TI TPA3100D2, Amazing Class D Amp

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I guess things depend on your design, when you say this part needs a I/P transformer. I believe that it is an unnecessary expense.

I can see that perspective. The amp is so cheap and an no doubt function well without. Decent transformers would increase costs substantially.

I am less concerned about budget sanity, as I have spare transformers and this approach appeals to me (for some reason!).

A nice solution may be to pair the amp with a Vout dac with transformer output stage (there are many solutions of this kind described in the forum). As long as the secondaries were isolated from ground the amp could be connected directly without coupling caps. Of course this may necessitate digital volume control, which raises its own issues.

Looking forward to playing with the board.
 
I have heard that a transformer as a I/V sounds great, something to try out. Nothing wrong with experimenting.

IIRC, TPA3110 low F response is also a function of gain. I use min gain setting and make up the gain in the pre-stage. Maybe this is why you use 10u?

Something I should do in my setup menu, to select TPA3110 gain settings and listen for a diff. I do it for the MAX9729ETI- 3 channel audio selector, volume control, headphone amp.
Have fun
 
Last edited:
Founder of XSA-Labs
Joined 2012
Paid Member
I just received my Sure 3110 board from Parts Express. The thing is tiny and amazingly sounds very good powered by 12v 2amp SMPS wall wart from the Lepai T2020. I am very impressed by the minimal components used (no output inductors!) and the nice and convenient screw terminals for power input and speaker output. Nice thru hole solder pads are available for audio input direct connection if you want to bypass the 3.5mm jack. The input gain is good too as I am able to drive it with usual sources and not need a preamp. Very happy with it and for $10 the value is there. The sound is similar to the very nice 3116 but with less power.
What are good mods to try with this board? Upgrade main power rail electrolytic?
 
...
What are good mods to try with this board? Upgrade main power rail electrolytic?

Yep, do the PS caps. I've tried Pan FM 470 uf/25V, Pan HFQ 1000uf / 25V and Pan FC 2200 uf/ 25V (for bass feeding a pro 15" through a 10mH coil!) :eek:

I grounded the gain pad under R4 on one of mine but just moved R4 on another to lower the gain to 20dB (looks like you can ground it or leave it floating as mentioned in the TPA31XX datasheet)

TPA31XX gain.JPG

TPA3118D2 Evaluation Module - TPA3118D2EVM - TI Tool Folder


Next will be C1 and C2 (L+, L- input caps) and C10 and C11 (R+, R- input caps)
 
Last edited:
I've just assembled a basic power amp with the Sure TPA3110D2 board, using an external medical grade 100W 12V smps. I removed the input coupling caps from the board (C1, C2, C10 & C11) and wired the + & - connections for each channel directly to a pair of ungrounded XLR input sockets.

I'm driving the amp from the balanced outputs of my Music First Classic v2 with ground lift engaged (so the transformer output is floating). Sounds very nice so far. Less weighty perhaps than the ACA it replaced but possibly more dynamic. Clean and smooth. Less obviously class-d than the tripath amps I've used before. The TDA3110D2 with 12V PSU is powerful enough for my Tannoy Little Reds at normal volumes.
 

Attachments

  • image.jpg
    image.jpg
    563.9 KB · Views: 656
Hey everyone,

I am late to this post. I'm having trouble with a TPA3123 amp. Very similar, or identical to these ones on eBay:

http://www.ebay.com/itm/TPA3123-Cla...pt=Vintage_Electronics_R2&hash=item338678cf52

I got my $11 or whatever refunded from the seller, but still out of curiosity and sheer Welsh stubbornness I'd like to get this going.

Trouble is this: it distorts, but usually only piano sounds. Other than that, sounds wonderful. And of course it's intermittent.

I re-flowed every solder joint I could think of, but no change.

Any thoughts on what to do? Possible swap out certain capacitors. Most of the rest of the components are surface mounted devices and of course very difficult to mess with.

Mark
 
Last edited:
We'll need more details Mark, such as what are the PS voltages, how efficient are your speakers, etc..

Hey everyone,

I am late to this post. I'm having trouble with a TPA3123 amp. Very similar, or identical to these ones on eBay:

TPA3123 Class D Digital HiFi Mini Assembled Amplifier Board 25W 2 | eBay

I got my $11 or whatever refunded from the seller, but still out of curiosity and sheer Welsh stubbornness I'd like to get this going.

Trouble is this: it distorts, but usually only piano sounds. Other than that, sounds wonderful. And of course it's intermittent.

