TPA3116D2 Amp

Does this analogy work?

The analogy works but the amp is different from the people drinking the water, which is where it falls over (perhaps 'falls short' though is more accurate). People don't care about how much water's in the bucket when they arrive to quench their thirst, so long as they don't encounter an empty bucket. But the amp (and hence the SQ) is sensitive to that level in the bucket (we could say the 'water pressure' which is analogous to voltage) - we want it to be kept right up to the top, all the time.

So people who are consumers of water (which here is the analog of charge) - are in effect current sinks in their demand for water. They care nothing about the pressure (voltage) so long as enough current flows. The amp though does.
 
I have been listening to some original YJ Blue/blacks now since April, and like them, I would like to adjust the gain, but am OK with the factory setting for now.

I would like a suggestion for the best bang for the buck in a 2.1 amp using this chip.

I need this for a desktop set-up for my son's computer. USB DAC-Amp-Fullrange sattelites, and sub.

Need to be able to adjust the crossover freq, and bass level independently.

I've been following this thread and the myriad of tweaks since the beginning (almost) but I've lost track of late.

Thanks,
John
 

Attachments

  • WP_20141213_001.jpg
    WP_20141213_001.jpg
    770.6 KB · Views: 722
Last edited:
I'm setting up my switch boxes so I can A/B the Sure boards with and without the OSCON cap mod, and with the little tube preamp I got.

Two Way Amplifier Switch Selector Loud Speaker Switch | eBay


Bridge rectifiers in output Sure3116 will guide the oscillation voltage spikes that occur without speakers connected towards the PVCC capacitors. Post#1 tells us these could burn 50V filmcapacitors in open output instantly, TI confirms in next posts. The 25V Oscons are far weaker than any 50V filmcapacitor, they might not burn, but get damaged for sure I would say.

So what happens every time you switch speakers here???
 
Bridge rectifiers in output Sure3116 will guide the oscillation voltage spikes that occur without speakers connected towards the PVCC capacitors. Post#1 tells us these could burn 50V filmcapacitors in open output instantly, TI confirms in next posts. The 25V Oscons are far weaker than any 50V filmcapacitor, they might not burn, but get damaged for sure I would say.

So what happens every time you switch speakers here???

Has anyone had an Oscon go bad?
 
A TPA3116 isn't a free energy machine, it won't charge your PVCC capacitors to a high enough level to threaten their voltage ratings while running off those same capacitors.

Good to know why those bridge rectifiers are on the amp though... *marks down change to cAMP design*
 
A TPA3116 isn't a free energy machine, it won't charge your PVCC capacitors to a high enough level to threaten their voltage ratings while running off those same capacitors.

Good to know why those bridge rectifiers are on the amp though... *marks down change to cAMP design*

So you mean 50V MKP or MKT filmcapacitor does burn but a 25V Oscon doesn't ? Surge voltage for those filmcaps is in hundreds ?? of volts usually, for Oscon ~30V ??????
 
The analogy works but the amp is different from the people drinking the water, which is where it falls over (perhaps 'falls short' though is more accurate). People don't care about how much water's in the bucket when they arrive to quench their thirst, so long as they don't encounter an empty bucket. But the amp (and hence the SQ) is sensitive to that level in the bucket (we could say the 'water pressure' which is analogous to voltage) - we want it to be kept right up to the top, all the time.

So people who are consumers of water (which here is the analog of charge) - are in effect current sinks in their demand for water. They care nothing about the pressure (voltage) so long as enough current flows. The amp though does.

At the risk of beating a dead horse... what if I change the analogy a bit: instead of a bucket, it's a water tower (a big water reservoir up in the air). And now the people do care about water pressure, because they want to not just quench their thirst but also put out fires, take showers, wash their cars, etc.

So there's clearly some device (the power supply in this analogy) that fills the water tower (the capacitor). But as long as the tower stays full, there's both pressure (voltage) and supply (current) to meet the town's various needs.
 
So you mean 50V MKP or MKT filmcapacitor does burn but a 25V Oscon doesn't ? Surge voltage for those filmcaps is in hundreds ?? of volts usually, for Oscon ~30V ??????
Two different caps for two different applications.

Output: use film capacitors.
PVCC decoupling: use Oscon + ceramic. Film caps big enough to have enough uF for the job have too much ESL for the job.
 
Two different caps for two different applications.

Output: use film capacitors.
PVCC decoupling: use Oscon + ceramic. Film caps big enough to have enough uF for the job have too much ESL for the job.

Sorry you miss the point here.
If you couple output to pvcc capacitors like Sure3116 with bridge rectifiers and we know 50V filmcaps that are less sensitive to overvoltage than Oscons, that those 50V filmcaps burn instantly and that TI isn't surprised at that, what will happen to 25V Oscons that do not have surge voltage margin? I would say at least you damage them every time you disconnect speakers and output without load oscillates.
 
Bridge rectifiers in output Sure3116 will guide the oscillation voltage spikes that occur without speakers connected towards the PVCC capacitors. Post#1 tells us these could burn 50V filmcapacitors in open output instantly, TI confirms in next posts. The 25V Oscons are far weaker than any 50V filmcapacitor, they might not burn, but get damaged for sure I would say.

So what happens every time you switch speakers here???
 
No need. You want fast/schottky for situations like clamping an inductor in a SMPS, but in this scenario, the output capacitor limits the dv/dt so an ordinary silicon rectifier is OK.

I was thinking more of the resonance frequency of L and C.

10uH and 0.68uF is about 610KHz.

Driving an L-C circuit with a step function will cause ringing.

Add tolerance of L and C and you could have a resonance close to the switching frequency...then you get voltage gain.

Yes there is also some damping R-C on the output but why take a chance.

🙂