Is the Sure AMP (AA-AS32171) a fraud, or am I doing something wrong?

[LEECH666]

Hi,

I've got a technical question. I bought this tiny little Class D AMP via Ebay, which is supposed to have 50W (RMS) Rated output power. However in my test I am unable to measure a power draw anyway near that rated power.

http://store.sure-electronics.com/audio/audio-amplifier-170/modesty-series/aa-as32171-1414

It's based on the ST TDA7492.

It's been some time since I've tinkered with electronics and now I am unsure if my measurements are correct.

My test setup: Mains Power -> Power Meter -> Manson PSU -> Fluke 87 IV (Current) -> Sure AMP -> Speakers (8 Ohms?)

My Laptow / my Smartphone (Galaxy S3) is used as the audio source.



When dialing the volume encoder of the amp to the maximum setting and setting the audio source to maximum volume the power draw displayed on the power meter is a measly 8 watts (playing a 40 Hz sine wave). Same with the readout display on the Manson PSU 12V/423mA -> 5 watts.

Is that as good as it gets with TDA7492 based amps or am I doing something wrong here?

Help appreciated.

Cheers,
L666

[Odysseus]

From Sure's webpage, they quote the TDA7492 datasheet for output power specification and test conditions.  Unless you replicate these conditions, you cannot guarantee you will meet the spec. They are:  VCC=24V, Fin=1kHz, Rload = 6ohms, none of which I think you have met.

[LEECH666]

I didn't want to go with 1Khz, because I don't currently have a dummy load resistor, and 1KHz is a bit hard to bear at those volume levels.

I could raise the voltage to 24V tho. Have tried 20V before but it didn't seem to make a huge difference.

It's a bit late here to do loud volume tests. So I will probably continue tomorrow.

Thanks for the replay.

Cheers,
L666

//EDIT

Okay did a quick test with 24V / 1KHz Sine, the speakers impedance is supposed to be 8 Ohms iirc.



Sorry for the overexposure in that picture.

I still find those numbers to be far from the claimed values. It seems to be more like a 15W amp not a 50W one.

[suicidaleggroll]

Is your smartphone driving the input of the amp fully?  Are you getting clipping or just maxing out the volume?

[LEECH666]

The volume is controlled through a rotary encoder which just stops increasing the volume at some point. I can't increase the volume any further, that's what I am experiencing. Unless I use the software equalizer or preamp setting on my smartphone (AIMP) / laptop (Winamp / Foobar2000).

I was under the impression, that the regular laptop sound card was capable to produce standard line levels (2Vpp iirc?), but I have not verified this.

[janoc]

I was under the impression, that the regular laptop sound card was capable to produce standard line levels (2Vpp iirc?), but I have not verified this.

That would be the first thing to check, actually - if the amp is not driven to the limit, it will not produce the full rated output neither.

Ideally you should have a scope on the output and keep an eye on the distortion - once it starts to clip, you are overdriving it. On the other hand, this may not be actually built to produce the full 2x50W the chip is capable of (with 10% distortion though - you likely wouldn't want to listen to that).

[DrGeoff]

You must use a resistive dummy load to perform this test. Speakers do not represent a constant impedance. The impedance of the speaker changes with frequency and depends on the cabinet it is mounted in.

Also monitor the output waveform and make sure you are not clipping.

[janoc]

From Sure's webpage, they quote the TDA7492 datasheet for output power specification and test conditions.  Unless you replicate these conditions, you cannot guarantee you will meet the spec. They are:  VCC=24V, Fin=1kHz, Rload = 6ohms, none of which I think you have met.

Even with 8Ohm load, 20V instead of 24V and lower test frequency the amp should be taking more current, that's way too much difference - with 8 Ohms load it is still specced at 40W.

I don't think that the amp has a built-in big ass high-pass filter that would attenuate those 40Hz so much - the frequency characteristics dips at 0.5dB at 40Hz only.

[DrGeoff]

It's a switching power amp, so the current waveform will be all over the place and probably not measurable using a simple RMS meter.

[lutkeveld]

Does a sinewave max out power? Don't you need pink noise for that?

As said: make sure that it's input is saturated and that you test the output with a dummy load. You can use an old speaker and rip the cone assembly out the basket.

[PeterFW]

Oh yea... the power output of a amplifier...
Others have pointed out some of your mistakes but i will answer your question from the subject.

These amplifiers work just fine, they are not a scam.
I have been running two of the 100W and one of the 50W single supply and two rather big ones with a dual supply.
Can not remember wich chips they use...
They work, they do what they are supposed to and sound good.

