some help on TDA7293 boards off Ebay

[sardonx]

TDA7293 parallel BTL mono power Amplifier borad 350W - eBay (item 120595963571 end time Oct-12-10 07:58:28 PDT)

Well I just ordered a couple of these boards on a whim and i'm curious about a few things.

1. Will my 2x20 VAC 400VA trafo work and if so, how low of an impedence will i be able to drive?
2. Are 2 10000 mf filter caps enough? I have two DNMs which were expensive and don't wanna buy another 2 or add cheaper ones.

[abraxalito]

The impedance you can drive is more a function of the current capability of the chips than of the transformer. 20V and 7% regulation would give you about +/- 29V off-load. You'd then be able to swing up to +/-52V in very short bursts - this is a peak power of 670W into 4R. I'd not want to use bigger than 10,000uF caps on a 400VA transformer myself.

[random007]

Please post some photos of what arrives - I'd be very interested.

The amount they charge is half what I'd expect to pay wholesale locally so I'm wondering if the company depends of fake parts to break even.

Of course, it would be unjust, unfair and wrong of me to accuse them of anything apart from amazing good value without inspecting the parts that ship.

If it all works and the result is good, I'm going to be shopping even more from China in the near future, but if something blows up without warning, I won't be all that surprised. I'm probably just paranoid.

[sardonx]

abraxalito: Thanks, their page states 350watts into 4ohms. It just strikes me as odd that most of the single chip boards on there ask for just as much voltage or more, for quite a lower output wattage. But then again, i don't know much about working this stuff out, that's why i'm asking here.. lol.

random007: I'll borrow a good camera and take a few photos when they arrive. Should be 10 to 20 business days. Do you know how to tell if the components are fake? And if they are fake does that mean sound quality will be lower? I'd liek to assume since they're in China, which is where the original parts are probably made anyway, these guys can just get them at a much lower cost than us. And that would not surprise me considering what these guys on ebay charge to assemble a typical board ($6 or something). So we shall see.

[random007]

Originally Posted by sardonx

abraxalito: Thanks, their page states 350watts into 4ohms. It just strikes me as odd that most of the single chip boards on there ask for just as much voltage or more, for quite a lower output wattage. But then again, i don't know much about working this stuff out, that's why i'm asking here.. lol.

random007: I'll borrow a good camera and take a few photos when they arrive. Should be 10 to 20 business days. Do you know how to tell if the components are fake? And if they are fake does that mean sound quality will be lower? I'd like to assume since they're in China, which is where the original parts are probably made anyway, these guys can just get them at a much lower cost than us. And that would not surprise me considering what these guys on ebay charge to assemble a typical board ($6 or something). So we shall see.

Comparing the physical units to known genuine ones is the only way I know of to tell the difference... at least without trying to split the unit's case open and look at the silicon wafer itself! People (mad, strange and very dexterous people) have done that.

All the fake parts I've heard of appear to be high power transistors where the silicon dies of lower powered and cheap components have been repackaged to look like more expensive units. Fake components burn fairly quickly because the cheap silicon is vastly overloaded even when the genuine component should be well within spec.

I've not yet run into any reports of fake amplifier chips: possibly due to their complexity vs single transistors. Take heart - I'm being paranoid when I voice a concern that these cheap kits have to have something dodgy about them - it's unfounded apart from the amount being charged for them which is hardly evidence.

Much of the fake transistor racket is disturbingly professional (though, amusingly, many have simple yet glaring typo errors!). It looks like
most come from India not China.

[abraxalito]

Originally Posted by sardonx

abraxalito: Thanks, their page states 350watts into 4ohms.

That's the continuous, RMS power, given a big enough transformer. I think its a little optimistic, not very - I think the datasheet is a bit optimistic and I guess they got the power by taking the figure for a single chip and multiplying by 4.

It just strikes me as odd that most of the single chip boards on there ask for just as much voltage or more, for quite a lower output wattage.

My guess is - the majority of boards on there are not bridged. Bridging doubles the output swing and therefore quadruples the power.

I agree with random btw - the price looks spectacularly low. I've heard there are plenty of fake TDAs floating around, but I haven't definitely found one... Yes, parts prices in China are considerably lower than the small quantity prices from western distributors. I bought some TDA7293 a couple of years ago, I think the price was under $2. Have yet to try them so no idea if they're fake...

[wakibaki]

670W instantaneous peak power is a bit misleading. A meaningful and reliable measure of the maximum power output of an audio amplifier - or the power handling of a loudspeaker - is continuous average sine wave power. When you see '100Watt amplifier' on the forum here, what is meant is 100W continuous average sinewave power (often called 100W RMS), which is equal to 200W instantaneous peak power. Instantaneous peak power is a pretty meaningless term and is rarely used.

w

[abraxalito]

Originally Posted by wakibaki

Instantaneous peak power is a pretty meaningless term and is rarely used.

Its by no means 'pretty meaningless' - it means what it says. Additional meaning could be had from saying how long this 'instant' lasts though.

<edit> It has occured to me that perhaps you meant 'pretty meaningless in relation to how loud it will sound in practice' in which case yes, I agree. Continuous average power is a better measure for that, but still has shortcomings.

Continuous average power, while useful for engineers testing amps into resistive loads (and thereby for CE/UL mark safety testing) isn't particularly useful for finding out how loud an amp will play when driving music into real loudspeakers. That's because music has a crest factor considerably higher than that of a sinewave and because the load a speaker presents is reactive and varies with frequency.

[pacificblue]

It is important to note that achievable power does not only depend on supply voltage and load impedance, but also on heatsinkabilty. The datasheets states that you should not ask for more than 50-60 W output power per IC in normal class AB mode under normal conditions due to thermal resistance. It recommends e. g. a ±25 V supply for a bridged amplifier connected to an 8 Ohm load.

±29 V on a bridged amplifier that feeds a 4 Ohm load lead to ~170 W of heat dissipation. That is a lot even when it is spread across four ICs.

Anyhow, wouldn't it make more sense to use 4 ICs in parallel for a 4 Ohm load than in bridge-parallel configuration?

[abraxalito]

With the ±29V rails and unbridged, the output would be around 85W. 4 ICs in parallel would certainly be overkill for this power level.