Denon AVR 5803 and power

George · Oct 3, 2003

Hi Folks,
I have a query which perhaps someone out there may have an answer to.
The power ratings on the 5803 is 170wpc. A review(not as detailed as Gene's but had some other tests) done by Homecinemachoice.com(UK  March 2003), yielded a significantly lower rating. THere does seem to be a tendency to overrate power output which can sometimes be a wee confusing.
Is it accurate at 170Wpc(20Hz to 20KHz @ 8Ohms)? If not then what is the actual measured output.
The reason for the question obviously has to do with speaker loads to be driven and choice of speakers.
Can you bi amp a speaker using the speaker terminals on the Denon and terminals on a power amp connected to the pre-outs on the Denon, if so which would be sent to the HF and which to the LF.
Still waiting with baited breath for the arrival of the 5900. Rumor has it that DVI is on its way out, and is being replaced by something new, fact or fiction?
George.

gene · Oct 3, 2003

Hi George;

Is it possible you were reading their review of the European equivalent to the AVR-4802?

Here is a link to their review of the Euro AVR-5803 or AVC-A1SR that shows it will deliver 189.9 watts into 8 ohms and 308.9watts into 4 ohms. For a receiver, thats impressive, especially one with linear power supply and Class AB amps.

HomeCinemaChoice Review of Denon AVC-A1SR
<table border="0" align="center" width="95%" cellpadding="0" cellspacing="0"><tr><td>Quote </td></tr><tr><td id="QUOTE">
LAB RESULTS

Manufacturer's rated output: 170W (20Hz-20kHz, 8ohms, 0.05% THD)

Measured power output @ 1kHz: 189.9W (8ohms, 0.18% THD); 308.9W (4ohms, 0.18% THD)

Fidelity firewall: 183.3.7W @ 0.04% THD (8ohms)

Distortion @ 50W: 0.0021% (8ohms, 1kHz)

Frequency response: 20Hz-20kHz ±0.14dB
</td></tr></table>

Becareful with Bi-amping speakers using different amplifiers with different gain structures. I recommend following my Set-Up Tip in the AVR-5803 review to calibrate the system before doing so.

Bi-Amping on the AVR-5803

We are also looking forward to reviewing the DVD-5900. If you get yours before we do, please provide commentary when you get a chance.

[edited: added lab results since direct link to the HomeCinemaChoice review could not linked]

PaulF · Oct 3, 2003

George,

In general, the same amp will result in a lower measured power in the US than in other parts of the world. Sometimes the manufacturers will even spec it differently. The US requires that power amps be rated by the IHF method which measured maximum continuous power. Other parts of the world still allow peak power or even music power.

To capture the dynamic power that some of the above pwer ratings include, the US now has a dynamic headroom rating.

Paul

Yamahaluver · Oct 3, 2003

Also the line voltage during tests are critical and most testing places overlook this very important aspect. I know for sure that most Yamaha receivers are sensitive to line voltage and a low line voltage would come with a lower than rated power reading.

gene · Oct 3, 2003

Yamahaluver;

You touched on a subject that is often overlooked and misunderstood by consumers and reviewers that fall prey to the power game.  Years ago S&V magazine faced off the Yamaha RX-V1 and Denon AVR-4800 in the same issue and incorrectly showed the latter (which was cheaper) was more powerful.  The reason being was two different reviewers did the test.  The reviewer of the RX-V1 did NOT hold the line voltage constant, while the reviewer of the AVR-4800 did, thus making the test comparison unfair.  A few months later they noted this in very small print, perhaps b/c Yamaha as well as myself made a fuss about it to them.  However, the damage was done, and the impression was left in S&V readership that the Denon was more powerful.

In reality the correct way, or "real world" way of measuring power would have been NOT to hold the line voltage constant.  However, this yields smaller and less impressive numbers to consumers.  Thus S&V as well as all other audio publications to my knowledge hold the line voltage constant during power tests.  

In our up and coming reviews and measurments of power amps/receivers we will NOT do test like this and explicitely state it.  Unless the consumer has a Variac or regulated AC line, those figures wont be very realistic.  Thus our objective is to test equipment in realistic "real world" settings.  This does not take away from what others are doing, and we do feel S&V is one of the best remaining audio magazines out there.