Unless you buy high end equipment (PA stuff from the big players) you can take 25% off anny power rating you read when it it is rated in RMS.
If it is sinus, music or PMPO... stay away from it.

You can not trust anny power ratings for consumer products because consumers still think "The Watt means how loud it is." and even the folks who should know better still calculate with "Watts" to meet sound pressure requirements.

What you want to look out for is the efficiency of your speakers, because you can pump 1000W in a shitty speaker and still not be as loud as 200W into a good one.

[suicidaleggroll]

Does a sinewave max out power? Don't you need pink noise for that?

Power ratings are generally spec'd for a single frequency sine wave near 1 kHz.

[LEECH666]

Thanks for all the answers and please excuse the crudity of my measurements. I didn't even intend to measure the amp. Initially I just wanted to see if it works of a 12V source, since I'll be going to a festival this weekend and I want the amp to power two speakers so me and my pals can listen to some music on the campgrounds.

I guess I was just expecting higher numbers. However I might investigate some more (out of curiosity) after I return.

I need to go to work now and after that it's time to load all my festival equipment into my car.

Thanks again.

Cheers,
L666

[LEECH666]

Oh yea... the power output of a amplifier...
Others have pointed out some of your mistakes but i will answer your question from the subject.

These amplifiers work just fine, they are not a scam.
I have been running two of the 100W and one of the 50W single supply and two rather big ones with a dual supply.
Can not remember wich chips they use...
They work, they do what they are supposed to and sound good.

Unless you buy high end equipment (PA stuff from the big players) you can take 25% off anny power rating you read when it it is rated in RMS.
If it is sinus, music or PMPO... stay away from it.

You can not trust anny power ratings for consumer products because consumers still think "The Watt means how loud it is." and even the folks who should know better still calculate with "Watts" to meet sound pressure requirements.

What you want to look out for is the efficiency of your speakers, because you can pump 1000W in a shitty speaker and still not be as loud as 200W into a good one.

Just wanted to note, that I am aware of efficiency / sensitivity of speakers.

My Speakers use these chassis:

http://www.traumboxen.de/ad-audio/t8015.htm -> SPL 94 dB @ 1W/1m
http://www.traumboxen.de/ad-audio/h257.htm -> SPL 110 dB @ 1W/1m

And thus the speakers (ULB 8 Pro Variant) should be decently efficient.

That's actually the reason why I am a little disappointed in the amp. I thought the tiny amp with it's claimed 50W RMS (or 40W @ 1% THD) would actually somewhat match the performance of my bigger AMPS. I don't know where the problem is. Could be the input voltage of the soundcard, but using the same soundcard on my other amp (Pioneer A-77x, 100W @ 8 Ohm) plays much louder, and sounds much better at higher volumes.

[continuo]

Just bring enough beer along to the party and no one will notice 

[PeterFW]

That's actually the reason why I am a little disappointed in the amp. I thought the tiny amp with it's claimed 50W RMS (or 40W @ 1% THD) would actually somewhat match the performance of my bigger AMPS. I don't know where the problem is.

I just had a look, i got a TDA7492 board from sure, not the same as yours but with a 24V supply it got more then loud enough on a pair of full range drivers but thats it.
I never measured the output, you get a reasonable amount out of them but i would put them in the "desktop use" category.

If i understood you right you have a pair of Jobst Audio ULB 8-Pro connected, the amp is seriously underpowered for those speakers.
You need way more headroom, music is fairly dynamic, the amp will clip at anny decent volume level on the peaks.

The next bigger one, the 100W TDA7498 boards, gets closer to what you want but since the whole thing is on a log scale...
But i have two of the TDA7498 board used on a semi active 2-way system for about 20 people in a small room in several occasions.

But i upgradet that to IRS2092 amplifiers with a beefy dual supply smps.

[Lightages]

From the data sheet for the TDA7492, they have test curves for voltages and such. Their test curve for output power vs VCC only starts at 15V. At 15V the output power into 6 ohms is around 13W with 1% distortion. A rough extrapolation would have it giving around 10W at 12V, perhaps even less. Your 5W measurement is probably not far off.

To keep the distortion below a more desirable 0.1%, the supply at 25V, you need to stay below 25W. Extrapolating this to 12V VCC I believe that you would want to stay below 5W for reasonable distortion at 1kHz.

But wait! It gets worse.

The distortion rises to almost five times more at 20Hz and 7kHz. So to keep things below 0.1% across the audio band you need to stay below 1W output!

And now drive an 8ohm load and you are looking at staying below 0.5W.

