Who wants to measure how much power (voltage) is REALLY needed?

[Ski Bum]

1.5v, times four equals 6v peak, or ~36w.  That was pretty loud, but if I had a few additional beers I think I could have set the reference level even higher!  Cool test, but it seems a bit of a rubber yardstick, as while the 0db max level of digital sources is ensured, my amp's clipping status at post-beer volumes is still questionable.

 

   

 

  

[mark1101]

1 hour ago, tube fanatic said:

Mark, you are not getting it!!!  No one is trying to push the use of flea power tube amps with Klipsch speakers in all situations. How about measuring your power requirements and showing us some numbers so we can understand your particular situation?  You have been critical of Pano's test procedure, but have not backed up your assertions with any science!  This thread is intended to be educational (which is why I would enjoy seeing measurements by the high power SS guys as well as the tube guys).

 

Maynard

So my Klipsch speakers are an MCM Grandeur setup.  How about if I measure just the Klipsch 684 subs next time I fire up?  I have a 900 wpc QSC amp connected to them.  I'm sure I'm only using about 7 watts........sheesh.

 

I have 3350 wpc connected............and not a lot of time for this actually.  I evidently speak a different language.  .........so funny.  Look at the cinema ratings.

 

I think it's advantageous for the low watt guys to do this test.  It can show where you might be pushing it.

 

 

 

[mikebse2a3]

4 hours ago, mark1101 said:

What the thread gets on the edge of is that.............all you need is flea power and all else is waste.  Has not been stated but can be read between the lines.  Not taking a shot at ANYONE..........but again, some new folks are around and we want you to know if you invested in some wattage your still OK with some of us. 

 

You "new folks" watch out for the "Flea Watt believers" they are saving all that wasted power to take over the WORLD...........

 

 

miketn

[alzinski]

7 hours ago, Ski Bum said:

1.5v, times four equals 6v peak, or ~36w.  That was pretty loud, but if I had a few additional beers I think I could have set the reference level even higher!  Cool test, but it seems a bit of a rubber yardstick, as while the 0db max level of digital sources is ensured, my amp's clipping status at post-beer volumes is still questionable.

 

   

 

  

 

If you measured 1.5v on the voltage test then you are looking at 2.25 watts into an 8 ohm load.

 

[alzinski]

4 hours ago, mark1101 said:

So my Klipsch speakers are an MCM Grandeur setup.  How about if I measure just the Klipsch 684 subs next time I fire up? 

 

 

 

 

If you want to participate that would be the way to do it. I didn't measure my sub for the test.

[alzinski]

For the folks that haven't done the test the test tone files can be found on page 1 thanks to John Albright for graciously uploading them for us.

 

I want to recap everyone that has done the test so far. I am posting the highest voltage from either test tone.

 

Derrickdj1 -        2v

 

Tube Fanatic -   1.4v

 

Wirrunna -         1.8v

 

alzinski -            1.8v

 

Skibum -            1.5v

 

 

 

Starting to see a trend here.

 

 

For anyone wanting to know how this will relate to power, especially Klipsch speakers I will try and explain, hopefully others can help me.

 

For the SS guys. Most SS amps are going to be close to an ideal voltage source. Derrick had some good questions I will also try and address later. Fundamentally a perfect voltage source has 0 output impedance and won't lose any power, it all goes into the load, voltage stays the same and depending on load only the current changes. In real life there will always be some output impedance but it's low enough to assume a voltage source. When we are doing these measurement tests with a low output impedance device we want to know how much power one needs. I would go ahead and say it's safe to assume that somewhere in the low frequency range you'll see a dip on an "8" ohm speaker go as low as 4 ohms. If you are using a SS amp and doing the calculations I would just go ahead and do the maths for a 4 ohm load to be on the safe side. Derrick has the highest voltage reading thus far at a whopping 2v, for a 4 ohm load he will need an 8 watt amplifier. Seriously that's it, and that's being generous.

 

 

For the tube guys it gets a bit more complex. Most tube amps use an impedance matching transformer at the output. The speaker load actually reflects from the secondary back to the primary as the load for the output device/devices. It's a safe assumption that most tube amps can handle a 100% load mismatch so seeing a 4 ohm load (at particular low frequencies) when connected to the 8 ohm output tap isn't a big deal. So for the tube amps users no need to worry about the lumpy speaker impedance of Klipsch speakers, just do the calculation for the average impedance which is 8 ohms. So for Derrick and his 2v test tone reading he only needs a 4 watt tube amp.

 

Hopefully I have this somewhat correct.

[alzinski]

On 5/12/2017 at 5:49 PM, derrickdj1 said:

Great discussion so far.  Now on the next question.  Most amps are voltage driven, how do the high current amps affect the picture.  Many of the top SS amps are current driven.  Should we be talking about the 1st watt or just what an amp can deliver at lower wattage?  How does current influence low watt tube amps?

 

 

As you say most if not all amps are voltage drive. They have low output impedance and as the load impedance vairies so does the current. For a voltage source providing 8 volts into 8 ohms you will have 1 amp of current. The speaker will dip down at some frequencies to say 4 ohms this voltage source still puts out 8 volts but now it needs to deliver 8/4= 2 amps of current.

 

I would go ahead and say that the top SS amps are not current driven. A current drive amp or also known as a transconductance amp is a whole different beast. Almost all of the speakers you encounter are designed for a voltage source and not a current source. A current source will have high output impedance, a perfect current source will have infinite output impedance. It will deliver a precise current and depending on the load the voltage will change.

 

How does current influence a low watt tube amp? Your question is unclear to me. Maybe it's worded backwards? How does a low watt tube amp influence current? A low watt tube amp, especially a SET with zero global feedback (around the output transformer) can have  in todays standards a relatively high output impedance, say 3 ohms. This is where the amps starts to somewhat behave like a current source. Say the device is delivering 10 volts at 100Hz and the impedance is 6 ohms at 100Hz you can view it as a potential divider. 6 / (6+3) = .66   and so .66*10 = 6.6v   You will have 6.6v across delivered to the speaker and 3.4v is lost across the output impedance of the amplifier. Now say the impedance is 70 ohms at 2.5kHz, 70 / (70+3) = .95    and so .95*10 = 9.5v   You will have 9.5v delivered to the speaker and only .5v is lost across the output impedance of the device.

 

Remember you can have high and low output impedance amps with either tube or SS.

 

Output impedance effects damping. Some speaker have better mechanical damping than others and so they don't require ultra low output impedance. But the majority of speakers out there assume a low output impedance and it's needed to dampen the back EMF created by the moving mass of the woofer. A lot of older amps had variable damping to cater to different speaker manufacturers. Today's speakers are different.

[Ski Bum]

Some highly subjective qualifiers...

 

My speakers are stand mounted, and I have considerable eq down low to restore what's lost perching them up there (and being away from any corners/walls).  That includes a half octave wide 4-6 db boost at 35 hz, with a steep lpf below that.  It results in being quite the power sucker, particularly when the big bass arrives.

 

Two watts just isn't enough.  Audible, objectionable clipping easily achieved with the flea watters in this setup (they fared much better when the fortes were stuffed into a smaller room).  The big bass hits don't have a chance.  On the plus side, it sounds pretty lovely at low levels, and with less bassy content they can stretch a bit louder.

 

I can bring several amps rated at 50w to their knees before reaching outrageous volumes, evidenced by the strident sound that results from audible clipping.  These are from cheap to respectable (Yamaha receiver to ATI power amp).

 

If I hook a deep breather (Crown or Yamaha pro amps, some of the big NAD amps that can puke out hundreds of watts, at least for a little while), my ears become the limiting factor.  Club-like, pant flapping volume, no sense of stridency or compression, even at the most ridiculous levels.  It's quite exhilarating, actually, but you guys already know this.  

[dtel]

19 hours ago, Ski Bum said:

my amp's clipping status at post-beer volumes is still questionable.

Many things "post-beer" could be questionable, your right, including volumes. 

 

So to me I would rather have my ears say stop, then the speaker releasing the magic smoke because of clipping, post beers of course. 

[Don Richard]

If a person is listening to 1 watt average power, the amplifier will need to put out 20 clean watts to pass peaks undistorted.

 

Don't forget every time the listening distance is doubled SPL falls by 6 dB. ! watt average power into a Khorn is 92 dB SPL 16 feet away.

 

Don't forget to allow for more power if EQ is employed. A 3 dB EQ bump doubles power requirements at that frequency.

[mark1101]

Also would be nice to see the test at say 35hz instead of 120hz, and see the test include a proven undistorted reproduced signal.

[racebum]

my car 22v x4 @ 4ohms to the morel mids/tweets , subwoofer sees about 40v @ 2ohms {this gets pretty loud as it translates to a bit over 1000w} 

 

livingroom {klipsch} cranking i'm seeing about 28v. normal listening is considerably less. i'm not sure if it's the rf82ii / rc62ii combo or what but like others hint at. they do sound a bit better not being driven hard. contrast that to the B&W in my bedroom they are a bit more linear. they require more power to get the same SPL but maintain a similar sound ramping up

 

[garyrc]

2 hours ago, Don Richard said:

If a person is listening to 1 watt average power, the amplifier will need to put out 20 clean watts to pass peaks undistorted.

 

Don't forget every time the listening distance is doubled SPL falls by 6 dB. ! watt average power into a Khorn is 92 dB SPL 16 feet away.

 

Don't forget to allow for more power if EQ is employed. A 3 dB EQ bump doubles power requirements at that frequency.

I believe that PWK found that in an ordinary room, neither outside, nor in an anechoic chamber, doubling the distance caused about a 3 dB decline in SPL.  In his experiment, three doublings reduced the SPL by an average of "8 or 9 dB."   It seems this is true in the critical bass areas, as well, occasionally rising to 10 or 11 dB for three doublings.  His room (I believe) was about 16 x 24 feet, ceiling height not reported.  Most of this is in a humorous article criticizing reverberant field speakers, called "THE GREAT MAJOR BREAK THROUGH No. 29," but the curves are for real. 

 

[Endo]

On 5/15/2017 at 3:35 PM, tube fanatic said:

... I would enjoy seeing measurements by the high power SS guys as well as the tube guys).

 

Maynard

Maynard, my numbers may be of limited use--as the display on my digital multimeter will not register less than one-tenth of a volt (eg. 0.0, 0.1, 0.2, etc... ).

 

A few days ago, when I first tried the test--I thought the meter was incapable of the lower measurements; because after many minutes I couldn't get it to read anything other than '00.0'. I tried it again today and turned the volume up beyond normal listening levels... To my surprise, the meter began bouncing between 0.0 and 0.1, sometimes 0.2, with infrequent peaks to 0.3-0.4.

 

Without touching the volume control, and with an EXTECH sound level meter--from my listening chair, 14 feet from the speakers--the music levels hovered in the 72-80 dB range, with peaks to 84-85 [This is louder than my normal, maximum levels by about 5 dB, but necessary to get readings for this "trial"].

 

So, with the numbers I'm seeing--that translates to... ?

 

•Amp = YAMAHA SS, 70W+70W @ 8 Ohms

•One pair '83 Cornwalls, with B3 crossovers

 

 

Addendum: I had my meter set to low range AC; but I'm thinking I still did something wrong... those numbers can't be right?

 

Edited May 20, 2017 by Endo
Addendum

[tube fanatic]

Thanks for taking the time to post Rick (I'm very disappointed at the apparent lack of interest here- the DIY thread has run continuously for 5 YEARS and has generated 122 pages of responses!).

 

Well, on paper at least, only 20 MILLIwatts/channel into your CWs should give you around 74 db @ 14 feet.  So, 0.4V on the musical peaks isn't out of reason (keep in mind, however, that most meters won't read a transient shorter than about 200 milliseconds, so the actual voltage peaks are probably higher).  What we're really looking for is the voltage you measure with the test tones.  So, if you have time, crank the volume control back to where you are getting 85 db spl peaks at the listening position and, using the same source device, play the test tones (without touching the volume control) and record the voltages you obtain.  From that we can see how much power you would need under a worst case situation.

 

Maynard

[WMcD]

I did some back of the envelope calculations.  It looks like the power requirement is based on a 15 dB crest factor and an 8 ohm speaker load. 

 

That is pretty close to classic assumption of a 17 dB crest factor (which PWK may have described).

 

Noteworthy is that we’re requested to use our recording having very high dynamics.  It is part of the measuring stick in the calculations.  An inherent assumption is that the test recording does actually have about 15 dB crest factor and any other source of music we play is comparable, or lower.

 

Overall, a very insightful approach to testing without expensive equipment. 

 

 

[tube fanatic]

Gil, Pano's test is based on 12 db of headroom out of the source component.  So, what did your measurements show you, and what kind of level did you achieve with that particular volume control setting?

 

I ran the test for the guy down the road.  He's running a new (to him) Audio Research D-76 into his CWs, and his typical listening level is in the 105-110 db range.  With the volume control of his preamp set to his limit of tolerance the test showed that he needs just under 36 wpc. So, with that in mind, he's comforted that the approx. 75 wpc capability of his amp is just fine.

 

Maynard 

[Zen Traveler]

I am a multichannel guy and definitely want to try this test.....First I need to pick up a volt meter.  I gather I can pick up one at Lowes?

[Endo]

On 5/21/2017 at 6:06 AM, tube fanatic said:

we're really looking for the voltage you measure with the test tones

Maynard,

ran trial again with test tones:

•120 Hz produced 96 dB at 12 feet, 2.8 volts

•220 Hz produced 88 dB at 12 feet, 0.9 volts

 

If these number look wrong--they may be. I could have done something wrong--and also, my multi-meter is a cheap one (though my sound level meter is not).

 

Amp = YAMAHA SS, 70W+70W @ 8 Ohms; one pair '83 Cornwalls, with B3 crossovers

 

Again, this is perhaps 5-10 dB louder than I normally listen.

[alzinski]

Do you have the EQ set flat? Or do you have loudness setting on?