Anyone Had Any Luck With TPA3116D2 Power Amp>?

Tube or solid-state, this section goes to eleven!
User avatar
J4mes
Breadboard Brother
Information
Posts: 100
Joined: 29 Feb 2024, 03:40
my favorite amplifier: Too many to choose...
Location: UK
Has thanked: 34 times
Been thanked: 26 times
Contact:

Post by J4mes »

I've had some of these in a drawer for about a year and not really thought much of it.
Image
Until I realised I could build a 100W (aside) stereo HiFi with full range 4Ohm Loads and putting a cab sim etc through it.
Problem is, it sounds very lacklustre in its overall characteristics no matter the tone shaping... But they're cheap as chips.

Has anyone used these with good results or any documentation around?

User avatar
RuptorGB
Information
Posts: 30
Joined: 30 Jun 2024, 08:52
Has thanked: 17 times
Been thanked: 10 times

Post by RuptorGB »

J4mes wrote: 30 Oct 2024, 14:04Problem is, it sounds very lacklustre in its overall characteristics no matter the tone shaping
What do you mean? Are you referring to a lack of high frequencies? They might be filtering the input to cut noise above 20 KHz. I have one of these with the same chip in a subwoofer that sounds good to me but it is low frequency so I can only say it is good for bass.
https://www.ebay.co.uk/itm/315581462788 ... BMvvmC4Ntk

User avatar
J4mes
Breadboard Brother
Information
Posts: 100
Joined: 29 Feb 2024, 03:40
my favorite amplifier: Too many to choose...
Location: UK
Has thanked: 34 times
Been thanked: 26 times
Contact:

Post by J4mes »

RuptorGB wrote: 30 Oct 2024, 15:26
J4mes wrote: 30 Oct 2024, 14:04Problem is, it sounds very lacklustre in its overall characteristics no matter the tone shaping
What do you mean? Are you referring to a lack of high frequencies? They might be filtering the input to cut noise above 20 KHz. I have one of these with the same chip in a subwoofer that sounds good to me but it is low frequency so I can only say it is good for bass.
https://www.ebay.co.uk/itm/315581462788 ... BMvvmC4Ntk
There's plenty of high end with some tweaking and with the addition of the baja presence and resonance. It just sounds dull... No life.
When you listen to some of the older SS amps from the 70's, they sound open and almost like they have a airiness about them. Class D stuff seems to be stuck to the floor. It works, but doesn't give nearly the same transients as a valve. Obviously, I'm not looking to replicate a valve here, but something that is pleasant to play through and don't feel like you need to compensate with other gear. As a foundation I guess it's good, but not nearly what you would expect a guitar amp to feel like.

I have read other threads regarding the LM3886 and how they sound great with a feedback mod. I bought two LM3886's and haven't assembled anything yet. I wouldn't mind something for practising, but even a 5 watt valve amp is too cumbersome for my needs.

User avatar
plush
Cap Cooler
Information
Posts: 679
Joined: 08 Dec 2015, 09:29
Location: Moscow, Evil Russia
Has thanked: 37 times
Been thanked: 185 times

Post by plush »

Have you measured its frequency response?

User avatar
J4mes
Breadboard Brother
Information
Posts: 100
Joined: 29 Feb 2024, 03:40
my favorite amplifier: Too many to choose...
Location: UK
Has thanked: 34 times
Been thanked: 26 times
Contact:

Post by J4mes »

plush wrote: 30 Oct 2024, 18:43 Have you measured its frequency response?
I don't want to sound thick, but I've not had an oscilloscope for years, and I don't have anything else. Could I use spectrum analyser software? I'm sure I saw somebody doing this with adequate results.

User avatar
FiveseveN
Cap Cooler
Information
Posts: 624
Joined: 25 Oct 2009, 21:20
Location: Romania
Has thanked: 227 times
Been thanked: 126 times
Contact:

Post by FiveseveN »

J4mes wrote: 30 Oct 2024, 16:41 dull, life, open, airiness, stuck to the floor
None of these things convey any actual information about the amp's performance.
But if you want to talk about transient response, I'd bet on de Forest's moustache that it would beat any transformer-coupled amp if you assess it using measurement equipment instead of vague epithets and childhood memories.
Could I use spectrum analyser software?
Yes, but something with more bandwidth than an audio interface would be better, to check for any ultrasonic funny business. Datasheet performance isn't always guaranteed with these anonymous Chinese modules.
Ignorance more frequently begets confidence than does knowledge. (Charles Darwin)

User avatar
J4mes
Breadboard Brother
Information
Posts: 100
Joined: 29 Feb 2024, 03:40
my favorite amplifier: Too many to choose...
Location: UK
Has thanked: 34 times
Been thanked: 26 times
Contact:

Post by J4mes »

FiveseveN wrote: 30 Oct 2024, 21:32
J4mes wrote: 30 Oct 2024, 16:41 dull, life, open, airiness, stuck to the floor
None of these things convey any actual information about the amp's performance.
But if you want to talk about transient response, I'd bet on de Forest's moustache that it would beat any transformer-coupled amp if you assess it using measurement equipment instead of vague epithets and childhood memories.
Could I use spectrum analyser software?
Yes, but something with more bandwidth than an audio interface would be better, to check for any ultrasonic funny business. Datasheet performance isn't always guaranteed with these anonymous Chinese modules.
It'd be great if I knew where to start! Everything I've learned in electronics has been online.
Until I learn, I'm afraid you're stuck with childhood memories and anecdotes of times gone by.

I do appreciate that you guys have been doing this for years and know it like the back of your hand. It must be frustrating when the obvious thing to do isn't so obvious to others. But surely, given your intellect, you can understand (if you've played guitar for some time) what I'm getting at.

I'm running this into a Vintage 30 8Ohm. I made a couple of cab sims (using omnicab) and tone stacks (using tonemender) etc. In the current setup I believe it's meant to be running through HiFi speakers, as with all the modules. Can you imagine what that sounds like, or do you need a full analysis?

I'm wondering if anyone has any prior experience or guidance, because my knowledge is evidently limited...

User avatar
FiveseveN
Cap Cooler
Information
Posts: 624
Joined: 25 Oct 2009, 21:20
Location: Romania
Has thanked: 227 times
Been thanked: 126 times
Contact:

Post by FiveseveN »

Alrighty, let's try to find the issue systematically then.
I'm running this into a Vintage 30 8Ohm
First of all, make sure you're not using a cabsim since you have an actual cab. The amp doesn't care (or know) if it's driving a guitar cab or full range speakers, so that shouldn't be an issue.

Have you used a different poweramp with this cab and liked the result?
Ignorance more frequently begets confidence than does knowledge. (Charles Darwin)

User avatar
plush
Cap Cooler
Information
Posts: 679
Joined: 08 Dec 2015, 09:29
Location: Moscow, Evil Russia
Has thanked: 37 times
Been thanked: 185 times

Post by plush »

J4mes wrote: 30 Oct 2024, 18:55
plush wrote: 30 Oct 2024, 18:43 Have you measured its frequency response?
I don't want to sound thick, but I've not had an oscilloscope for years, and I don't have anything else. Could I use spectrum analyser software? I'm sure I saw somebody doing this with adequate results.
Yep, can be done by using your audio interface - https://www.voxengo.com/product/span/

Configure your inputs and outputs in your DAW
Feend frequency sweep to poweramp input.
For measurement use _only_ resistive load at the power amp output. Don't even try measuring frequency response with active loads or speakers.
Don't forget to attenuate the signal amplitude you are measuring by using some resistors or a potentiometer/resistor combo. Something like 100k/10k should work.
Observe near flat frequency response and consider revisiting your circuit design skills in terms of toneshaping and stuff, yeah.
:)
And, yeah, buy a decent scope, something like Rigol DHO804, which is only 400 bucks, but packs a huge.

