terms no?

    As you might/might not know/care Bee i use Kef 203's and 202c for front's not a million miles away in quality from your own but perhaps not the same swings in impedance.
    I have recently moved from Rotel 993's (one per speaker) which allowed tri-amping at a Rotel rated 600w into each, dynamics were fine, soundstage depth less so, i have recently moved to Tag 250mr mono-blocks (via 125m's which are now on the backs) and consider the overall "picture" to be markedly improved (i do have some "idea" of quite how and why a transistor amplifier "works") i am no stranger to (Tag) THX reference levels and have to say whilst the presentation is a litlle less "influenced" this is in no way to be considered as a dynamic compression (to which i really cannot detect losses by comparison) allied to greatly improved delineation of detail and imaging (i am aware of the Tag amp decryers who do not need to shout it down - I LIKE WHAT THE 250's DO FOR ME - END OF - yes i have played with other usual suspects!)
    So whilst large outputs (of which i have always advocated their benefits of here) are necessary/desirable/essential? there will always be the low output velvet sounding valve unit lovers with their sensitive speakers who will choose this option over absolute output (or indeed sacrifice absolute spl).and thus for the more mainstream. The choice for most of us (or has at least for me since pretty much day 1) is perhaps output or "flavour", few of us can reach the no compromise (or their abouts in amp terms) level.

    Regards

    Steven
     
  5. Steve.EX

    Steve.EX
    Active Member

    Joined:
    Oct 17, 2001
    Messages:
    3,161
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    68
    Ratings:
    +69
    Also, is not a purpose designed amplifier for a specific speaker the prefered option (is this not the cry of all active speaker owners), if you could design a unit that for example maintained (however improbable) a linear damping factor etc etc be the best solution over generic watts?


    Regards

    Steven
     
  6. Spligsey

    Spligsey
    Active Member

    Joined:
    Jan 16, 2003
    Messages:
    2,319
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    66
    Location:
    UK
    Ratings:
    +195
    As an owner & advocate of active systems, i will support that notion Steve.


    Adz.
     
  7. alexs2

    alexs2
    Well-known Member

    Joined:
    Nov 13, 2002
    Messages:
    13,957
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Ratings:
    +1,671
    Thats a big one,Beekeeper!.....interesting question though....in the case of valve amps,especially SET's,it's not just a case of a watt being a watt,as you have to factor in not only load sensitivity,but the quality of the output transformer,and especially in the case of SET's,the likelihood of core saturation with decreasing frequency.

    As for solid-state amps,I totally agree...a lot of nonsense is written about the capabilities of certain amplifiers,and it brings to mind the Crown DC300A of the late 70's....enormous power but useless sound quality.

    I think I'll stick to my Krells
     
  8. bamber

    bamber
    Standard Member

    Joined:
    Dec 2, 2002
    Messages:
    216
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    16
    Ratings:
    +0
    Brillant post Beekeeper and just what I was looking for. I'm about to order the Nautilus 800's for L/R and was thinking of Biamping with 2 x Bryston 7B SSTs and 1 x 4B SST (already own this). Using your formula this appears to work for my proposed setup.

    Is this the way to go in terms of amps or could my money be better spent?

    Should I use the balanced output to one amp and unbalanced output to the other amp from my Bryston SP1.7? Alternately is a Y I/C the way to go?

    Gaz
     
  9. General Skanky

    General Skanky
    Active Member

    Joined:
    Jul 28, 2000
    Messages:
    4,247
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    71
    Ratings:
    +43
    Oooooh, I see debate looming.:D

    It will also depend on how loud a person does actually listen to their system. (I know though, you're trying to explain the ideal).

    Yes, a good system will go to 100db+ clearly without distortion etc. However, isn't that really quite loud?

    If my spl meter is even a smidge accurate, I'd say we rarely get close to that level. Therefore my amps aren't required to get to the same levels you're talking about. Also, they shouldn't be working hard either, but driving my speakers along nicely without too much undue effort.