George · Oct 4, 2003

Hi Gene,
The test you refer to, was the one done in ?April '02.
There was also a test done March '03, where the receiver was pitted against the Marantz SR 9200, the big HK as well as the Yamaha and the Pioneer. That is the one where the tests yeilded lower results.
If you go to their home page www.homecinemachoice.com you will then see a page of reviews from the March issue. Click on the AV A1 SR, and that will show differing results from the previous one.
George.

gene · Oct 4, 2003

George;

I saw that one and I think it was a misprint for the 4ohm rating, especially since they claimed the Denon was the most powerful from what I recall.  They didn't even spec the correct power rating from the manufacturer (Should be 170wpc, they have 130wpc).  You may wish to contact them.

<table border="0" align="center" width="95%" cellpadding="0" cellspacing="0"><tr><td>Quote </td></tr><tr><td id="QUOTE">LAB RESULTS

Manufacturer's stated output: 130W (20Hz-20kHz, 8ohms, 0.07% THD)

Measured power output (front channels, two driven) @ 1kHz: 155W (8ohms, <1% THD) 145W (4ohms, <0.5% THD)

Fidelity firewall (front channels, two driven): 150.W @ <0.1% THD (8ohms)

Distortion at 50W (front channels, two driven): 0.0007% (8ohms, 1kHz)

Frequency response (front channels, two driven): 20Hz-20kHz ±0.07dB

Measured power output (front channels, five driven) @ 1kHz: 120W (8ohms, <0.10% THD); 105W (4ohms, <0.1% THD)

Measured power output (centre channel, five driven) @ 1kHz: 120.W (8ohms, <0.1% THD); 110W (4ohms, <0.3% THD)

Measured power output (rear channels, five driven) @ 1kHz: 130W (8ohms, <0.1% THD); 120W (4ohms, <0.2% THD)

FEATURES

Formats supported: Dolby Digital 5.1 & EX; THX Surround EX; DTS 5.1 ES Matrix & Discrete (with THX post processing); DTS 96:24; Dolby Pro-Logic; Dolby Pro-Logic II; THX Ultra 2 processing; DTS Neo:6

Connections: Inputs - 10 line audio inputs (eight A/V composite & S-video and three tape) & phono input; 11 digital - seven optical, three electrical, RF; three component video (assignable); three x eight-channel audio inputs. Outputs - three AV (composite, S-video, audio); two audio-only; two optical digital; component video; preamp outputs for all channel; inputs for pre-main centre and front effects; RS232C; Denon Digital Link

Features: THX Ultra 2 amplifier rated at 7 x 170W; all channels driven 20Hz - 20kHz, 8ohms, 0.05% THD; Denon Digital link multichannel wideband digital input; Composite/S-video/component video conversion; Dolby Headphone; supplied with touch screen programmable, coded remote control with macro capability & USB port

Dimensions: 434(w) x 216(h) x 486(d) mm

Weight: 29kg

</td></tr></table>
[edited added reference to review from HomeCinemaChoice]

Dan Banquer · Oct 5, 2003

The great bulk of audio power amp power supplies are unregulated. This means that when the AC line voltage goes down the output voltage of the power supply goes down. It also means that when you demand current from the unregulated power supply the output voltage of the power supply goes down.
When a power amp is required to provide a musical peak, both the AC line voltage and the voltage of the power supply goes down. Holding the AC line voltage steady under test assures the tester that what will be found for output power is only at a constant 120 volts from the AC line. This does not happen in the real world.
We also need to look at the transformer rating used in a power supply. Transformer ratings are usually given in VA ratings. (volt /amps) What is not known is how many volts and how many amps is the transformer actually rated for. In actuality the transformer could have a high voltage rating but a low current rating. If that is the case then the transformer will saturate much more easily in a higher current demand. Holding up the AC line voltage at 120 volts can mask this and not holding up the AC line voltage and monitoring the AC line voltage can tell you a lot more, because when that transformer saturates, it will draw even more current from the AC line and from the explanation given above the AC line voltage will drop even further and so will the output voltage of the power supply.
I think at this point you have some idea on why power ratings can be so different on the same unit.I could go into a much longer post on power supplies, but I think I will stop here. The above explanation is just one of the many reasons why I use a fully regulated power supply in my power amps. Hope this helps.

Dan Banquer
R.E Designs

PaulF · Oct 6, 2003

Dan,

That's a very good point. However, many regulated supplies use plain old series regulators that sap away available power and turn it into heat. I remember looking at the specs of some high end monoblocks a couple of months back. In one case, a pair from one manufacturer were using 1600W of power just for being switched on.

Series regulated supplies do hold up under heavy current transients but the problem is they use much of the available power themselves and grossly drop down the efficiency of the amplifier.

