How do I know if my receiver has enough power not to "clip"? - Page 3 - Home Theater Forum and Systems - HomeTheaterShack.com

Old 03-05-13, 11:32 PM
Senior Shackster

Join Date: Dec 2012
Location: Lexington, VA, USA
Posts: 155
Quote:
AudiocRaver wrote: View Post
There are a lot of factors, here is a one way to approach it:

Per channel:
[*]Start the calculation at 0 dB.[*]The TSi300s have an efficiency rating of 90 dB, I assume with 1 W RMS at 1 m, that is the "standard." For Dolby reference level (85dB), subtract 5 dB (85 dB - 90 dB = -5 dB).[*]For headroom (highest instantaneous peaks), the Dolby standard says add 20 dB.[*]For home theater seating at 10 feet (3 m), (distance x 3) add 4.8 dB (6 dB theoretical for each doubling of distance in open space, but in a typical HT/listening room it is closer to 3 dB).

This example adds up to (0 - 5 + 20 + 4.8) = 19.8 dB relative to the 1 W reference point for the TSi300s, or 95 W available amplifier power needed (19.8 dB is 95x for power). It would allow you to run at Dolby reference level and experience the full dynamic range of Blu-rays & DVDs.

Fudge factors:
[*]If your individual preference was to run your movies a bit louder then the Dolby reference level once in awhile, cranking it up 6 dB, which is roughly 1.5x louder, that quadruples the peak power you would need, but the speakers are only rated to 150 W peak, only 2dB higher than 95 W, so they couold not handle that amount of power.[*]If for some reason the room is REALLY DEAD, lots of sound absorption (you hear no echo or reverb at all with a good hand clap), then the 4.8 dB for distance above becomes 9.6 dB instead of 4.8 dB, another 3x the power. Not likely, though, and, as with the last calculation, your speakers wouldn't be able to handle it.
This is a great mathematical model. Thank you for posting it.

One portion I found especially interesting:
"For headroom (highest instantaneous peaks), the Dolby standard says add 20 dB."

Manufacturers are always gaming their specs. Following the FTC guidelines is the best way to compare amps. But when discussing headroom and instantaneous peaks, dynamic power would be a better way to compare amps, if there was a standardized way of measuring the ability of an amp to produce these instantaneous peaks.

This discussion has been going on in audiophile circles for at least 40 years: what is the best way to measure amplifier power that relates to listening to music? Are tightly regulated power supplies desirable, or are ones that allow more dynamic power? I think the FTC guidelines have their shortcomings in regards to being able to measure an amps ability to reproduce dynamic sounds. Yet these same guidelines are a huge improvement over the frequently fictitious peak power vs RMS ratings that used to occur.

I am less impressed with the need to run all 5 or 7 channels at full power at 20 - 20 kHz. There is ample testing to indicate that this never happens in playback of music or video and chasing this spec just adds needlessly to manufacturing costs with no real world benefit.

I agree with the comment that if one needs a large increase in power, that separates are the way to go. It gets to be awfully tough to cram 7 channels capable of 250 watts plus into an A/V chassis. But one should keep in mind that going from 125 to 250 is only a 3 db jump and that is pretty small at what is likely to be a huge jump in cost. A better investment would be into more efficient speakers or pass off some of the load into two powered subwoofers.

http://www.audiokarma.org/forums/arc.../t-329286.html
TheHammer is offline

Old 07-18-13, 01:55 AM
Shackster

Join Date: Feb 2011
Location: SA, Australia
Posts: 18
Re: How do I know if my receiver has enough power not to "clip"?

Thanks to AudiocRaver for his mathematical model - it’s the first time I’ve seen something like that. I was able to understand the model after a little head scratching (I’m a mechanical engineer, not an electrical engineer – I wasn’t taught this stuff). After some more reading and head scratching, I created the attached spreadsheet which might be of interest to others . I also made a couple of tweaks, one of which is to account for the number of speakers. I found the model helped to put things into perspective, and get a feel for which values make a big difference.

