Behringer A800 Stereo Amplifier Review

[restorer-john]

@amirm Excellent review and really happy to see the expanded measurements.

 

[oog747]

Amir (@amirm), thank you very much for your review.

Any thoughts on running A800 with lower impedance than spec'ed loads? Is it mostly limited in specs to 4 Ohm/8 Ohm bridged due to limited cooling solution or is it parts/design limitation? Behringer seems to be very vocal on "runs almost any speaker", but quite shy in putting that in specs.

In my own current side by side testing of XLS1502 and A800, I am leaning towards A800, and 240 WPC @4 Ohm is more than I expected. Still wondering, if it has potential to be pushed little more, if needed.

 

[bogart]

Very interesting! I’m glad to see this quietly go about its business. Although it isn’t the best measured item we will see this year, it seems to be solid value and quite competent.

Personally, I’m short-listing this as a compact and cool running option to serve a living room system. Curious whether folks would go for this or one of the modest powered NCores like at Audiophonics. I suspect the Behringer is solving for total power on hand, while the NCores would offer (less) better measuring power.

 

[oog747]

... I could do without the notches in the volume control though as they are fake (they are fully analog) and made exact setting harder (it would creep into the notch). Not a big deal though. ...

My $.02 on volume controls on A800. I found myself using them more than expected on power amp for very slight adjustments of balance on sources like older concert DVDs, where I am sometimes finding it bit off as well as vinyl (where it's off more often). Usually those are left untreated for lack of easy control or not wanting to affect the sound more intrusive ways. And those notches help (at least to me) for easier small adjustments and going back. And even between the notches, gap seems to be wide enough to stay quite well in place.
To my advantage, I keep dials at around 2 o'clock position for input sensitivity on RCA at around 1.4V, so it feels more flexible/acceptable to adjust them (and also more nominal for analog potentiometers) than having them at max. With "digital" dials on XLS1502 set at max that doesn't "work" same smooth way.

 

[Rja4000]

In Europe, price is now 169€, shipping included.
That's pretty cheap.

Now Behringer hardly qualifies for 'Pro' label, given a reputation of so-so reliability...
(You may not care at home, especially given the price and length of guarantee period, but when your income and reputation are dependent on your gear's reliability, that's very different.)

But that's not what matters here.
For that price, you can hardly beat it.

 

[JIW]

The A500 I have had had slightly different gains for the two channels. To get similar output (as indicated by the LEDs), I had to set the potentiometer for one channel two notches lower at low levels and one notch lower at high levels.

On the A500 you tested previously, the gains of the channels using the data from the dashboard are 27.2 dB and 27.6 dB respectively giving a difference of 0.4 dB. On the A800 tested here, the gains in the dashboard only differ by 0.08 dB. Do you recall how the pots were set on both the A500 and A800 to get the data for the dashboards?

Alternatively, could you test how the gain for the channels is depending on the setting of the pots? Setting the pots to 9 o'clock, 12 o'clock, 3 o'clock and full respectively (or the closest notch) and measuring gain for the same input voltage should be sufficiently resolving in my estimation.


Also, could you test bridged mono?

 

[amirm]

On the A500 you tested previously, the gains of the channels using the data from the dashboard are 27.2 dB and 27.6 dB respectively giving a difference of 0.4 dB. On the A800 tested here, the gains in the dashboard only differ by 0.08 dB. Do you recall how the pots were set on both the A500 and A800 to get the data for the dashboards?

I fine adjusted them on the A800 between steps and hence the comment I made. It would keep turning a bit and fall in the notch and become different. I did not do that with A500. I say they are the same in this regard.

 

[JIW]

I fine adjusted them on the A800 between steps and hence the comment I made. It would keep turning a bit and fall in the notch and become different. I did not do that with A500. I say they are the same in this regard.

Without actively measuring the output voltage, how reproducible (and maintainable) and hence practically relevant do you think your test setting is?

I understand that you want to standardise testing, but if a device is difficult to set up accordingly, then the standardised test is unlikely to be practically relevant. Testing the device in the practically relevant setting closest to the standard while explaining why the standard is difficult or impossible to reach makes the most sense to me.

In this case, maybe the data from the A500 gave you an easy way out, but elsewhere this might not be the case.


Also, do you know (or can you guess) how much of an effect a channel imbalance of 0.4 dB (i.e. 5%) has on accurate localisation, e.g. how far to one side is the centre image (or a mono signal) shifted?

 

[amirm]

Most amplifiers I test have channel imbalance of this amount without a way to fix it. So in that regard, these amps are superior. I have not tested to see how audible the difference is. Since speakers are not precisely matched either, you could be worse off or be OK when combined with the amp.

 

[amirm]

Here is the bridged mono measurement:

Would be good to drive subs this way. Are there cheaper options for half a killowatt of power?

 

[JIW]

Here is the bridged mono measurement:

View attachment 43291

Would be good to drive subs this way. Are there cheaper options for half a killowatt of power?

On the back of the A800, it says the minimum load for bridge mono is 8 Ohm. Since you did not seem to run into problems, it seems it also works fine at 4 Ohm.

Anyways, the above measurement gives a maximum current of about 10 A RMS and a maximum voltage of about 40 V RMS. The maximum voltage for 4 Ohm stereo is about 31 V RMS per channel. Thus for an 8 Ohm load in bridge mono, the output is voltage limited to about 62 V RMS giving a maximum power of about 482 W RMS.

 

[muad]

Amir (@amirm), thank you very much for your review.

Any thoughts on running A800 with lower impedance than spec'ed loads? Is it mostly limited in specs to 4 Ohm/8 Ohm bridged due to limited cooling solution or is it parts/design limitation? Behringer seems to be very vocal on "runs almost any speaker", but quite shy in putting that in specs.

In my own current side by side testing of XLS1502 and A800, I am leaning towards A800, and 240 WPC @4 Ohm is more than I expected. Still wondering, if it has potential to be pushed little more, if needed.

What makes you "lean" towards the A800?

 

[Eddiechilli]

Thanks Amorim,

sonthe amp is not the best but not a dog. Does not compete with Hypex and unfortunately does not live up to Behringers marketing;

‘A800 boasts remarkable specs and unrivaled sonic performance.

Or

‘The resulting output is no longer dependent on the actual load, but is rendered perfectly ruler flat via impedance compensation.’

However, surely in your write up there is an error when you say, ‘That amp though, generated much worse SINAD because of its non-switching power supply’. I don’t know for sure but expect this class D to have switch mode PS.

Would you be kind enough to describe how the amp was powered for test? Is it like for like and standardized through all tests? otherwise it’s not apples and apples and the comparative Sinad bar chart data is undermined.

I first started reading you some time ago, when researching LPS. In fact your blog completely changed my power habits. I thank you.

Kind regards.

 

[restorer-john]

The above is with a resistive load. I have been asked frequently how the amp performs with a real speaker. Stereophile uses a simulated load of a 2-way speaker that was given to them by NHT. I built a similar load based on schematic from Doug Self:

I do question the skyrocketing high frequency impedance in that Stereophile simulated load, but I applaud you building it for adding to reviews.

My impedance sweeps of many 2 way bookshelf speakers do not show anything like the rise in impedance at HF they suggest as typical. Many 2 way speakers use shunt resistors over the tweeters to bring down levels which prevent such high impedances.

 

[restorer-john]

It is fine with me if others think I am a cretin. But I can't imagine a better system for what I want. Ultimately, it is all that matters.
Being 58 years old I can say...my opinions would be different if my ears were 8 years old again. But they aren't.

Not remotely a cretin. Your reasoning is solid, as is the value proposition and the overall enjoyment you are getting from the unit.

Secretly, all of us wish we weren't obsessed with the minutiae, as it's by far the most expensive part of the high fidelity pursuit.

 

[LTig]

I have been asked frequently how the amp performs with a real speaker. Stereophile uses a simulated load of a 2-way speaker that was given to them by NHT. I built a similar load based on schematic from Doug Self:

View attachment 43267

Here is the impedance of my load versus frequency:
View attachment 43268

My response is more damped in low frequency than what Doug Self and Stereophile show. I will investigate this further but for now, it is close enough. Here is the phase response:
View attachment 43269

And Stereophile graph of the two:

WRT impedance the stereophile graph is not the same as the one shown by Douglas Self in his book (Audio Power Ampfier Design Handbook, 1996/reprint 1997, page 165). The latter might be a simulation using the values of the circuit. A real circuit will have other values though as the inductors have their own resistance.

• For the two inductors used in series you can account for this by reducing the value of the series resistor accordingly.
• For the inductor connected in parallel it's not so easy - you have to increase Rr to counter act the undesired damping of the Q-factor by the inductor resistance but I'm not sure if this works perfectly.

If I find some time I'm going to simulate the circuit.

 

[LTig]

Stereophile uses a simulated load of a 2-way speaker that was given to them by NHT. I built a similar load based on schematic from Doug Self:

View attachment 43267

I'm don't want to step on Dougs feet (who am I compared to him?) but I think this does not look like a realistic model. It seems to simulate two chassis (tweeter to the left, woofer in closed housing to the right) and then connect them in parallel without any crossover. One should add a typical crossover (e.g. 2nd order) and recalculate the impedance. If it does not differ significantly I'd stand corrected.

 

[MZKM]

Can you do 20Hz-20kHz wattage?

 

[LTig]

Any thoughts on running A800 with lower impedance than spec'ed loads? Is it mostly limited in specs to 4 Ohm/8 Ohm bridged due to limited cooling solution or is it parts/design limitation? Behringer seems to be very vocal on "runs almost any speaker", but quite shy in putting that in specs.

You have to see the context. The target are professional speakers (PA) and here you will not find such exotic loads as in highend audio. AFAIK almost all passive PA speakers have 8 Ohm impedance spec and the amp maker has to expect that the user connects 2 of those in parallel, resulting in a 4 Ohm load spec. The pro user takes the spec into account and acts accordingly.

 

[tomtoo]

At the moment it's at Thomann.de 143€. For the power and measurements that's fu**ing cheap..Add to this the Beringer Audiointerface and you are under 250€ for a good measuring Dac and amp with a lot of Power.