    Does anyone have a chart or link to how loud dB equte in real life, ie, 110 dB = jet plane taking off 3 feet away or whatever it is.:)
     
  10. General Skanky

    General Skanky
    Active Member

    Joined:
    Jul 28, 2000
    Messages:
    4,247
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    71
    Ratings:
    +43
    Found this for reference.


    Rocket Launching 180
    Jet Engine 140
    Thunderclap, Air Raid Siren 1 Meter 130
    Jet takeoff (200 ft) 120
    Rock Concert, Discotheque 110
    Firecrackers, Underground Train 100
    Heavy Truck (15 Metre), City Traffic 90
    Alarm Clock (1 Metre), Hair Dryer 80
    Noisy Restaurant, Business Office 70
    Air Conditioning Unit, Conversational Speech 60
    Light Traffic (50 Meter), Average Home 50
    Living Room, Quiet Office 40
    Library, Soft Whisper (5 Meter) 30
    Broadcasting Studio, Rustling Leaves 20
    Hearing Threshold 0

    In which case any ordinary system could reach peaks of 90 - 100 dBs more often than we think.
     
  11. nathan_silly

    nathan_silly
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    I don't agree with BK's OTT figures. I've been using low powered and high powered amps. There's really not that much in it in real world use. I'm sure people using hioh quality 30W amps weren't "in the wrong", and perfectly able to drive their speakers.

    I have no "experience" in techie talk. I just listen to it. I've used 60W and 200W poweramps. At -10dB they perform on a similar level. I don't listen at reference (too loud)

    This makes me wonder BK- do you work for a amplifier company?
     
  12. Gordon @ Convergent AV

    Gordon @ Convergent AV
    Distinguished Member AVForums Sponsor

    Joined:
    Jul 13, 2000
    Messages:
    14,201
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    166
    Location:
    Living in Surrey, covering UK!
    Ratings:
    +3,036
    Mr Bee:

    V Interesting but I have questions I need answered.

    From your comments it suggests that the only reason an amplifier may sound better is because it has a higher rated power output before distortion sets in correct? If this is so how can lower rated amplifiers sound better than higher rated ones when not played at clipping levels (or am I suffering from sighted dem syndrome....not)

    Also how do we figure in amplifiers that can be bridged to increase power output but then become current limiting?

    Are you just talking about how much power for given volume of sound at listening level or do you really think you can use this formula for quality as well?

    I'm not having a go I want to understand

    Gordon
     
  13. Gordon @ Convergent AV

    Gordon @ Convergent AV
    Distinguished Member AVForums Sponsor

    Joined:
    Jul 13, 2000
    Messages:
    14,201
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    166
    Location:
    Living in Surrey, covering UK!
    Ratings:
    +3,036
    Beekeeper is NOT an amplifier company rep. Perhaps he will let us know what he does. He is a scientist though......

    Gordon
     
  14. S H A D O

    S H A D O
    Well-known Member

    Joined:
    Mar 16, 2001
    Messages:
    10,434
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Location:
    United Kingdom
    Ratings:
    +1,804
    This may be of interest to this thread :)

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

    According to that I can hit 114db SPL at my listening position.

    (90db Sensitivity speakers, 170 watts per channel, 9 ft Distance, 3 Speakers and near corner placement).

    Whether it's accurate or not I don't know! LOL
     
  15. alexs2

    alexs2
    Well-known Member

    Joined:
    Nov 13, 2002
    Messages:
    13,957
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Ratings:
    +1,671
    Gordon...I think that quality of sound often has little to do with voulme,as those with SET amps will no doubt confirm,as the sound quality within the obvious limits in terms of power that most SET's have,can be absolutely stunning.

    The important thing is driving the amp/speaker combination within it's limits,to get the best possible quality out of both.....there's little point having a very good low powered amp struggling with B&W 800's or hanging an FPB750 on the end of a set of horns.

    It is interesting how often though the 100dB level can be peaked at,even when the overall SPL is quite low most of the time.
     
