TPA3255 Amplifiers

[ajc9988]

So, we all know about the A07 from AIYIMA. Toid did a video on it.

Well, there are other boards with the same Ti chip in it available, two of which I bought and can give my impressions on when they get here (taking the slow boat from China along with some 1000W (really 800W continuous) power supplies).

In my search for amps to use at a good price, I discovered the designer of an amp who sells the designs in spun-off versions for other manufacturers.
https://www.jlelectronicsph.com/product/sylph-d200-tpa3255-pffb-stereo-ready-to-run-amplifier-module

From here, I found 3eAudio’s version of this same amplifier.

https://www.aliexpress.com/item/10000090083368.html?pdp_ext_f=%7B%22sku_id%22:%2220000000119654870%22,%22ship_from%22:%22CN%22%7D&gps-id=pcStoreJustForYou&scm=1007.23125.137358.0&scm_id=1007.23125.137358.0&scm-url=1007.23125.137358.0&pvid=c25cb47e-794e-43b5-8b39-5278c5484c15&spm=a2g0o.store_pc_home.smartJustForYou_821093785.0

TPA3255-2CH-260W

And then the further cutdown design (which may or may not be licensed or may just be a design someone else made based off of the above):
https://www.aliexpress.com/item/1005003213211009.html?spm=a2g0o.cart.0.0.37013c00JIN0Kh&mp=1

I ordered the latter two, with the lowest cost option coming in first. I’ll update once I have them in hand. The lower cost one I only plan running a single channel on, and it will be the sub channel, so it should be fine. The upper two channels (woofer and tweeter) will be on the 3e-audio amp.

Anyone wanting to talk about TPA3255 amps, feel free to chime in.

[123toid]

I will be really interested in your take with the chip. Do you think that heatsink will limit the continuous power output?

---

https://www.youtube.com/123toid

[elliottdesigns]

Let us assume 90% efficiency:

250w max for 1% 4ohm (BTL). 0.1*250=25w heat output, it could probably cope with that size of a heatsink.

2*77w for 1% thd 4ohm (SE). Will be even lower.

Things will however change when using lower ohmage and/or higher %thd.

Also worth noting that as the chip heats it’s efficiency decreases, so the heat’s rate of transfer between the chip and the heatsink is critical, if not good enough the chip will get too hot, efficiency would decrease exaggerating heat output… Chip then shuts off or if it happens quickly enough, dies.

My answer: it probably needs testing, but I’d say it should be fine if you stick within the limits I’ve given, even that’s an estimate though.

[ajc9988]

That is a concern. Long term plans for mounting on the plate that I am making for the amplifier is to get at least a 10.08″x4″x2.5″ heatsink, sand the side that will attach to the aluminum plates I already have (12″x12″x0.125″), then get a low temp solder and a heat gun, drill some holes, then basically bolt them together with the low temp solder between and use the heat gun to flow the solder (which could take awhile to flow the solder because heatsinks that size can take some energy). As you heat it, you have to create a way you can tighten intermittently, that way to squeeze it as you flow the solder. I suppose some metal clamps could do the same instead of bolts.

Then, buy some aluminum blocks, use the drill press to drill holes, then tap those holes on both side of the block and replace the heatsink on the amps with those new blocks. Then, drill matching holes in the heatsink/amp plate, then screw into the tapped holes on the top of the block. You also need to drill holes for standoffs. In effect, you are screwing down the amp upside down, but creating a better transfer of energy to what will be outside of the speaker. You want to use thermal paste both on the block attaching to the chip and on the block side attaching to the plate.

Meanwhile, I only need 150W@4Ohm for RMS of the 12″ drivers I am using (that will be single channel on its own amp), 80W@8Ohm for RMS of the woofer, and the tweeter claims up to 110W@6Ohm RMS, but likely much less, so that will be tuned last.

I plan on trying to cap the voltage out of the DPS, either through DPS controls or through a trimmer pot for the different channels after the DPS acts as an active splitter. To do that, it would just take the right value of resistor to mimic being a speaker (meaning 1 expensive 200W 4Ohm resistor, and a $9 100W 8Ohm resistor) and a current clamp. You will want to check clipping here as well, so an oscilloscope is handy. These boards do not have a gain/bias knob, unfortunately, and I am still researching mods to these specific boards, so talk of that will come in time.

With the clamp, you can turn the trimmer pot to lock down the voltage coming from the DPS as a pre-amp, thereby effecting the gain of the amplifier, which effects its output wattage. By using a current clamp, you can watch and get the output wattage close to what you want before you ever hook up a speaker. At least that is how I have thought it out to date.

After figuring out if clipping has occurred and setting the output wattage from the amps to RMS levels of the speaker, that should both limit the amps to stay within safe levels and help to manage heat from the amps through limiting how much heat they will generate, since they will not be pushing to the limits of the amp.

Also, with tuning of the tweeter, if you are measuring the SPL of the speakers after locking down the sub and the woofer, you can volume level whichever of those two is louder (meaning reduce the other to make both flat), you can then sneak up on the tweeter for volume leveling.

If this works to set it up, it would act to protect the speakers for the long-haul, as they would be limited to the lowest common denominator on output, thereby preventing me from turning it up to possibly blow my speakers.

If you guys see anything wrong with my thoughts here, let me know.
I’ll also be checking heat and output before I ever do the block mod to them to strap them to the plate.

[123toid]

I have nothing of substance to add to this discussion. But I do want to say, I am really enjoying the new notification icons. It really helps me keep up with what is going on.

---

https://www.youtube.com/123toid

[elliottdesigns]

First of all when running the chips mono (BTL), you only want to be running them 250w max on 4ohms (1% thd), if you go just a bit lower 200w gives you 0.01% thd at an 4ohm load.

150w max on 8ohms (1% thd), but 100w gives you 0.003% thd.

Further down in the spec sheet there is wattage information for the chip in stereo(SE), but only for 1% thd (and 10% but those are just plain stupid so I won’t include them): 120w for 2ohms, 98w for 3ohms, 77w for 4ohms

[ajc9988]

I will be running the 4Ohm load at 150W to match the RMS of the speaker. For the 8Ohm load of the woofer, that is 80W, so should also be fine. So, the actual distortion when running those to RMS should be minimal and is well within the wattage that these amps claim to be able to do.

So, has anyone here ever tested the FFTs of switching op-amps? Basically, these have socketed op-amps, so really easy to swap them out. I’m thinking of grabbing newer Burr Brown op-amps (TI) with better specs. Not going crazy, things that cost like $5 a pop or less, but who have better distortion and handling. Trying to figure out if worth it.

[imcokeman]

These look cool I’ve played with the $5-$15 range amps tpa 311x amps on ebay, would be interesting to see how these work, worth it if good, and not too much money to risk assuming it’s close to what it says it is =)

[Author]

Posts