Possibility ₹1K worth TPA3116D2 Amplifier with ₹30K worth of good speakers

@somakmaitra Your words encourages me to try with some other power supply. I am digging out my PSU collection to find one. Let's see the outcome. Fun world!
 
I think it was meant to be quality of power, not qty.
Thanks @rikhav , I understood the same. That is why I want to try with some other power supply of good quality too. Apart from that just now I came to know that it is recommended to use Min 24V 5 Amps power supply for 2.0 setup and 7 Amps for 2.1 setup as per TI support. I have never tried with any SMPS of 24V and 5 Amps capacity. 7.5 Amps is the over current trip point for TPA3116D2. Here is the TI Support page discussion.
 
There is a 100 rupee tpa3118 mono board and do the modifications in this threadhttps://www.hifivision.com/threads/improving-tpa3118-class-d-amplifier.65258/
use a laptop power supply you will get a nice sounding amp . I have tried it and results are good . And it has a switch to mute turn it on before turning off the amp so there will be no pop . Give it a try total cost will be less than 1k
 
There are plenty of supposedly high end brands that use similar class D chips
e.g I have a Paradigm Link Amp that uses a TDA8954

I also use a TPA3116D2 and a TAS5630 board - I like them quite a bit especially because they can be tucked away neatly for setups where you can't plug in a big-a## amp due to aesthetic considerations. Also like the fact that the power draw at idle is so small that you can leave them always-on

The only difference I could make out between all of these (and also regular Analog amplifiers and AVR amps) is that the boards have a little bit of static hiss even with no input (although the Paradigm does not have a hiss)
The TPA3116D2 will distort at high levels but is damn near perfect at 50-60 db levels

All the speakers/boards/amps etc compared are in my sig and I have tried almost all possible permutations and combinations of them

As mentioned by other FMs, the key difference is the quality of the PSU and to a lesser extent, the quality of the passive components on the board.
I have got very good results with a 24V 5 Amp grounded SMPS that I snagged off amazon for a little over 1K
Given the low cost of admission, I'd say get a good PSU and if you hear what you like, move on the next phase of replacing the passive components
 
There are plenty of supposedly high end brands that use similar class D chips
e.g I have a Paradigm Link Amp that uses a TDA8954

I also use a TPA3116D2 and a TAS5630 board - I like them quite a bit especially because they can be tucked away neatly for setups where you can't plug in a big-a## amp due to aesthetic considerations. Also like the fact that the power draw at idle is so small that you can leave them always-on

The only difference I could make out between all of these (and also regular Analog amplifiers and AVR amps) is that the boards have a little bit of static hiss even with no input (although the Paradigm does not have a hiss)
The TPA3116D2 will distort at high levels but is damn near perfect at 50-60 db levels

All the speakers/boards/amps etc compared are in my sig and I have tried almost all possible permutations and combinations of them

As mentioned by other FMs, the key difference is the quality of the PSU and to a lesser extent, the quality of the passive components on the board.
I have got very good results with a 24V 5 Amp grounded SMPS that I snagged off amazon for a little over 1K
Given the low cost of admission, I'd say get a good PSU and if you hear what you like, move on the next phase of replacing the passive components
Thanks. Exactly what I am doing now. Before give up, I just ordered 24v 5Amps PSU. By this weekend I should be able to test this Amp board with PSU ordered. As you said if it is decent enough I can tweak further like passive components replacement like capacitor and pop sound arresting if any.
 
Thanks. Exactly what I am doing now. Before give up, I just ordered 24v 5Amps PSU. By this weekend I should be able to test this Amp board with PSU ordered. As you said if it is decent enough I can tweak further like passive components replacement like capacitor and pop sound arresting if any.
Search for an OEM laptop charger with same power ratings. I'm using one from Lenovo for my INSMA board and cheapo speakers and they sound very good for the money and purpose of use.
 
Search for an OEM laptop charger with same power ratings. I'm using one from Lenovo for my INSMA board and cheapo speakers and they sound very good for the money and purpose of use.
Laptop chargers are reasonably clean but will typically be at 19V and peak out beyond 2 Amp (and that's pushing it)
Good enough for low volume levels but will cause clipping at moderate to high gains levels
 
...and peak out beyond 2 Amp (and that's pushing it)
...