User avatar
J4mes
Breadboard Brother
Information
Posts: 100
Joined: 29 Feb 2024, 03:40
my favorite amplifier: Too many to choose...
Location: UK
Has thanked: 34 times
Been thanked: 26 times
Contact:

Post by J4mes »

FiveseveN wrote: 31 Oct 2024, 08:12 Alrighty, let's try to find the issue systematically then.
I'm running this into a Vintage 30 8Ohm
First of all, make sure you're not using a cabsim since you have an actual cab. The amp doesn't care (or know) if it's driving a guitar cab or full range speakers, so that shouldn't be an issue.

Have you used a different poweramp with this cab and liked the result?
I built a 5 watt el84 SE when I was about 17, and so built the cab to go with it. Sounds great, and have used the cab ever since. Any amp I've put through seems to enjoy being coupled (meaning, it sounds good). The amp itself is where I'm stumbling. Although it does sound good, and I could probably gig with it, it isn't "there" yet. When you know you know. But in my case, only my ears are telling me what's wrong.
The amp packs a punch, and I'm guessing even more so if I coupled the 2 vintage 30's I have in series. (I actually meant parallel. 4Ohms) I know doing that will change the overall feel and characteristics of the setup. It'd be good to try, but before doing any of that I'm gonna get a scope. Probably an outlay I don't need right now, but I can learn a lot with a scope. I see guys fault finding in mere minutes as apposed to hours.
Thanks for the insight and your time. It's the most precious thing we have
Last edited by J4mes on 04 Nov 2024, 13:30, edited 2 times in total.

User avatar
J4mes
Breadboard Brother
Information
Posts: 100
Joined: 29 Feb 2024, 03:40
my favorite amplifier: Too many to choose...
Location: UK
Has thanked: 34 times
Been thanked: 26 times
Contact:

Post by J4mes »

plush wrote: 31 Oct 2024, 14:18
J4mes wrote: 30 Oct 2024, 18:55
plush wrote: 30 Oct 2024, 18:43 Have you measured its frequency response?
I don't want to sound thick, but I've not had an oscilloscope for years, and I don't have anything else. Could I use spectrum analyser software? I'm sure I saw somebody doing this with adequate results.
Yep, can be done by using your audio interface - https://www.voxengo.com/product/span/

Configure your inputs and outputs in your DAW
Feend frequency sweep to poweramp input.
For measurement use _only_ resistive load at the power amp output. Don't even try measuring frequency response with active loads or speakers.
Don't forget to attenuate the signal amplitude you are measuring by using some resistors or a potentiometer/resistor combo. Something like 100k/10k should work.
Observe near flat frequency response and consider revisiting your circuit design skills in terms of toneshaping and stuff, yeah.
:)
And, yeah, buy a decent scope, something like Rigol DHO804, which is only 400 bucks, but packs a huge.
I'll give this a good read, and I've bought some books, because online learning is too sporadic. Thank you for your time and effort. It's gravely appreciated

User avatar
ppluis0
Diode Debunker
Information
Posts: 931
Joined: 14 Jul 2010, 18:33
Location: Argentina
Has thanked: 128 times
Been thanked: 311 times

Post by ppluis0 »

Hi J4mes,

Searching the datasheet for this chip (49 pages :shock: ) I have seen that to obtain 100 W in mono connection the load impedance must be 2 ohms.

If you use two 8 ohm speakers in series, you would be making it work way outside the normal range :roll:

Try to connect both speakers in paralell and see what happens.

Cheers.
Jose

User avatar
RuptorGB
Information
Posts: 30
Joined: 30 Jun 2024, 08:52
Has thanked: 17 times
Been thanked: 10 times

Post by RuptorGB »

J4mes wrote: 30 Oct 2024, 16:41It just sounds dull... No life.
Assuming you are running comparisons between amps with the same power the difference between digital & analogue outputs is the filters to knock out the digital switching so I wonder if that is what makes you hear it as dull. I am not sure why filters are required other than to give smooth perfect frequencies in the audio range but valve & analogue transistor amps don't have them. What about taking the filters out of circuit for a test? The speaker will filter out the 400 KHz switching frequency that you can't hear anyway & no output filters will not block other frequencies. If you try it let us know the result like if It causes any weird harmonics & instability or makes no difference at all.