I think most people who’ve contributed to this thread know the following already, but for those less familiar:
• Placing a speaker near a wall or corner will change its sensitivity. This should depend on the port arrangement and the distance to the wall(s), and should be borne in mind. This effect is also frequency dependent.
• As I understand it (again, I’m not an EE), these calculations are based on speaker sensitivity at a fixed impedance. Impedance will vary at different frequencies, so this isn’t 100% accurate, but it gives you a decent start.
• There’s been some discussion in this thread about manufacturers being a bit cheeky with their specs. I’ve seen some speaker reviews include sensitivity testing, so this could help in that regard. In terms of amps, there are a few things to be careful of (as discussed in this thread). One of the better examples on this front is that Emotiva go so far as to publish THD plots and a range of other data for their amps. It might pay to look around a little.
• In the spreadsheet, the values in the darker blue are the main variables. The lighter blue values are also variables, but perhaps better left alone in most cases.
• I took some measurements with my SPL meter to get a feel for listening volume. I found 65dB to be a quiet background volume – the kind of volume I’d have when watching the news. 75dB was loud but sustainably so – I could comfortably listen for quite some time at that volume. 85dB was “cranked” – I wouldn’t be comfortable listening to more than a single song when it’s that loud (but I’d sure enjoy it while it lasted). These values might help to provide guidelines for some, when working to a listening volume.
Attached Files Amp Power Calc.xls (31.0 KB)
Wolfgang is offline
Old 07-18-13, 03:55 AM
Shackster

Join Date: Jun 2013
Location: Romania, too far away
Posts: 32
Re: How do I know if my receiver has enough power not to "clip"?

goodear said '' loud enough''. This is a very important point to start. There are fellows who love 120db as in rock live concerts. I would also point out that the tweeters are mostly blew out by underpowered amps. In the pro area, the recommended amp power is from 1,5 to 2,5 of the speakers power. I am using a preamp + individual power amp/channel. I am using an old Aphex Studio dominator as a limiter and I never blew out any speaker.
tba
tba is offline
Old 07-18-13, 07:44 AM
Senior Shackster

Join Date: Sep 2012
Posts: 725
My System
Re: How do I know if my receiver has enough power not to "clip"?

Quote:
goodears wrote: View Post
How do I know if my receiver has adequate power for 7.1 HT? So I can play it loud enough to get good accuracy and clarity. I don't mean ear bleed but loud enough but at the same time not clip and blow up a speaker tweeter. What power rating should I look for? Is this why people buy amps to go along with their receiver (so they don't blow something up)?
I'm purposely leaving the Dolby specs out because very few people listen to their Hometheater at Dolby reference levels because its plainly just too loud. Things to consider are, the size of the room, distance of the speakers relative to the listener position, and what SPL you are truly going for.

With the exception of NAD who advertises full bandwidth x all channels driven, few AVR manufacturers list full bandwidth at all channels driven. Why? Its not that relevant. There are very few DVD/BluRay that contain equal volume across all channels. I own over 300DVD/BluRay and there is not one in my collection that I can point too that contains equal volume across all channels. There are exceptions but are rare as hens teeth. Couple that with the fact that any bass material is routed to the sub when setting speakers to small. I'm more interested in the 2 channel test results because most of the information in a movie soundtrack is contained across the main left/right and center channels. Most AVR specify their out power in 2 channel driven using full bandwidth.
Old 07-18-13, 09:11 AM
Elite Shackster

Tony

Join Date: Sep 2007
Location: Edmonton, Alberta, Can
Posts: 14,816
My System
Re: How do I know if my receiver has enough power not to "clip"?

Quote:
3dbinCanada wrote: View Post
few AVR manufacturers list full bandwidth at all channels driven. Why? Its not that relevant. There are very few DVD/BluRay that contain equal volume across all channels.
Actually its not required so they choose not to do it. Many movies have full levels sent to all channels but it usually only lasts a few seconds (enough to drain any reserves that the amps may have), I have a full 8 channel meter array hooked up and have confirmed this with many action movies.

Quote:
I'm more interested in the 2 channel test results because most of the information in a movie soundtrack is contained across the main left/right and center channels. Most AVR specify their out power in 2 channel driven using full bandwidth.
They are only required to test using a 1kHz test tone not full bandwidth Some do choose to post their full bandwith 2 channel readings but most bench tests I have seen (and thats many) do not come very close at all to what they state. They also only test at a distance of 1 meter (who sits that close to their speakers).