  16. nathan_silly

    nathan_silly
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Thanks for that link Jase. For my front three (90dB @1W) and 100W poweramps at six feet I get 112.5dB.

    Although these speakers are 4Ohm - so my poweramps are at least 150W into 4Ohm, possibly more..so entering in 150W get's me 114.3

    For my rears (87dB @1W) and 100W poweramps I get 111.3dB. Again these are 6Ohm, so add a bit more power- say 112.3dB

    Oh yeah I do recall actual output is a bit higher- I believe the 8000S (60W) was measured 75W into 8Ohm... so if the similar measurement for the 100W PX's are applied- maybe 120W actual into 8Ohm?
     
  17. alexs2

    alexs2
    Well-known Member

    Joined:
    Nov 13, 2002
    Messages:
    13,957
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Ratings:
    +1,671
    An interesting link,Jase...I can see a contest arising here!....certainly seems about right for my room and speakers etc.
     
  18. S H A D O

    S H A D O
    Well-known Member

    Joined:
    Mar 16, 2001
    Messages:
    10,434
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Location:
    United Kingdom
    Ratings:
    +1,804
    Using that link to calculate my surrounds I get 122db SPL at the listening position. (90db, 170watts, 3ft, 4 speakers near a wall).

    No wonder it sounds loud. LOL
     
  19. alexs2

    alexs2
    Well-known Member

    Joined:
    Nov 13, 2002
    Messages:
    13,957
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Ratings:
    +1,671
    A bit less here...peaking about 110dB,but listening distance is getting on for 15 feet and mains speakers not near a wall.....I suppose I could always move a bit nearer if I felt the need!
     
  20. EvilMudge

    EvilMudge
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Might be calling down a swarm of killer bees on my head, but this is certainly one of those areas where properly executed DSP can help, if only to share the load around a 5.1 setup.
     
  21. nathan_silly

    nathan_silly
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    That calculator is only useful of you have identical speakers & poweramps?

    My front three speaker efficiency/impedance are different to my sides, and also different to the rears.
     
  22. alexs2

    alexs2
    Well-known Member

    Joined:
    Nov 13, 2002
    Messages:
    13,957
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Ratings:
    +1,671
    Yes...because it assumes the same efficiency/power values for all.
     
  23. EvilMudge

    EvilMudge
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Ah but does it assume that the in-room efficiency of the L&R speakers is the same as the centre?:devil:
     
  24. alexs2

    alexs2
    Well-known Member

    Joined:
    Nov 13, 2002
    Messages:
    13,957
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Ratings:
    +1,671
    LOL,'Mudge...I think like all of these things,it's a bit of an informed guess at the end of the day,and it certainly doesn't take into account room furnishings etc which can have a massive effect on sound quality and level.
     
  25. pwiles1968

    pwiles1968
    Active Member

    Joined:
    May 15, 2003
    Messages:
    773
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Leicestershire - England
    Ratings:
    +9
    I think this should read –6dB for each doubling of distance, not for each 1m increase.

    I believe it is only true if the cones that are being driven are identical, in most cases one of the them will be the tweeter which will output comparatively very little SPL compared to the Mid/Bass drivers, therefore in reality Bi-Amping gives no SPL gain,
    This confused me when I took Before and after measurements of my system when I Bi-Amped, I had expected a +3dB lift which I did not get even at 1m. It did give benefits in control, soundstage etc.:clap:
     
  26. kryten

    kryten
    Active Member

    Joined:
    Apr 9, 2003
    Messages:
    557
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Location:
    Swindon
    Ratings:
    +18
    I'm sorry, but I have to disagree with these 'target' figures for sound output.

    Anyone who listens regularly to music at anything over 90dB is going to have _serious_ ear problems within a fairly short space of time. 100dB+ is complete insanity!

    I spent years continually improving my car stereo - always striving for sound quality as opposed to the boy racer 'max power' approach. When I finished, I had well over 6 grands worth of kit and an amp that delivered 2x150w into the main speakers plus 300w into the subwoofer and could easily hit 118dB+ (somewhat helped by the cabin gain of the car).