We would all be better off with switchmode power supplies in our audio gear. Thankfully Sony and some others have started going that route.

Dan Banquer · Oct 6, 2003

If your have an amp that draws a lot of current at idle than yes you will have a lot of heat on the pass transistors of the power supply. However that is really not the case if the amp has reasonably low idle currents. For all the testing I have done I can definetly tell you the pass transistors on the monoblocks I manufacture have far less heat on them than the output transistors, even at full power testing. ( the voltage goes down across the pass transistor as more current is drawn from the unregulated section of the supply.)
Here's a thought for you: What if power amps that are using hard biased outputs are using them to overcome the output impedance of an unregulated power supply (partially)? If that's the case, which I suspect has some merit, than you could bias the outputs lower with a regulated power supply.
Just a thought.

PaulF · Oct 6, 2003

Dan,

No argument here so far, it seems that you know some about amplifier design. I'm not that concerned about the heat itself (although as an end user I naturally prefer less) but as you state the pass transistor(s) does consume power. At full power the voltage probably goes down as you suggest, but the current has gone up. As it's the power that generates the heat, it would be hard to predict under what conditions an amplifier will produce the highest loss across the regulators.

In general though series regulated supplies will need to lower the rail voltage with a given unregulated section in order to account for the headroom required for the regulator to function. So I can see why manufacturers who are after higher power numbers, and not too concerned with high current transients will go the unregulated route. It's also cheaper.

My main gripe is that is it highly inefficient. We often advise people to look for amplifiers with well designed regulated supplies but we should also point out the cost. As an example I randomly selected an amp, the Pass Labs X600, a 600W monoblock,  it uses 600W at idle and 1800W max. That's highly inefficient, much of this consumption coming from the regulators, hence why even idle power is impacted so much. Again please note I picked Pass Labs at random, there are plenty of similar examples from other well known amplifier manufacturers. It's the sort of power consumption a high power multichannel amp could easily reach also.

These are the sort of designs that deliver performance at all costs, in this case efficiency. My guess is you could probably design the same amp with a switchmode power supply and increase efficincy by 25% or more. You would also halve the size of the box.

Dan Banquer · Oct 6, 2003

Paul: Anything is more efficent than linear design. However the Pass labs amps have really hard biased outputs and generate lots of heat. What I am trying to say here is that amps don't have to have hard biased outputs, and still do an excellemt job of reproducing music.
Switchmode supplies are taking over in sound reinforcement. Less weight and more efficency cetainly are important in that area.The problem of HF noise in these type of supplies may not be important to sound reinforcement but I know it is important for home and studio.
One good thing we need to remember about engineering:
There is no free lunch
d.b.

Yamahaluver · Oct 6, 2003

Just wanted to add that class A amps would always consume high current even at idle and this includes the ones with the sliding bias design. This is due to the inherent nature of class A operation.

However I am willing to make the trade off as class A amps always manage to sound better than their class AB and other counterparts, somehow they are more musical and have excellent transients. The sliding bias designs consume less idle current than the hotter, heavier running full class A amps and so in this case, they seem to be a good compromise.

Guest · Oct 7, 2003

Well Yamahaluver; do you consider 35 watts per channel at idle a lot of power?
                d.b.

PaulF · Oct 7, 2003

Dan,

This is an excerpt from The Absolute Sound's review of the X250.

"The X250 is basically a Class A amplifier, but it shifts to Class AB operation beyond the Class "A" bias point - limiting both the electric bill and room heat."

Well lucky for us it already employs an energy saving sliding bias circuit. Quite frankly, it's hard to find a high end amp that doesn't these days.

However that's not my point. If it was I'd be arguing that class D amps and their derivatives are far more efficient than class A or class AB. But that's another thread.

What I am saying is that while power amps with well regulated power supplies will improve transient response, they do so at a cost.

1. For a given unregulated supply section, the voltage rails must be dropped in order to allow the series regulator(s) to function, thus dropping the continuous power spec.

2. My main point. Old clunky series regulators are highly inefficient. They are essentially acting like an intelligent series resistor between the power supply and the amp. They vary the voltage drop across themselves in order to keep the voltage at the amplifier rails constant.

I am suggesting that people should consider all the options. Many amps such as Jeff Rowland, Halcro, and Linn have started using switchmode designs. Sony's new receivers have also, all but the 9000 due to its power needs. The HF noise is a very solvable problem, these PSs are running somewhere between 60Khz and 200Khz.