User avatar
Frank_NH
Resistor Ronker
Information
Posts: 250
Joined: 12 Jun 2013, 14:18
Has thanked: 123 times
Been thanked: 154 times

Post by Frank_NH »

I would think that your preamp circuit would be crucial in getting a good tone since the chipamp’s job is to reproduce your input signal without any distortion or coloring of the sound.

Here’s some related info on the EHX 44 Magnum.

viewtopic.php?f=2&t=11733

User avatar
karul
Cap Cooler
Information
Posts: 530
Joined: 02 Jan 2011, 20:28
Has thanked: 639 times
Been thanked: 273 times

Post by karul »

ppluis0 wrote: 01 Nov 2024, 10:19Searching the datasheet for this chip (49 pages :shock: ) I have seen that to obtain 100 W in mono connection the load impedance must be 2 ohms.
On these tests that board with2 ohms load, freq response drop 6db down at 10kH and cannot deliver 100W

User avatar
ppluis0
Diode Debunker
Information
Posts: 931
Joined: 14 Jul 2010, 18:33
Location: Argentina
Has thanked: 128 times
Been thanked: 311 times

Post by ppluis0 »

Hi karul,

You're right.
The datasheet has an schematic in the page 26:
https://www.ti.com/lit/ds/symlink/tpa31 ... g.com%252F

There are one IC connected in bridge to drive an subwoofer and another one in stereo to feed two main speakers.

Cheers,
Jose

User avatar
karul
Cap Cooler
Information
Posts: 530
Joined: 02 Jan 2011, 20:28
Has thanked: 639 times
Been thanked: 273 times

Post by karul »

J4mes what kind of a power supply you use, what voltage? Did you tried it with different voltages? Did you try to play with that vol trimpot on the board, does it affect the sound in upper register?

This is the webpage of the manufacturer, no documentation. Best info for the IC you can get from it's datasheet.
https://www.tztstore.com/goods/show-5103.html

User avatar
J4mes
Breadboard Brother
Information
Posts: 100
Joined: 29 Feb 2024, 03:40
my favorite amplifier: Too many to choose...
Location: UK
Has thanked: 34 times
Been thanked: 26 times
Contact:

Post by J4mes »

karul wrote: 01 Nov 2024, 16:14 J4mes what kind of a power supply you use, what voltage? Did you tried it with different voltages? Did you try to play with that vol trimpot on the board, does it affect the sound in upper register?

This is the webpage of the manufacturer, no documentation. Best info for the IC you can get from it's datasheet.
https://www.tztstore.com/goods/show-5103.html
I've tried at 24v - 20v - 16v - 12v and so forth until it switched off.
The overall headroom and volume was altered as you'd expect. It got grainier and more saturated with the decrease in voltage. It does have a raunchy edge to it as it breaks up, but nothing to write home about. I'll have a go at the analysis tomorrow and report back. Also, I'll remove the output filtering on one of the boards and do some form of spectral analysis on both for reference.
The magnum44 doesn't sound to bad as with a bit of colouring, so hoping to get that sort of result. They're stupidly loud for the size of them.
If all fails I've not really lost anything. After all, these are cheap and if they work, then everyone can have a go, but my real desire is to try the lm3886 with the FB board on one of the threads here. I haven't seen (or heard) any demo's of said circuit, but I take your guys word for it.
I'll keep you posted

User avatar
HENK
Breadboard Brother
Information
Posts: 51
Joined: 05 Sep 2008, 17:15
Has thanked: 312 times
Been thanked: 6 times

Post by HENK »

Over at Pedalpcb-Forum there is a Tread about TPA3118/3116.
Your Device seems to be a Fake-no TPA3116-Chip.

User avatar
J4mes
Breadboard Brother
Information
Posts: 100
Joined: 29 Feb 2024, 03:40
my favorite amplifier: Too many to choose...
Location: UK
Has thanked: 34 times
Been thanked: 26 times
Contact:

Post by J4mes »

HENK wrote: 02 Nov 2024, 11:11 Over at Pedalpcb-Forum there is a Tread about TPA3118/3116.
Your Device seems to be a Fake-no TPA3116-Chip.
So what's being used instead?

Post Reply