    I regularly listened at volumes far higher than sensible and now have significant hearing loss (-18dB at 6kHz), tinitus and great difficulty understanding people when there are several conversations going on - none of this is fixable :(

    Dolby reference level is far too loud for me now.

    Aside from this, different amplifiers clip at different points in their range and this is often power supply based. As long as the amp can drive the speakers with around 100w of _clean_ power then that is usually enough at most listening levels.

    You need 10x as much power for you to perceive the sound as being twice as loud (10dB) so to get something twice as loud as your 100w amp, you need to go to 1kW!

    The only other thing I'd say is that power amplifiers are truly the one piece of equipment where the numbers mean absolutely nothing. Take a random sample of 5 amplifiers (lets say a cheapo own make one from dixons, a rotel, an arcam, a chord and a krell reference) - now, give someone the specifications alone and ask them to decide which will sound best - bet they don't even get close!

    The only thing that's worth doing for poweramp choice is deciding a budget and then going and listening to the amps that are in your budget!
     
  27. EvilMudge

    EvilMudge
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Did you biamp with amps of the same gain? Did you then check the calibration or try to measure the maximum output?

    I doubt you'd see the calibration levels change at all, but the maximum output should have gone up.

    Oh and tweeters normally have a much higher acoustic efficiency, due to the small wavelengths, which is why tweeters can get away with being much smaller than mids or bass units. That's also why in active speakers it's common to find the amp driving the HF unit(s) is usually lower powered than the one driving the bass.

    And, if you think about it in terms of harmonic distortion, the lower the fundamental frequency, the more the harmonics of this intrude on the mid-range.
     
  28. alexs2

    alexs2
    Well-known Member

    Joined:
    Nov 13, 2002
    Messages:
    13,957
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Ratings:
    +1,671
    kryten...I get your point about "target" SPL's....certainly if you listened to an average SPL of 100dB or even close to that,you would be at serious risk of hearing damage.
    I think that what most people are aiming for is a system capable of reproducing transients in music with a wide dynamic range without clipping or other forms of distortion...certainly what I want from my system,but like you,I would never want to have it running constantly at that level.

    The other pint about specs on amplifiers counting for little is quite right.....SET's in particular measure very poorly in general,and yet a properly designed one can offer superb sound quality...likewise my example of the Crown DC300A....superb measurements and power delivery,but awful,flat sound.
     
  29. pwiles1968

    pwiles1968
    Active Member

    Joined:
    May 15, 2003
    Messages:
    773
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Leicestershire - England
    Ratings:
    +9
    Yes they have the same gain, I used an SPL meter on a tripod about 4 feet from the RH 603 this was not moved through the experiment, the Bi-Amped system did not give me any more SPL than the system with just the MA6100's the 6100's gave me about 1dB more than the SR5300 but I put this down to a greater ability to drive the speakers at low frequencies.

    I am not saying it was not am improvement, it is the best money I have ever spent, just no increase in SPL.
     
  30. slingshot

    slingshot
    Active Member

    Joined:
    Jan 31, 2001
    Messages:
    1,212
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    48
    Location:
    UK
    Ratings:
    +32
    I'm sorry but this is all fine in theory, but please everyone go home tonight and listen to how really awefull your HiFi/Home Cinema sounds at reference level (hope it's not just mine anyway).

    Get anywhere near reference level and even after doing obvious stuff like blu-tak on pictures etc. I get light bulbs buzzing, hinges ratling, floor boards up stairs wobbling, glass in picture frames ratling, not too mention neighbours banging on the door and windows.

    I'm sure if I had 600 watts instead of the totally inadequate 85 watts I really have my room would still sound bad.

    Sure It's nice to know what your system's capable of, but limiting yourself to only auditioning amps the produce at least X watts of power doesn't sound like a clever idea to me.

    Slingshot
     

Share This Page

Loading...
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice