Power into M80s from my cheap Denon

Axiom Home Page

Yesterday I mentioned that I would attempt to find out how much power I can get from my cheap Denon receiver into a pair of M80s.

A few words about my receiver: it's the Denon AVR-2105 rated at 90W/channel into 8 Ohms with 0.08% distortion and 125W/channel into 6 Ohms with 0.7% distortion. Dynamic power is 200W x 2 channels into 2 Ohms. The receiver cost me $350 a couple of years ago brand new.

So a few words about what I did: I wanted to use the worst possible load the M80s could present to the amp. As shown on the graphs below, this load presents itself at about 50 Hz and is 4 Ohms. At this frequency, the speaker has a resistance of 2.83 Ohms and an inductive reactance of 2.83 Ohms due to the 45 degree phase angle (R=4*cos45, XL=4*sin45). So I injected a 50Hz test tone from the Realtraps CD (thanks again Randy

) and measured the RMS voltage across the speaker terminals. I had both M80s going but I only measured one of them. The real power is ((V^2)/4)*cos(45), the reactive power is the same formula but substitute sin instead of cos. The complex power is the square root of the sum of the squares of the real and reactive.

The data below (see blue column) shows that I could source as much as 306 Volt-Amps (Watts) from the Denon at maximum volume. As I indicate on my notes on the graph, I only applied the test tone for about 5 seconds (long enough to get a reading) because frankly I was scared I might do some damage

.

I did not do any distortion measurements as I don't have any equipment to do that. I can tell you though that up to +3 on my main gain knob, the 50Hz tone sounded "good" to me. After +3, I had to stop and put in earplugs. But I watched the M80 drivers a few times after +3 and they looked to be travelling linearly to me.

This tells me a few things:

1. Denon amps deliver! If my bottom-of-the-line unit can do this, I am sure the rest of them can do more.

2. I would certainly not hesitate to run this unit continously at the +3 level. As I've mentioned before, I have left it turned on continously for an hour at maximum volume with music while I left the house with no shut-down (come to think of it now maybe it did shut down and then re-start but it was certainly playing when I got home).

3. This was a test tone at the worst possible frequency. For real music and with a sub, the receiver would never experience these kinds of torturous conditions.

4. I don't know how anyone, in a room that is 4,000 cubic feet or less can possibly listen to any levels approaching 0 on the gain knob. I've listened to some classical at +7 but absolutely no more.

5. If this is the capability of my cheap Denon, what does that say about Odyssey, Emotiva and other more "esoteric" brands? Be careful that you don't fry your M80s!!!

I hope this is helpful to anyone out there looking for an amp. I am sure that the discussions about power will rage on but at least this is some quantitative data that we can draw on.

Most interesting results under, as you say, extreme conditions that would never occur in ordinary listening(but what was the classical item you listened to at +7, which would be horrendously loud unless the passage in question was very low level itself). In particular, five seconds of 50Hz drawing 216 watts aren't going to happen except on an experimental basis.

i'm newbie but very interesting to know the test you did. Is that mean Denon does a good job? Or M80? coz i'd looked for a NHT absolute Zero (6ohms) for my denon(8ohms), but finally i back to Axiom M3(8ohms). And now im still confusing about the risk of a sound system which seting up with compenents (receiver/speakers)come with diff ohms.

Mark

Well Mark, the test was mainly of the 2105's capabilities. As I commented, especially at that +15 mark, there's nothing remotely like that in real-world home listening to music or movies, so it speaks highly of the capabilities of the 2105(and I suspect that there're other receivers in that price class that would also do well).

JohnK,

My apologies. I didn't mean to write +7 but rather -7. Good catch

Just goes to show you don't have to spend alot to get a good amp. If memory serves, the 2105 only had a about 5 receivers above it and numerous ones below in the Denon line up for 2005. Even though it didn't cost alot I would still consider it an upper middle receiver, certainly not cheap, just inexpensive.

Hi Mojo,

Excellent testing and I'd agree with you that the Denons, even the inexpensive models and the Sherwood Newcastles will drive the 4-ohm M80s without problems in average-sized rooms.

However--and it's a big "however"--you would really have to put an oscilloscope on the output and monitor the audio waveform to see when the Denon approached or went into clipping. I suspect that may be happening because our detection threshold for distortion with music and soundtrack material has to reach quite gross levels--from 1% to 3%--before we notice a grotty, edgy or nasty edge to the sound quality.

Psycho-acoustic studies have demonstrated this repeatedly. I'll be writing about this subject in an upcoming newsletter article, but what's surprising are the momentary peak power requirements in some music that you'd think wouldn't be demanding.

It's a matter of acoustical "ease" at loud listening levels, even with smaller loudspeakers when they're fed unlimited amounts of power with no clipping. I was testing some Axiom W2 in-wall/on-wall hybrid speakers driven by two older Denon Class A/B monoblocs rated at 250 watts per channel into 8 ohms with no subwoofer running--the tests were double-blind so I had no idea what I was listening to until after the test. I couldn't believe how clean and loud the sound was and how much good bass there was from the two W2's unassisted by a subwoofer.

Later, the tests were repeated again with several different and excellent subwoofers mated to them and the same Denon monoblocs driving the W2s. I wrote down comments like "excellent"; "really transparent and clean"; "very natural and uncolored", "wonderful deep and powerful bass", and "these are great speakers--what am I listening to? M80s, M60s?"

I was amazed that the satellites in these tests were W2s and I attribute my comments to the large amounts of unclipped power (and of course the inherent quality of the W2s).

Regards,

Thanks for posting the data, though I'm not sure you can conclude much without more testing. For example,though you couldn't measure it, the likelhood is you are clipping that amp and causing compression and unwanted harmonics from the 50hz signal. In other words the SPL from the woofer is not increasing as gain exceeds rated output of the amp. If you were running a mixed frequency signal you would hear the distortion from the HF drivers as the dynamic range compressed.

Clipping and damaged voicecoils go hand in hand so be careful.

Thanks Alan for your comments supporting the use of monoblocks with more power and your results. I know a lot of math and engineer gurus around really push the whole 1 watt mumbo jumbo. However in real situations, as I believe Ian has pointed out, the more power/headroom/etc. the better results.

Yes my Denon does a good job, but it does not compare to my new Odyssey mono's to me in overall sound quality from lower volumes to higher volumes. Now I can easily tell when my Denon is starting to strain, my Odyssey's keep on going. Also, at louder SPL's there is no ear fatique as a result of no clipping, distortion, etc.

I'm a firm believer in seperates now. I wonder if a lot of people who are on the AVR bandwagon have actually experienced a true high-end amplifier.

Quote:

I wonder if a lot of people who are on the AVR bandwagon have actually experienced a true high-end amplifier.

Not to beat a long dead horse, but I wonder how many people on the separates bandwagon have actually experienced a true double-blind listening test between separates and AVRs.

Dead horse, nothing. That thing's a zombie.

Hi Randy... a few weeks ago you said:

Quote:

Good point Rick, and one I've been wanting to ask as well. Seems many like to reference articles and base everything on math, etc., but have they actually been involved in real world test themselves.

I'm looking at a Russound AB-2 switcher so I can jump back/forth between my Denon and Amps. I plan to invite some friends over for some tests. I'm not even going to tell them what they are testing.

How bout putting those Odyssey's to the test?

Peter and Ken,

Just curious, I believe both of you had heard Dennis's Halo system in his large room and high spl volumes. Do you actually think a cheap Denon could provide the same performance?

I plan to figure out a blind test scenario at some point, but to my ears I can easily hear a difference on most songs when using my Denon by itself versus throwing the Odyssey's into the equation.

No, I don't. Where I have my doubts is at low(er) volumes, say up to around 80dB. Shutdowns are patently, provably the case with higher volumes, as both you and he showed. Presumably distortion occurs before those shutdowns.

What I also know (from my own experience) is that audio memory is remarkably short--on the order of seconds. Instant switching is the only way that one can trust one's ears--and brain.

When deciding to buy my RV-8, I auditioned one driving speakers with it's internal amplifier, and then through an LX-7 amplifier with the RV-8 feeding it as a preamp. Although there was some time for "switching" involved (it wasn't an easy A/B switch), I couldn't hear any difference, so I figured this AVR was good enough to my ears.

My AVR is rated at 140W per channel into 8 Ohms fully driven, and the LX-7 was 200W per channel I think. So I'm sure I would have heard a difference once the RV-8 was pushed current-wise past what it could deliver, where the more powerful separate amp would give a little more. I guess we didn't get it that loud in our testing, but it was plenty loud enough.

I've found its difficult to tell the difference at low volumes where there is no clipping or compression. At moderate volumes with music containing a mix of quiet and loud passages you can hear the onset of compression and distortion. It manifests itself as loose one note bass, smeared drums and shrill, unclear music. Switching to a better amp, usually a more powerful amp, means you can play the speakers much more loudly without the nasty artifacts you get from a clipping amp trying to reproduce quick musical transients.

There is a great jazz cover of "Walking on the Moon" by The Yuri Honing Trio that has excellent drum and sax attack and quiet plucking that make it ideal for testing your system. Bass, drums and sax go from barely audible to not hearing yourself speak and back again several times. You'll know if your amp clips or how well your speakers perform with this one.

When my brother visits he brings his favourite CDs and DVDs and keeps raising the volume to much higher levels than I am used to and complains my system doesn't sound loud enough.

He is used to hearing much more distortion from clipping on his system at lower volumes.

Quote:

Dead horse, nothing. That thing's a zombie.

lol.

I agree with Alan's and Jakeman's posts on distortion. As I mentioned, I didn't have the necessary equipment to conduct distortion measurements. This was a "poor engineer's" characterization

. One day, I'll borrow a scope to satisfy my curiousity.

It's worth noting, again

, that I keep my Denon at -25 for music and around -20 to -15 for movies. This translates to about a half watt to 2 watts (measured under a different experiment). So when listening to movies, this gives me about 22dB of headroom and music 16dB of headroom (referenced to 90W). This is more than enough headroom. Now admittedly I don't know for sure if my receiver is clipping at 90W but I certainly don't hear any clipping. BTW, these power levels translate to 80 to 90dB SPL in my room.

Another item worth noting is the effects that high power have on your listening experience. Higher power to your speakers translates to more heat that must be dissipated. Speakers are not heatsinks and they are not terribly efficient at cooling themselves. The sinister aspect to this is that as the voicecoil temperature rises, its resistance increases. As its resistance increases, the power to the speaker goes down. As power to the speaker goes down, SPL decreases. So now a "portion" of your high powered amplifier is acting as a heater

. The question of course is to what degree? If you are putting 300W into a speaker, is 150W going into heating? I'd need to work the math out on this one.

And now something even more sinister. Ever wonder what happens to the speaker's passive cross-over design with these heating effects? Very simply, the cross-over frequencies will shift with volume.

And there are also deleterious effects on the magnetic circuit...I won't go there

.

Most of us need not concern ourselves with these worries. But there's a few out there that may want to take note. Those that want more power may not only need a larger amp. They may also need larger speakers. And perhaps even active cross-overs. And ear plugs

Quote:

Another item worth noting is the effects that high power have on your listening experience. Higher power to your speakers translates to more heat that must be dissipated. Speakers are not heatsinks and they are not terribly efficient at cooling themselves. The sinister aspect to this is that as the voicecoil temperature rises, its resistance increases. As its resistance increases, the power to the speaker goes down. As power to the speaker goes down, SPL decreases. So now a "portion" of your high powered amplifier is acting as a heater . The question of course is to what degree? If you are putting 300W into a speaker, is 150W going into heating? I'd need to work the math out on this one.

And now something even more sinister. Ever wonder what happens to the speaker's passive cross-over design with these heating effects? Very simply, the cross-over frequencies will shift with volume.

I wonder if Ian knew this when he designed the M80s

Well, I wasn't specifically referring to the M80s but speakers in general. The M80s are rated to 400W. So I am sure they don't experience any of these issues until above that level. 400W is an amazing figure for a speaker, period! Especially a pair that sounds so darned good!

My bad, I just figured this being an Axiom forum and this little tid bit "The question of course is to what degree? If you are putting 300W into a speaker, is 150W going into heating?" The M80s are the only speaker here rated for such power.

The 400W spec is a conservative figure, I believe Alan and Ian have mentioned this in the past. They have been driven with 700w continuous for 24hr/day for 2wks straight with peaks exceeding 1100+ watts in recent tests.

The post started out about my receiver and then I drifted off babbling about speakers. My bad, not yours

.

But for all the powerheads out there, upgrading to M100s may not be a bad idea

To respond to Alan, my buddy came over earlier today with his scope primarily to determine at what power level the Denon starts to clip. So here's what we found out.

Recall that we are injecting a 50Hz tone in order to extract the maximum possible current out of the amp. The M80s have an impedance minimum of 4Ohms at this frequency. By the way, we did confirm the symmetry, period and "cleanliness" of the Realtraps 50Hz tone.

The Denon started square-waving at +9dB which represented a complex power of 225W/channel. However, well prior to this at -2.5dB, which represented 112W/channel, we noticed the onset of slight frequency or slew rate distortion on the positive half of the sine wave. The negative-going portion of this top half lost symmetry by decaying slightly from 90 to 135 degrees and then assumed proper symmetry from 135 to 180 degrees. The effects of this artifact were not audible (at least not to us). We did not see any other distortion artifacts throughout our tests such as cross-over distortion, hysteresis or frequency shifting. Recall that my amp is rated a "clean" 90W into 8Ohms.

We made some other observations:

The left and right channel DC levels were -10mV and -20mV respectively. These levels are practically insignificant in my opinion as I see no obvious "sucking in" of the cone.
At maximum gain with no source, we saw 6mV peak-to-peak, 60Hz noise with some high frequency artifacts on top. Again, this has practically no consequence.
Speaking in a normal voice 2 feet away from the M80 caused a 2mV p-p signal to be picked up at the speaker terminals. When we noticed this, the concern was cross-talk from one speaker to the other. But we didn't dig any more into this and now in retrospect we should have tried to identify how this may vary with sound pressure. The yard and beer beckoned however .
The Radio Shack meter that I used for the initial readings is accurate for sine waves.
I listen to stereo at levels of less than half-a-watt per side.

The conclusions are:

I have headroom of more than 21dB.
The amp exceeds spec at least at 50Hz.
Under realistic music conditions, I would expect the amp to start square-waving at a higher power level
I am happy with my amp.

On another note, does anyone know why the charts are missing in the original post? I still have them on photobucket.

Mo, very interesting results again, which confirm your prior conclusions(the original chart and graphs still appear on my screen). Incidentally, the slight distortion of the waveform noted around 112 watts wouldn't seem to be related to slew rate, since at 50Hz the slew rate required would be extremely low even for much higher power levels.

Quote:

I am happy with my amp.

The most important thing imo

Hi John,

Thanks. The graphs are back on my screen now. Maybe I'm having these problems with the site because I use Opera. The problems have appeared in the last month.

The big difference with my prior conclusions is the power level after +9dB. I had made the assumption (Wid doesn't like it when I assume

) that the waveform remains sinusoidal right up to maximum volume. That assumption has now been proven wrong. Therefore, the voltage and power levels in my chart after +9 should be discounted since the waveform is being clipped.

As for the slew rate, I have to admit that I don't know the technical term for what we witnessed but it looks like the picture below. I'm wondering if the issue is related to a leaky power supply cap on the positive rail. If this was appearing at lower volume levels, I'd have my Denon apart right now but I just don't see this issue as being a practical concern.

Mojo,

Where the hell do you get the time to do this stuff??

Hey Mike,

Audio has been a long repressed passion that has now been re-awakened. Now I have to subdue my passion before it subdues me

Mo, when I said that your previous conclusions were confirmed, I meant your belief that you had more than enough clean power for your use. Clipping as the level went over 200 watts doesn't change that. By the way, a quick slew rate calc(6.28 x frequency x peak voltage)/1,000,000, indicates that at 50Hz and that power level about .01 V/microsecond is adequate. Modern equipment typically has rates of 10V/microsecond or more and has a slew rate far more than adequate for the full audio range.