TPA3255 - all about DIY, Discussion, Design etc

[454Casull]

From the latest PFFB application report, it can be seen that PFFB performance gain with good inductors is ~6dB, like bringing down 0.003% to 0.001%

http://www.ti.com/lit/an/slaa788/slaa788.pdf

Take note that some fact in the tables are labeled wrong.

Thanks for the link.

 

[DPH]

I m using a tpa3251 board.

Will there be a problem if I use 40000uf capacitor bank after the large trafo and rectifiers ?

Linear power supply? I mean it's going to take a good amount of time for the bank to discharge after shut down, and meaninglessly excessive capacity, but pretty harmless. (Caveats of power limiting start up acknowledged)

 

[SanjeevM]

Linear power supply? I mean it's going to take a good amount of time for the bank to discharge after shut down, and meaninglessly excessive capacity, but pretty harmless. (Caveats of power limiting start up acknowledged)

I am expecting that due to the large PS caps, during bass transients supply voltage will not sag as much, so bass will be better.

 

[SanjeevM]

I am expecting that due to the large PS caps, during bass transients supply voltage will not sag as much, so bass will be better.

As you rightly say, startup current will be a problem. To counter that I am going to use a soft start circuit.

 

[F4n]

Someone solved the problem with glowing diodes? On tpa3255 on page 30

 

[emedia4u]

Someone solved the problem with glowing diodes? On tpa3255 on page 30

This solution didn't work and startup is not a problem - the shut down is because of huge power filter capacitors on board.

 

[doctormord]

I am expecting that due to the large PS caps, during bass transients supply voltage will not sag as much, so bass will be better.

This might work for capacitance values of minimum 1,000,000uF and up.

 

[SanjeevM]

This might work for capacitance values of minimum 1,000,000uF and up.

But I have read in Class A forums that they use 100000 uf per channel, doesn't it help?

 

[wdecho]

But I have read in Class A forums that they use 100000 uf per channel, doesn't it help?

My understanding of class A is that they need better filtering than either class D or A/B. I am using 30K uf just because. I feel like 15K uf would probably be plenty enough.

 

[DPH]

I am expecting that due to the large PS caps, during bass transients supply voltage will not sag as much, so bass will be better.

One it's fairly simple to figure out how much the rails will sag under such load, especially if one takes a conservative assumption that you don't get any help from the transformer supplying current. Two the low frequency psrr is liable to be very good (can't remember if there's a plot in the datasheet, but you're down where there's a ton of loop gain). Three, are you actually pushing these amps so hard as to truly load down the rails? Pro sound maybe, home, not likely at all.

 

[abraxalito]

But I have read in Class A forums that they use 100000 uf per channel, doesn't it help?

Was the amp SE classA where the active element/pass transistor was acting as a shunt? If so the capacitance requirement is decreased very substantially as in such a case the amp draws a constant current under all signal conditions.

 

[SanjeevM]

What I did was to connect a DMM to the supply lines of my 3251 amp and playing it at moderate volume.

The digits were moving all the time, and variation was more when drums sounded. I dont remember the reading, but I can redo the experiment anytime.

 

[SanjeevM]

This was with a linear psu with 10,000uf capacitance.

 

[hajj]

Anyone tried a Cap Multiplier with the TPA3255?
I am planning on building thehttp://www.diyaudio.com/forums/solid-state/297921-jumas-easy-peasy-capacitance-multiplier-28.html#post5215388 to try with my TPA3255 boards

 

[alexttt71]

good afternoon. did anyone have a problem with sibilants? TPA3255 300W +300W DC50V 3900UF/100V*2 Channel Digital Power Amplifier Board | eBay

 

[impuls60]

Interesting question!
I only have the 3e-audio board and it has no sibilance if anyone wondered. I used to get that with some of the cheaper tpa3116 boards before mods.

 

[SanjeevM]

I am also using a 3eaudio 3251 board. No problem of sibilants.

 

[alexttt71]

Thank you

 

[DPH]

What I did was to connect a DMM to the supply lines of my 3251 amp and playing it at moderate volume.

The digits were moving all the time, and variation was more when drums sounded. I dont remember the reading, but I can redo the experiment anytime.

It'd really be better to actually get a trace, as that is hardly a robust test. And far more important, are you seeing scads of distortion from PSU modulation? If not, and I'm going to assume not, this is digging around trying to invent a problem that isn't there.

 

[doctormord]

+1