Home theater:
Onkyo 805, Yamaha YDP2006EQ, Samson Servo 600 amp
3 EV Sentry 500 monitors across the front, 4 Mission 762i's Surrounds, SVS PB13U sub, Panasonic BDT220, Harmony 1100, Nintendo WiiU
Panasonic PT-AE8000 on a 120" 2,35:1 fixed screen

Living room system:
Sherwood/Newcastle R972, Mission 765's, SVS SBS02's, A/D/S MS3u sub, Yamaha YDG2030EQ
Yamaha KX-393 Tape deck, CDC 805 CD changer, Panasonic BD60, Sony turntable PS-T20
Panasonic TC-P50ST60, HD-PVR & WDTV Live, Harmony 900

tonyvdb is offline
Old 07-18-13, 09:48 AM
Senior Shackster

Join Date: Sep 2012
Posts: 725
My System
Re: How do I know if my receiver has enough power not to "clip"?

Quote:
tonyvdb wrote: View Post
Actually its not required so they choose not to do it. Many movies have full levels sent to all channels but it usually only lasts a few seconds (enough to drain any reserves that the amps may have), I have a full 8 channel meter array hooked up and have confirmed this with many action movies.
I would agree with your assessment if and only if the AVR is expected to power all channels full range using full range speakers. However, this usually NOT the case as most typical set ups involve setting all the surrounds to small which channels most of the bass energy to a sub. Since the vastly predominant setup in HT involves using a sub and steering bass away from all the channels, an AVR will easily handle the few seconds of all channels firing. My RX-V1800 can drive my PSB Image suite of speakers well into ear bleeding territory without strain or fuss.

Quote:
tonyvdb wrote: View Post
They are only required to test using a 1kHz test tone not full bandwidth Some do choose to post their full bandwith 2 channel readings but most bench tests I have seen (and thats many) do not come very close at all to what they state. They also only test at a distance of 1 meter (who sits that close to their speakers).
That's why I read only those audio mags that provide test results and leave the subjective "fluff" reviews alone. Most AVR manufactures now provide a full bandwidth test into two channels. Check out their sites.
Old 07-20-13, 07:24 AM
HTS Moderator
Reviewer

Wayne Myers

Join Date: Jun 2012
Location: Lincoln, NE
Posts: 4,609
Re: How do I know if my receiver has enough power not to "clip"?

Quote:
Wolfgang wrote: View Post
Thanks to AudiocRaver for his mathematical model - it’s the first time I’ve seen something like that. I was able to understand the model after a little head scratching (I’m a mechanical engineer, not an electrical engineer – I wasn’t taught this stuff). After some more reading and head scratching, I created the attached spreadsheet which might be of interest to others . I also made a couple of tweaks, one of which is to account for the number of speakers. I found the model helped to put things into perspective, and get a feel for which values make a big difference.

I think most people who’ve contributed to this thread know the following already, but for those less familiar:
• Placing a speaker near a wall or corner will change its sensitivity. This should depend on the port arrangement and the distance to the wall(s), and should be borne in mind. This effect is also frequency dependent.
• As I understand it (again, I’m not an EE), these calculations are based on speaker sensitivity at a fixed impedance. Impedance will vary at different frequencies, so this isn’t 100% accurate, but it gives you a decent start.
• There’s been some discussion in this thread about manufacturers being a bit cheeky with their specs. I’ve seen some speaker reviews include sensitivity testing, so this could help in that regard. In terms of amps, there are a few things to be careful of (as discussed in this thread). One of the better examples on this front is that Emotiva go so far as to publish THD plots and a range of other data for their amps. It might pay to look around a little.
• In the spreadsheet, the values in the darker blue are the main variables. The lighter blue values are also variables, but perhaps better left alone in most cases.
• I took some measurements with my SPL meter to get a feel for listening volume. I found 65dB to be a quiet background volume – the kind of volume I’d have when watching the news. 75dB was loud but sustainably so – I could comfortably listen for quite some time at that volume. 85dB was “cranked” – I wouldn’t be comfortable listening to more than a single song when it’s that loud (but I’d sure enjoy it while it lasted). These values might help to provide guidelines for some, when working to a listening volume.
Wolfgang:

Nice job on the spreadsheet. That is a handy tool.

