Power supply voltages.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
(Warning: beginner's question!)

When building a chip amp using a chip which will accept variable voltages, is it best to have higher voltages and lower current or lower voltage and higher current?

Just curious.

Background:

Currently making a little amp with a TDA7370B, running at 13.8 volts. Sounds great to me! Just wondering if it's worth going for a higher voltage power supply (can be a max of 18 volts) when making the permanent PS.

Thanks in advance.
 
When one designs a power supply, you first determine how much voltage AND current a device will draw. As this chip may indeed go up to 18 Volts this rating is for maximum power output. The ideal or optimum voltage is at the stated 13.8. When you know the voltage you are choosing (13.8) and the current it will draw at that voltage........design the supply to source that voltage and current requirement. The often argued general rule of thumb is to have 1 1/2 times more current capacity than the device can draw.

_____________________________________________________Rick.........
 
Last edited:
I would tend to go for the higher voltage as this gives a better headroom for transients.
You can get some pretty high transients well before the amp takes too much power.

I got caught out with power amps, my first was a low voltage and clipped horribly before it gave full power.
The next amp I built I increased the supply rails by around 50% and had much better results.
 
On a related note how do you determine what voltage to run at when you build an amp based on an op amp. Most op-amps seem to accept a range of voltages. Example OPA627 runs between ±4.5V to ±18V. PGA2310 runs between ±4.5V to ±15V

Do you try to use the highest voltage as possible or 80% or somewhere in the middle? I understand with a higher voltage you can get better dynamic range. What are the trade offs with going with a higher voltage?
 
Looking at the performance curves of the OPA627 there was only one curve with varying supply voltage....that was the Bias supply to power supply. Not being able to back up any claims by data (That would be alot more performance curves!) ..my "hunch" is THD+n will find a happy voltage somewhere in between. The clipping issue is a hollow one as all one needs to do is choose a loudspeaker with a higher sensitivity rating (SPL in Db), to get to an acceptable Db level.......Any amp will clip, just depends on the volume level it starts clipping (Is it high enough?) Another 'issue' I have with running at full rated voltage, is generation of heat.........got heatsinks??

__________________________________________________Rick...........
 
. . .The ideal or optimum voltage is at the stated 13.8.. . .

That is true of TDA7370B. One of the tripath also uses that voltage.
Is there actually some weird little car that runs 13.8v with the engine on? That's like 1v less than normal cars. Healthy alternators do 14.8v.

Maybe the 13.8v is because the car chip designers live in an RV with a trickle charger? :) That's a real mobile application with 13.8v.
 
The old LOW tech lead acid batteries gassed profusely when overcharged.
They required regular checking and occassional "topping up".

To avoid this, battery manufacturers were asked to develop a low maintenance lead/acid battery.

We have gone even further and now have zero maintenance lead/acid batteries.
The chemistry inside these NEW technology batteries is quite different from the OLD.
One consequence of zero maintenance and the "no topping up" is that these new batteries do not tolerate overcharging.

The charging voltage and the temperature compensation of same must be very accurately set. 13.8Vdc is that voltage, to maintain the zero maintenance and a reasonable battery life.
 
Really?
Less might be magical.
More voltage might be the opposite.
Its because stuff doesn't sound good overrun.

With more voltage than optimal, expect a harder sound with lower resolution, no increase in full decibels at the speaker, and no benefit whatsoever. Every time you double the power, you get +3db more, and that won't be happening--any increase within the operable range of this chip won't even do 1db more. It would be a fraction of a decibel. There's absolutely no need to increase the voltage on that amplifier. Higher voltage would go with a completely different amplifier.

If you're switching power supplies, try a finer quality power supply, not a higher voltage power supply.
 
Good advice from Daniel. If "higher voltage than 13.8" means something like the 14.4VDC that the chip is spec'd at, that's fine. If it means something like 17.9VDC instead, there really is little to no benefit to be gained.
I have a similar situation here, except my chips have a built-in current limit of 2.5 amps as stated in the datasheet. Same(?) 9 to 18 volt supply range, but you can see that the supply voltage isn't a primary issue. These are designed for car radio apps, and trying to get that last watt out is pointless.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.