I re-flowed every solder joint I could think of, but no change.

Any thoughts on what to do? Possible swap out certain capacitors. Most of the rest of the components are surface mounted devices and of course very difficult to mess with.

Mark
 
Founder of XSA-Labs
Joined 2012
Paid Member
Mark,
When you say it distorts piano sounds do you mean harmonic distortion or clipping? Not sure why it only happens on piano sounds? Have you double checked the inductor value used with TI's recommended value? It looks like a 22uH and maybe factory recommended is 10 uH in which case, you may be filtering too much of the HF.
X
 
Hello there X and Mr. Slim,
Thanks for replying.

Well, no, I am just using a board already built so I assumed the values of the inductors were fine. I notice all the ones on eBay are using the same value: 22uH.

I'm not sure I know the answer to your question about clipping vs. harmonic distortion. As in, I'm not sure I know the difference. To me, it just sounds strange/distorted and so far I've noticed it on piano sounds. Bass seems fine and higher sound seem fine.

Very weird. And not piano sound is distorted. Only some.

I don't know if changing out the large 1,000uF value caps with others might help or not.

Mark
 
Mr. Slim:

I run Klipsch La Scalas.

Various power supplies were tried: but both were 12 switch mode types. 2 amp.

Mark

Ok, so the La Scalas are plenty efficient (105db@1watt), so it shouldn't take much power to play them loud. According to the spec sheet, page 8, figure 11, given a supply voltage of 12 Volts, and 8 ohms (an assumption that the La Scalas are 8ohms...) then at 1% distortion, you would get an output of slightly over 2 watts, with the distortion rising to 10% at just under 3 watts. Is the sound less "distorted" at lower volumes? I recall hearing a story about David Hafler attempting to reproduce a piano and was surprised at how much power headroom was required to be able to produce an unclipped soundwave. According to page 15 of the spec sheet, the 22uH inductors are recommended for 4 ohm speakers, so that may also be an issue.

If you have access to one, it might be worth trying the amp with 24V power supply, which would give you about 9 watts output, at least 3 times more than you are getting with a 12 V supply
 
My first guess would be - the inductors, whilst being the right value aren't man enough in current handling for the job and are saturating. I've noticed other incarnations of TI amp chips are using toroids - they're much less likely to distort.

<edit> Eyeballing the dimensions, the inductors look to be 12mm, hence Coilcraft MSS1210. But Coilcraft are expensive in China, relatively speaking so they're almost certainly cheap Chinese copies. If you can get hold of the real thing i reckon your distortion will go away.
 
Last edited:
Hi there fellows,

Hmmm. Well, trying a 24VDC (cage type) power supply is certainly a possibility.
I have one on loan to my brother in law. He's trying out a Sure Electronics TK-2050 amp of mine.

The thing is: I never ran the amp in question at high volume levels. Levels in my small room are generally what I'd call "loud conversational", or even "conversational", which shouldn't tax this amp at all.

If my amp is defective then every single one being sold is too. It's just some $11 cheapie off eBay.

Yeah, you would think the bass would distort, but that seems fine. Just certain piano sounds, and at certain times. Not all the time.

I built a Jerry's Electronics amp which uses a similar chip (TPA3122D2) and that one sounds terrific. (I use just a 12vdc, switch mode, 2 amp, computer type power supply on that one as well.) I have two TPA7297 amps and those sound terrific as well. Same power supply. Same everything in the room, etc.

The bottom line is: If you guys don't think it's the large, 1000uF caps or something else I can swap out readily - I'll throw it in the trash and forget about it.

Mark
 
Founder of XSA-Labs
Joined 2012
Paid Member
Mark,
It looks like 22uH is correct size inductor per factory spec. It might be current saturation as Abraxalito says. I think it may be that the power caps are not big enough - what size are they anyway? You need at least 470uF per side and preferably double that to handle high current transients. If you have some 4 ohm speakers around try those and see of it gets worse or improves. I think it may just be power headroom limit of this amp. If you haven't already ordered the $17 tpa3116d2 from China - that may fix your problem with 50 watts.
 
magnet wire

I have magnet wire in several gauges. I'll have to look when I get home of what exact gauges I have in the parts box; but 24AWG, 26AWG, and 30AWG come to mind.

I could wind some inductors and replace the ones on this board if:

1) Some one would instruct me on how to do that. How to determine the proper value, etc.

2) If you guys feel it would work and be worth the effort.

Thanks,
Mark
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.