TPA3116D2 Amp

Try a 19 volt laptop brick. It will make it more musical as you say. I think there is just more power and the higher voltage reduces HD.

I didn't have a spare 19V laptop brick kicking around, but I did have a "clone" of the Astron RS-4 regulated linear power supply @ 13.8V/3amp. Some wire cutters and a couple banana plugs and I was all set. I think it did smooth out the SMSL TPA3118 a little bit compared to the 12V SMPS the unit shipped with.

BK
 
Another option instead of relays, I use a MAX9729 part, it works great, it can be used as a channel selector, bass boost, headphone amp that works great as a pre-amp for the TPA3100D2 that I used before the TPA3116 parts were available.

Cheers
Rick

No semiconductor on this planet can compete with relays for analog signal switching when signal integrity/sound quality is concerned. Manufacturers do their best to make their chips work great but the reason such chips are made is that they're way cheaper than relays and rotary switches. Society is cost driven, you do the math.

Why starting with industrial compromises as cost, board space etc. when DIYing ? The only somewhat valid points would be cost (as usual) and the possibility to drive stuff with microcontrollers/PICs/cell phones/apps/software and such convenience/"features"/bling bling stuff. In my words:"the blue LED syndrome". Almost all of them are contradictory to good sound quality (if that is the goal of course, I slowly start to understand that some might think that quality is not that important).
 
Last edited:
Jean-Paul, relax, take a chill pill or smoke a "J"!!
No semiconductor on this planet can compete with relays for analog signal switching when signal integrity/sound quality is concerned.
This is not the point that I am making. I offer a cheap hi-perforrmance solution that fits my requirements and maybe others. If you have golden ears then go with your relay solution, but this thread is discussing a class "D" amplifier, so you are basically shooting yourself in the foot in the ultra hi-performance front.
I wish some folks would not be too overly critical of component selection and actual listen to a design before they shoot it down in flames. It is akin to the same thinking of comparing a class "D" amp to say a well designed class "A" or "AB" amp. The whole premise of this thread is to discuss the sound quality of a class "D" amplifier. BTW I design both class "AB" and "D" type amplifiers!!
I measured the 9729 as 0.03% THD, as is documented, in my prototype and it works great, so in your case, all I can say go with mechanical parts if they suit your design requirements. I personally do not want to have to bother with pots/relays that wear out in time and when they wear out, they are obsolete.
The only somewhat valid points would be cost
If that is all you can come up with, then you are myopic.

Is there any specific change need to make on the MAX9792 circuit if use it as a pre-amp?
I have done nothing special, just followed the application information. I did use a 3.5mm jack with a switch so that when I plugged in the headphones, it interrupted the signal flow to the TPA3100D2.

My portable radio/media player design use these key components:
TPA3100D2
MAX9729
Si4735D60
PCM2902B
BU9458
Atxmega256a3
 
Last edited:
Well, I am as relaxed as can be and it is not my solution but Danzz designed a nice one. Just compare semiconductor volume control and selectors to analog equivalents en hear the difference. THD is not everything ! I have heard quite some semiconductor selector solutions and relays win every time hands down. Besides that signal (telecom) relays with gold plated contacts don't wear out in consumer use as they don't switch current. They are already available for 30 years or more and will be available for quite some years too. The difference being that current types use a lot less power to operate. Till now I have never had the occasion that relays and potentiometers or rotary switches have become irreplaceable opposite various semiconductor solutions that have meanwhile become obsolete. Please do not try to "win" a discussion with shaky arguments.

I don't see the point of class D being inferior to other amps. So if it is class D then any preamp will do ? Then why bother with class D at all ? Just to be busy waisting time on inferior amps as they're cheap or something like that ?

You took my words concerning "valid points" out of context by omitting the rest of the sentence so excuse me for not even trying to find the meaning of "myopic".

I think the board Danzz designed will outperform any semiconductor solution by a margin at reasonable cost. He just needs to add mounting holes 🙂
 
Last edited:
Well... May I suggest to add a coffee machine 🙂 To read the thousand posts of this thread, coffee is a requirement.

Now enjoy the coffee with my new TPA Amp. :cheers::drink:
 

Attachments

  • 20140724_000040.jpg
    20140724_000040.jpg
    223.1 KB · Views: 659
Ultimate Version 🙂
Below Features:
- More Inductor Footprint Support e.g SAGAMI 7G Series, SMT and Choke power inductors
- ALL TH Parts
- S/D PIN
- MUTE PIN
- SYNC PIN
- LED Power Indicator
- BTL and PBTL Option
- Anti-Pop Speaker Protection
- Two Big ECaps Footprint (Dia. 18.2mm, up to 6800uF/35V for Bass)
- Two Axial Caps Footprint (30x13.5mm, more replacement option
- Board Size: 65mm x 100mm

TPA3116_ULTIMATE.JPG


My Board come out. (This is Anti-Pop design)
- ALL TH Parts
- Use DALE Axial Resistors
- Use different types of Film caps and Ecaps with power inductors

Combination I
WIMA Cap 1uF/250V x2 , 1uF/100V x 3 and Nichicon Fine Gold 1000uF/35V x 2, SAGAMI Power Inductor 10uH

TPA3116_SPK_V1_1.jpg


TPA3116_SPK_V1_2.jpg


TPA3116_SPK_V1_3.jpg


TPA3116_SPK_V1_4.jpg



Combination 2
All Siemens Film Caps and Nichincon 4700uF/35 x 2, SMT Inductor 10uH
TPA3116_SPK_V2_1.jpg


TPA3116_SPK_V2_2.jpg


Combination 3
Power Coil

TPA3116_SPK_V3_1.jpg


TPA3116_SPK_V3_2.jpg


Volume Pot. With StandBy Switch.
TPA3116_SPK_LINK.jpg
 
Last edited:
Guys, I just had time to install the bootstrap snubber circuit and it works!

