TI TPA3255EVM

A little basic question to these chips, especially the TPA3255: To achieve more tham 300 watts from a single +53.5V rail, bridging two power amplifier sections is a demand. And yes, the evaluation board can also be configured as four separate amps. But how does it work to get monophonic 600 watts from this chip, as the specs state?

Best regards!
 
There are some graphs on the TPA3255 datasheet that reveals the power, supply voltage, load impedance and configuration required to get a particular power output.

There are graphs at 1% THD and 10% THD. AT 1% with a 2 ohm load and a 53 volt supply the most you will get is 550 watts, but that's at +75C. You might get a few more watts at a lower temp. Max power at 10% distortion is over 600 watts.
 
Figure 24 on page 13 gives output power VS case temperature for 3 different load impedances. For the 2 ohm case, about 670 watts are obtained at 0C but only about 595 watts are available at 100C. 75C provides about 620,,and 50C about 640, while 25C gets you about 655 watts.
 

Attachments

  • Fig 24.jpg
    Fig 24.jpg
    110.4 KB · Views: 989
diyAudio Editor
Joined 2001
Paid Member
So i’m Planning to use TPA 3255 to power my Linkwitz LX 521 woofers, which are multi-amped.

Each of 2 channels has two of these 4ohm subwoofers because they are open baffle so need some mojo. I’ve watched the speakers in action and the cones move a Lot.

Seas L26RO4Y 10" Subwoofer - 4 Layer VC - (D1004-04)

So I have a choice of putting the woofs in parallel for 2ohms, in series for 8ohms.
Looks like 2 ohms is Ok?

So my ultimate plan is one tpa3255 for each channel. I ordered a second 3255. Paid the $150 for this one, hope they let me get it. 10% distortion doesn’t work for me, even occasionally, so I want plenty of headroom.

Its occurred to me that a better approach might be to split each 3255 into 2ch and use one channel into each 4ohm driver per side.

I suspect I’m just a bit jumpy about a 2ohm load, but do these amps handle that fine? Which the best option?
 
Last edited:
diyAudio Editor
Joined 2001
Paid Member
Hey Tubelab, here’s something that will amuse you!
It plugs right into any of the high power TI amps in this series, and gets power from the amp. The sound is probably no better than a cheap pedal, with LEDs for distortion, but it has tone controls, and hey! it’s truly “Plug and Play your guitar"

SIDEGIG-GUITAREVM Guitar SideGig Plug-In Module | TI.com

http://www.ti.com/lit/ug/slau714a/slau714a.pdf

I ordered one just because I think it’s cool they were having fun. 600watts here I come!
 
Last edited:
Figure 24 on page 13 gives output power VS case temperature for 3 different load impedances. For the 2 ohm case, about 670 watts are obtained at 0C but only about 595 watts are available at 100C. 75C provides about 620,,and 50C about 640, while 25C gets you about 655 watts.

The difference is 36.6Vrms (0degC) vs 34.5Vrms (100degC) at 2R which corresponds to 0.5dB over the full temperature range. (0degC is a theoretical value as you'd need nitrogen to achieve this temp)

The connections and wire resistance will most likely have more impact here than the temps variation.

3m speaker wire (6m total for +/-) at 2mm^2 is 50mR, 200mR for a thinner 0.5mm^2 wire. Add 12.5mR per joint (connection point) give 100mR and 300mR.

Calculated with 35Vrms into 2R (without inductor loss) translates to:

35Vrms^2/(2R+100mR)=583.3Wrms

35Vrms^2/(2R+300mR)=532.6Wrms

:)
 
diyAudio Editor
Joined 2001
Paid Member
When you build a Linkwitz loudspeaker you pretty much have to stick with what he specs. It's a system: active crossovers with lots of EQ, specific speaker drivers, baffle size and shape. He even likes people to use the same amps for the 4 different frequencies. I won't do that, but they have to have the same gain or be adjusted to have that. Here's a link if you aren't familiar with them:

LX521 Description

I assume by "chassis" you mean a different model woofer? Well, these are woofers custom made for Linkwitz by SEAS and are extremely linear- at a cost of efficiency. He doesn't sell them, but they are designed for him. They have to be pretty linear because there is significant excursion in a smallish open baffle. AND he EQ's them in addition as I mentioned, which increases excursion also. And if you use a different driver then the EQ isn't going to match up with the response of the driver.

AND why worry about efficiency when you can get cheap Class D amps? I just want to use them in the best way possible!

I've heard these speakers at Burning Amp various times and at Mr Linkwitz's house and I love the sound. I also have simpler, full range speakers and love them also. It's fun to have systems with the 2 approaches!

@Variac, why not use a chassis with better efficiency? 80-85dB is pretty low end.
 
Last edited:
Hi ambiance,

Is there a reason why you want the 24V output model versus the 50V output model?

These power supply designs have my attention:

http://www.ti.com/lit/ug/slou293c/slou293c.pdf

HEP-600-MEAN WELL Switching Power Supply Manufacturer

In the meantime, the Mean Well LRS-350-48 switching power supply outputting 48.5V is doing a respectable job. I am looking for an improvement to get the most out of the TPA3255EVM.

My order from Mouser just came in, so the fun will begin with adding a pair of vintage Altec Peerless 15335A line level input transformers, and then implementing the PFFB mod.
 
Hi Thing,

I thought the model I chose had 2 outputs, 50V & 24V. Or am I wrong?

I am thinking the power supply can also power my DUG's 3116 board--just trying to make it flexible. Am I better off getting just the single 50V? I have no experience as how power supplies (with secondary output) affects the SQ. FYI, I will not be using both amps at the same time.

My other thinking is to use Class D Audio's power supply, like the ones they used for CDA 1000 W model. I think it is more expensive though. Do you think it's better?

Thanks for the input!