These are not what I would call good audio amps.

[macboy]

Does a sinewave max out power? Don't you need pink noise for that?

As said: make sure that it's input is saturated and that you test the output with a dummy load. You can use an old speaker and rip the cone assembly out the basket.

No, you can't get maximum output power with pink noise; the crest factor is too high.
A sine wave is the 'proper' or usual way to test amplifier output power. However, it is necessary to monitor the output waveform so that you can determine when the amp is clipping. Output power is measured at just before/at the onset of clipping. You can't just max out the volume and call it a good test.

[janoc]

That's actually the reason why I am a little disappointed in the amp. I thought the tiny amp with it's claimed 50W RMS (or 40W @ 1% THD)

Actually the datasheet (http://www.st.com/web/en/resource/technical/document/datasheet/CD00205863.pdf) says that it is 50W RMS @ 10% (!!!) THD (first page)

Then on page 15 you will find the curves for 8ohm load - 1% THD at 30W max and 24V power supply. For decent distortion you must stay way lower - at 20W you are at 0.1% and it shoots up sky high afterwards.

[LEECH666]

Just bring enough beer along to the party and no one will notice 

Thanks mate, that was may plan anyway and that's what I did.

[BradC]

My Speakers use these chassis:

http://www.traumboxen.de/ad-audio/t8015.htm -> SPL 94 dB @ 1W/1m
http://www.traumboxen.de/ad-audio/h257.htm -> SPL 110 dB @ 1W/1m

And thus the speakers (ULB 8 Pro Variant) should be decently efficient.

Yes, but.... what is your speakers *actual* impedance at 40Hz (including the crossover)? Most that I have experience with roll off badly and deliberately under 80Hz to protect the drivers.

[lutkeveld]

You can not trust anny power ratings for consumer products because consumers still think "The Watt means how loud it is." and even the folks who should know better still calculate with "Watts" to meet sound pressure requirements.

And then you have the 0.1% that use the correct speaker output in Watts: acoustical watts 
But saying that your PA set can only pump out a few acoustical watts doesn't impress much people 

[LEECH666]

According to this thread on DIYAudio forums there are four solder jumpers on the PCB (R6, R7, R10, R11) that set the gain of the AMP. I already changed the jumpers to the highest gain setting yesterday but was too tired test it.

According to the datasheet I now set the internal gain of the TDA7492 from 21.6 dB to 33.6 dB. Let's see how that turns out.

Im also considdering using one of these ...



to boost the voltage from ~12V to ~24V for my lead acid battery application.

Totally forgot I had one of these boost converters lying around ...

Cheers,
L666

[Neverther]

I tried the non current limited version of that on 2x100W TK2050 Surefire amp. It could not start as the surge current was too great. The 5A laptop supply went into current protection mode or something as the converter was taking in 7 amps.
Connecting the boost converter after it had powered up had some sparks on the DC jack every time as the capacitor array charged up.
Running on real 24V supply now.

[PeterFW]

Running on real 24V supply now.

On a side note:
Those 24V supplys from china are not that bad at all, if one can stomach the shipping time they are a good buy.

[miguelvp]

...

Im also considdering using one of these ...



to boost the voltage from ~12V to ~24V for my lead acid battery application.

Totally forgot I had one of these boost converters lying around ...

Cheers,
L666

Hmm, the silk screen shows OUT- where you have Output+
Make sure you look at the PCB to see what is the right way and of course measure it before plugin it into anything.

[mikerj]

I tried the non current limited version of that on 2x100W TK2050 Surefire amp. It could not start as the surge current was too great. The 5A laptop supply went into current protection mode or something as the converter was taking in 7 amps.
Connecting the boost converter after it had powered up had some sparks on the DC jack every time as the capacitor array charged up.
Running on real 24V supply now.

That's hardly the fault of the converter - doubling the voltage into a given (resistive) load means four times the power, and therefore four times the current draw from the 12v supply (plus some extra due to efficiency).

[LEECH666]

Hmm, the silk screen shows OUT- where you have Output+
Make sure you look at the PCB to see what is the right way and of course measure it before plugin it into anything.

You're right, in the picture I posted it looks like the silkscreen is saying Out- while the red annotation arrow says Out+. However on my physical unit the silkscreen is in compliance with what the red arrow say. I guess the plus sign just got milled away, when they routed out the board.

I already tested the converter (voltages only, as I lack a working dummyload at the moment) and the pinout shown by the annotation arrows seems to be correct.

Fed +12V in and adjusted the output to +19V. Haven't done anything else with it yet. Lazy weekend.

Cheers,
L666