TDA1562Q low freq roll-off

[DrLex]

I'm looking into using the TDA1562Q. The datasheet only gives a concrete value for the high-frequency roll-off and states "The low frequency power roll-off is determined by the value of the electrolytic lift capacitors." Unfortunately there is no mention of what value is needed to obtain a certain frequency, or what the roll-off is for the 4700µF caps shown in the application circuit.

Does anyone have an idea what would be a good value for C1 and C2 to get a -1dB roll-off at about 30Hz?

Or is there anything equivalent, i.e. another IC that can deliver a similar amount of power into a 4Ohm load on a supply up to 24V?

[pacificblue]

The -1 dB point is more or less twice as high as the -3 dB point. The -3 dB point can be calculated with f = 1 / (2 x PI x R x C). R is 75 k according to the datasheet. 150 nF capacitors on each input should give you the desired roll-off.

[DrLex]

Yes, that's for the input capacitors. But for this IC, the 'lift' capacitors C1 and C2 also affect the (power) bandwidth and I would like to know how. The datasheet for the similar TDA1560Q contains a table with values versus frequency, but the datasheet for the TDA1562Q doesn't.

[gfiandy]

Hi you could try the TDA7498 this will supply 60W into 6R on 24V at 10%. So theoretically it will supply 90W in 4R off the same supply - however due to extra losses in the PSU output inductors etc this theory never works but you will probably still get your 70W.

Regards,
Andrew

[DrLex]

Thanks for the suggestion, that is indeed a chip with good-looking specs. Unfortunately it's only available in SMD. Last time I had to solder a chip with 0.65mm pitch I almost went crazy already, and it only had 10 pins. Something with 36 pins is out of the question. I have found some other class D chips like the TDF8599B or the TPA3120d2 which also look promising, but again in the same hobbyist-unfriendly packages.

[DrLex]

I should have mentioned that the 4Ohm load is actually two 30W 8Ohm loads in parallel because I need to feed an identical signal to two speakers. In other words, two amps that can deliver 30W in 8Ohm are also OK. And behold: the TPA3122D2 delivers almost exactly that at 24V, and it is in PDIP. Gotta love TI!

[pacificblue]

Originally Posted by DrLex

Yes, that's for the input capacitors. But for this IC, the 'lift' capacitors C1 and C2 also affect the (power) bandwidth and I would like to know how. The datasheet for the similar TDA1560Q contains a table with values versus frequency, but the datasheet for the TDA1562Q doesn't.

Step the up by the same ratio as the input capacitors. The datasheet gives 4700 µF lift for 100 nF input. So for 150 nF input use 6800 µF lift, for 220 nF input use 10000 µF lift, etc.

[ashok]

The TDA1562 apparently just has a higher current capability to handle 4 ohm loads. So ( boost ) supply capacitance should be about double that used for TDA1560Q.
The TDA1560Q says for -3dB at 10 Hz you need 4700uF and input cap of 560nF. As Pacificblue mentioned -1dB is about 2x the -3dB frequency.

So for -3dB at 15 Hz ( -1dB at 30 HZ ) you need about 370nF ( 0.39uF ) at the input and something between 4700 and 3300uF for the supply cap. This would need to be double at 4 ohms and so 10,000uF recomended by Pacificblue sounds appropriate.

I lost my TDA1560Q ! So I never got round to trying out my sub amp !
I bought a new TDA1560Q today ! Since Murphy is always around I will find the first TDA1560Q chip after all components are soldered on the board !
One day Murphy will be shot .