Not true.

I bought two laptop chargers for my TPA3118 amps and they're both 4.5A, 19.5V. That's nearly 90W of power and enough for the 60W rating of each TPA3118 board.
 
Not true.

I bought two laptop chargers for my TPA3118 amps and they're both 4.5A, 19.5V. That's nearly 90W of power and enough for the 60W rating of each TPA3118 board.
They seem to be well rated - Where did you get them from?
There is actually a very easy way to test if they are actually performing at paper rated specs

Connect any decent DMM on the V out terminals - your reading should be be between 19.3-19.7V
Now start audio playback of a bass rich track and gradually increase the volume.
The volume levels (during bass rich sections) at which your reading starts dipping beyond 3-4% (i.e. 96-97% of the original reading) is the maximum volume you should be driving the amp so as to keep your speakers safe
 
Laptop chargers are reasonably clean but will typically be at 19V and peak out beyond 2 Amp (and that's pushing it)
Good enough for low volume levels but will cause clipping at moderate to high gains levels
Mine is 20v 4.5 amps :)
 
Let me put in my last year's experience
I started off with Nobsound NS-01 amplifier from AliExpress(TPA36112 amplifier) to drive my DIY speakers(JBL Coaxial drivers CS762)
I paired it with 3 types of power supply, while keeping volume at source to 70% :
  1. 12V3A wall adapter- using this, going beyond 11 O'clock on volume knob distorted sound unbearably (6 O'clock being Zero)
  2. 65W laptop charger(19.5V, 3.34A )- this "opened" up the amplifier. Now I was able to push upto 3 O'clock before sound distorted
  3. 90W laptop charger(19.5V4.62A)- this became the power supply for the amplifier. Marginal power increase was transferred to level control as well as now the peak was at almost 4 O'clock, 5 O'clock being the last position
As increasing power increased output level, I tried running it with 24V5A SMPS too. But in this case, the gain wasn't noticeable while the amplifier got hot more than ever before. As I needed a 2.1 solution, the setup of speakers running on Nobsound amp and subwoofer through car amp(powered by Computer PSU) was too much of a hassle, I switched a 2.1 amplifier, still TPA3116 based. It's specs looked promising and it was cheap enough to be experimented with.
With this amplifier there were 2 issues:
1. It's "Treble" knob at it's minimum setting and it's "Bass" knob at maximum setting used to cut audio from the speakers and rather a sound used to come out of the speakers, kinda of like when you are tuning AM radio stations. After lot of reading around, found out the problem lies in the board design. The "knobs" were 50K Pots and fix suggested using 10K Pots.
2. When bass heavy track was played like 'Bass I love You', as soon as Bass would kick in at volume levels greater than 60%, the signal would cut out and resume momentarily when the bass part was over. It was like as if the amplifier was going into "Protect" mode. My experience from Car audio suggested me it's a power issue as the amp was hooked on to a 90W laptop charger. Thus I got the 24V5A PSU. Still no dice.
After researching on this issue, eventually found out that output stage capacitors weren't powerful to drive the signal home. Swap of capacitors was suggested.

But I didn't want to mess around with SMD boards and had other priorities(Studies) so I packed it up.
Finally I after reading around and discussing on DIY forums I got this. It's a TPA3255 based board and I'm powering it with a Meanwell PSU This ensured that this amplifier isn't going to be starved of power at any moment. Audio output, connectivity, channel separation, everything is amazing. It drives those DIY speakers(2x45W@4ohms) and DIY Subwoofer(2x180W@8ohms, wired in parallel so effective impedance that amplifier bass channel sees is 4ohms) effortlessly. Even after running for 8-10 hours continuously, it gets just warm. There is a static hiss when nothing is playing but that's audible when you actually put your ear to the drivers.
But it too has it's fault and that is the Speaker Pop when switched On. It's considerable on L-R channel, so much so that even if volume/gain pot turned all the way down, it's almost 50dB Pop. If by mistake the subwoofer gain pot is anywhere above 0, it's like a gunshot going off. Solution to this is a speaker protection circuit but my jugaad involves using 2 switches, one each for each channel on the Positive line. The speakers are turned "off" using the switch and after powering On the amplifier, the speakers are switched "on". It's not the most elegant solution but works for the moment.