Well, I just done the job too...
For me the improvement is not as expected.
My configuration:
YJ blue board
Burns inductors
tdk bootstrap capacitors
wima input dc
changed dc power caps
bootstrap snubber circuit (330pf, 5%, 100v, 10ohm resistor)

speakers are Ditton 4 92db
The kind of 'hiss' on voices is always present, despite I tried other speakers, But I had improvements by changing the bootstrap capacitors

For me, the best mod was to change the bootstrap capacitors by tdk capacitors. It has improved the spacial feeling.
Wima capacitors and dc power caps improved the bass.

I am happy with this small chip. It sounds very good for the price.


Christian.
 
Last edited:
Well, I just done the job too...
For me the improvement is not as expected.
My configuration:
YJ blue board
Burns inductors
tdk bootstrap capacitors
wima input dc
changed dc power caps
bootstrap snubber circuit (330pf, 5%, 100v, 10ohm resistor)

speakers are Ditton 4 92db
The kind of 'hiss' on voices is always present, despite I tried other speakers, But I had improvements by changing the bootstrap capacitors

For me, the best mod was to change the bootstrap capacitors by tdk capacitors. It has improved the spacial feeling.
Wima capacitors and dc power caps improved the bass.

I am happy with this small chip. It sounds very good for the price.


Christian.

Hmm... it may be combination of capacitors and resistors that you have on post inductor? I know that they parts vary from design. I have 560 nF instead of 680 nF and 5.6 ohms instead of 3.3 ohms. I am using stock bootstrap film caps too. Anyhow, it may be that your speakers have a very sensitive high end.
 
My Board come out. (This is Anti-Pop design)
- ALL TH Parts
- Use DALE Axial Resistors
- Use different types of Film caps and Ecaps with power inductors

Combination I
WIMA Cap 1uF/250V x2 , 1uF/100V x 3 and Nichicon Fine Gold 1000uF/35V x 2, SAGAMI Power Inductor 10uH

TPA3116_SPK_V1_1.jpg


TPA3116_SPK_V1_2.jpg


TPA3116_SPK_V1_3.jpg


TPA3116_SPK_V1_4.jpg



Combination 2
All Siemens Film Caps and Nichincon 4700uF/35 x 2, SMT Inductor 10uH
TPA3116_SPK_V2_1.jpg


TPA3116_SPK_V2_2.jpg


Combination 3
Power Coil

TPA3116_SPK_V3_1.jpg


TPA3116_SPK_V3_2.jpg


Volume Pot. With StandBy Switch.
TPA3116_SPK_LINK.jpg

Hiamp,

Very nice board. Are these available through your EBay shop yet?

Regards,
 
It's about time Sure made one of these... they make nice stuff. I really like their $10 TPA3110D2 board. Now with pop suppression, and with built in RCA input jacks. Pretty nice.

Maybe this is off topic, but about Sure's $10 TPA3110D2 board... what's the reason it's marked 2x8W at 4 ohms? For most of the boards including for the TPA3116, they usually just quote chip specs for performance characteristics (with I suppose no guarantee the actual boards achieve exactly that in practice), but the $10 TPA3110 product seems to be particular about the output wattage, which is lower than what the chip can do in other circumstances.

What's missing or underspec or whatever to limit it like that? Is there an 8V regulator or something clamping the supply rail?

edit: oh, the chip has a PLimit functionality that's externally set. It's probably limited by that. I guess that could be modded, but at that point in the very least the caps look undersized and the minimal filter more concerning, so that's probably not worth it.
 
Last edited:
Maybe this is off topic, but about Sure's $10 TPA3110D2 board... what's the reason it's marked 2x8W at 4 ohms? For most of the boards including for the TPA3116, they usually just quote chip specs for performance characteristics (with I suppose no guarantee the actual boards achieve exactly that in practice), but the $10 TPA3110 product seems to be particular about the output wattage, which is lower than what the chip can do in other circumstances.

What's missing or underspec or whatever to limit it like that? Is there an 8V regulator or something clamping the supply rail?

edit: oh, the chip has a PLimit functionality that's externally set. It's probably limited by that. I guess that could be modded, but at that point in the very least the caps look undersized and the minimal filter more concerning, so that's probably not worth it.

Difference is somewhere in other parts, compare max Voltage, they also offer a 15w version that has the parts to be powered by TI voltage range. 2x8 board comes without Plimit.
 
I have run the Sure 3110 board with a 19v laptop supply and it works fine - I think you can get upwards of 20 watts out of it. They are just being conservative so that this amp doesn't step on the toes of their other offerings. The 8 watt spec is probably based on 12v into 8 ohms which is about right for low distortion.