Room gain: you assumed 6 db loss per distance doubling free field, 3 dB in a room. That means 3 db of "room gain," It will vary somewhat depending on the room. It can be measured and put into the formula as a variable, with 3 db as default if the user does not want to do the measurement. Best way to measure is wide-band pink noise, I think. There is a thread discussing it right now over here. Anyway you could say Play Pink Noise through one speaker; measure SPL at 1 m, measure SPL at 2 m, then Room Gain = (6 dB - (MEAS_at_1M - MEAS_at_2M)).

It is a little hard to know what to do about number of speakers (mono=1, stereo=2, home theater=3?, 5?, 7?), safety factor, etc...

Anyway, nice work.
AudiocRaver is offline
Old 07-21-13, 06:14 AM
Shackster

Join Date: Feb 2011
Location: SA, Australia
Posts: 18
Re: How do I know if my receiver has enough power not to "clip"?

Thanks AudiocRaver.

You're right about room gain of course. I'd deliberately left it as a variable for that reason, but I just didn't know how to work it out any more accurately than that. I now see your comment/formula on room gain, and this makes a lot of sense to me. I see in the thread you linked, there are effectively questions over how room gain is defined in terms of what frequency it's measured at. Pink noise seems a sensible practical concession - there'll still be a gain variation with frequency, but what else do you do?

I presumed the number of speakers would only be complicated if you have 5+ speakers. 1, 2, 3 (although not included in the spreadsheet) & 4 speakers seem relatively straight-forward, according to the reference I put in the relevant xls comments. When you get more than that, it seems there are some complicated cancellation/reinforcement actions taking place. For various reasons, I'm practically limited to a 2.1 or 2.2 system, so my care factor for higher numbers of speakers is somewhat diminished...
Wolfgang is offline
Old 07-21-13, 12:28 PM
Shackster

Join Date: Jul 2013
Location: Kansas City, Mo
Posts: 7
Re: How do I know if my receiver has enough power not to "clip"?

I think the best way to ensure your receiver will not clip when processing 7.1 is to get one with line level output jacks for the front channels (at least). As already mentioned using an external sub unloads the AVR from large amounts of LFE power handed off to the external amp. Getting an AVR with external connections, allows for connection of an external amplifier and "unloading" even more of a burden from the AVR. The front mains (almost always) have the highest power demanded by the source material, and will likely be the first to "clip".

If the calculator/spreadsheet is used to properly size the wattage of the receiver this will likely provide adequate power and clipping will be negligible. But if it is noticeable at some point in the future (or present), the jacks are an easy way to upgrade the system.

Also, the Dolby recommendation for 20 dB of headroom results in 100x the wattage.
CMiller747 is offline
Old 07-21-13, 12:48 PM
HTS Moderator
Reviewer

Wayne Myers

Join Date: Jun 2012
Location: Lincoln, NE
Posts: 4,609
Re: How do I know if my receiver has enough power not to "clip"?

Quote:
Wolfgang wrote: View Post
Thanks AudiocRaver.

You're right about room gain of course. I'd deliberately left it as a variable for that reason, but I just didn't know how to work it out any more accurately than that. I now see your comment/formula on room gain, and this makes a lot of sense to me. I see in the thread you linked, there are effectively questions over how room gain is defined in terms of what frequency it's measured at. Pink noise seems a sensible practical concession - there'll still be a gain variation with frequency, but what else do you do?

I presumed the number of speakers would only be complicated if you have 5+ speakers. 1, 2, 3 (although not included in the spreadsheet) & 4 speakers seem relatively straight-forward, according to the reference I put in the relevant xls comments. When you get more than that, it seems there are some complicated cancellation/reinforcement actions taking place. For various reasons, I'm practically limited to a 2.1 or 2.2 system, so my care factor for higher numbers of speakers is somewhat diminished...
Yes, there are a lot of variables one could play with. I like what you have done, it's a very nice tool.
AudiocRaver is offline

 Bookmarks

 Tags clip , power , receiver

Message:
Options

## Register Now

Random Question
Random Question #2

User Name:
OR

## Log-in

 User Name Password Remember Me?
Human Verification

In order to verify that you are a human and not a spam bot, please enter the answer into the following box below based on the instructions contained in the graphic.