Here's an example of what can be achieved with a well implemented switchmode design coupled with a regular class AB amp.

Linn Klimax

Scroll down to page 16, page 8 on the PDF.

Dan Banquer · Oct 7, 2003

Hi Paul;
You have made a number of good and accurate points. I don't think the issue of transient response of a well designed regulated power supply vs. an unregulated power supply is much of an issue. There is one real time advantage of a regulated power supply both linear and switcher. The output impedance is usually measured in milliohms. The output impedance of even a big unregulated power supply is usually measured in ohms.
One of the sacrifices of most switching power supplies is that the more efficent they get the tradeoff will be more switching noise, both conducted and radiated.
Hope this helps.
d.b.

gene · Oct 7, 2003

Paul;

Very interesting brochure.  Sounds like we need to check out some of Linn's latest product offerings.  One concern I do have is they don't spec SNR of the amp, unless I overlooked it.  

I agree there are tradeoffs to regulating power supplies for amplifiers:
1) Added power consumption and heat.
2) Drop out voltage of regulators reduce available rail voltage slightly, dending on design.

However I can tell you so far that Dan's amps run cool.  I have them stacked in a small space with only 6inches above and no backpanel in my rack and they only get warm to the touch.  Grant it, I haven't yet blasted them for long periods of time, but we shall see.

I personally think digital amps are the future.  DSP's have become powerful enough and switching speeds should be high enough to push the noise decades past the audio band.  However, until I hear them and measure their noise and distortion, I will stay with Class AB amps.

As for digital amps, lets also not forget the advantage of having 1 digital interconnect to hook-up a 7CH amp to a processor.  Personally I feel a fully integrated solution such as a Super receiver, if done right, would be optimal for space savings and keeping everything in the digital domain until amplified out to the speakers.

PaulF · Oct 7, 2003

Dan,

I didn't give the output impedance much thought but since you raised it, I partially agree. Under normal conditions where the power supply is acting as a contant voltage source, low output impedance is definitely an advantage as you state. Series regulators fall into this category. However under transient conditions, the load is asking the power supply for lots of current. Under these conditions a power supply acting as a high impedance current source makes sense, hence why a direct path to a huge capacitor bank probably makes sense. I also agree that for most people regulated vs unregulated is not an issue, they would probably not hear the difference, hence why most manufacturers just beef up the unregulated section.

As for radiated emissions Linn went the extreme route and built a case out of two solid pieces of machined aluminum, top and botton, however in their case (no pun intended) they did it as much for heatsinking purposes. Radiated emissions should not be an unsolvable problem, in fact Clint and others have often explored the use of HT PCs, all of which run off switchmode power supplies.

Gene,

Stereophile magazine did a review of this amp a few years back. They included measurements on THD and also the spectral breakdown.

Klimax THD measurements

While I am familiar with Dan from posts on other forums, I did not know he has products of his own. (Dan if you have a web site please post the link.)

It would be good to see more emphasis on both switchmode supplies and class D amplification. You may remember I posted a link to Mueta some time ago. They claim to have a class D amp design with extremely good common mode rejection, perhaps making it an even better fit for pairing with a switchode supply.

Dan Banquer · Oct 7, 2003

Hi Paul;
My web site is at http://www.redesignsaudio.com
I think we will see an increase in switchmode amps and supplies. The technology is getting better and cheaper. If I were you I would keep an eye on Apogeeddx. They are doing some interesting things.
d.b.
P.S. I would hate to do a low noise power amp without a regulated supply.

PaulF · Oct 7, 2003

Dan,

It looks like you've designed some nice stuff. I have to ask, why do you use dual power supplies in a monoblock amp?

Maybe you'll let me audition a pair of your amps just for fun next time I'm up your way.

Paul

Denon AVR 5803 and power

George

Audioholic Intern

gene

Audioholics Master Chief

PaulF

Audioholic

Yamahaluver

Audioholic General

gene

Audioholics Master Chief

George

Audioholic Intern

gene

Audioholics Master Chief

Dan Banquer

Full Audioholic

PaulF

Audioholic

Dan Banquer

Full Audioholic

PaulF

Audioholic

Dan Banquer

Full Audioholic

Yamahaluver

Audioholic General

Guest

Guest

PaulF

Audioholic

Dan Banquer

Full Audioholic

gene

Audioholics Master Chief

PaulF

Audioholic

Dan Banquer

Full Audioholic

PaulF

Audioholic

Latest posts

Sitemap

Sitemap

Site Information

Site Information

HTML