While power delivery especially in terms of bass has been good with these Class D amplifiers, the Highs have a plastic feel to them, a kind of shrillness. It's not fatiguing but the soundstage is hollow. The Mids are almost absent. Maybe it's my drivers, maybe it's the amplifier but I know what I have to do, move on from Class D.
 
Let me put in my last year's experience
I started off with Nobsound NS-01 amplifier from AliExpress(TPA36112 amplifier) to drive my DIY speakers(JBL Coaxial drivers CS762)
I paired it with 3 types of power supply, while keeping volume at source to 70% :
  1. 12V3A wall adapter- using this, going beyond 11 O'clock on volume knob distorted sound unbearably (6 O'clock being Zero)
  2. 65W laptop charger(19.5V, 3.34A )- this "opened" up the amplifier. Now I was able to push upto 3 O'clock before sound distorted
  3. 90W laptop charger(19.5V4.62A)- this became the power supply for the amplifier. Marginal power increase was transferred to level control as well as now the peak was at almost 4 O'clock, 5 O'clock being the last position
As increasing power increased output level, I tried running it with 24V5A SMPS too. But in this case, the gain wasn't noticeable while the amplifier got hot more than ever before. As I needed a 2.1 solution, the setup of speakers running on Nobsound amp and subwoofer through car amp(powered by Computer PSU) was too much of a hassle, I switched a 2.1 amplifier, still TPA3116 based. It's specs looked promising and it was cheap enough to be experimented with.
With this amplifier there were 2 issues:
1. It's "Treble" knob at it's minimum setting and it's "Bass" knob at maximum setting used to cut audio from the speakers and rather a sound used to come out of the speakers, kinda of like when you are tuning AM radio stations. After lot of reading around, found out the problem lies in the board design. The "knobs" were 50K Pots and fix suggested using 10K Pots.
2. When bass heavy track was played like 'Bass I love You', as soon as Bass would kick in at volume levels greater than 60%, the signal would cut out and resume momentarily when the bass part was over. It was like as if the amplifier was going into "Protect" mode. My experience from Car audio suggested me it's a power issue as the amp was hooked on to a 90W laptop charger. Thus I got the 24V5A PSU. Still no dice.
After researching on this issue, eventually found out that output stage capacitors weren't powerful to drive the signal home. Swap of capacitors was suggested.

But I didn't want to mess around with SMD boards and had other priorities(Studies) so I packed it up.
Finally I after reading around and discussing on DIY forums I got this. It's a TPA3255 based board and I'm powering it with a Meanwell PSU This ensured that this amplifier isn't going to be starved of power at any moment. Audio output, connectivity, channel separation, everything is amazing. It drives those DIY speakers(2x45W@4ohms) and DIY Subwoofer(2x180W@8ohms, wired in parallel so effective impedance that amplifier bass channel sees is 4ohms) effortlessly. Even after running for 8-10 hours continuously, it gets just warm. There is a static hiss when nothing is playing but that's audible when you actually put your ear to the drivers.
But it too has it's fault and that is the Speaker Pop when switched On. It's considerable on L-R channel, so much so that even if volume/gain pot turned all the way down, it's almost 50dB Pop. If by mistake the subwoofer gain pot is anywhere above 0, it's like a gunshot going off. Solution to this is a speaker protection circuit but my jugaad involves using 2 switches, one each for each channel on the Positive line. The speakers are turned "off" using the switch and after powering On the amplifier, the speakers are switched "on". It's not the most elegant solution but works for the moment.

While power delivery especially in terms of bass has been good with these Class D amplifiers, the Highs have a plastic feel to them, a kind of shrillness. It's not fatiguing but the soundstage is hollow. The Mids are almost absent. Maybe it's my drivers, maybe it's the amplifier but I know what I have to do, move on from Class D.
Awesome! Thanks for detailed experience report. It will be helpful for me.
 
Let me put in my last year's experience
I started off with Nobsound NS-01 amplifier from AliExpress(TPA36112 amplifier) to drive my DIY speakers(JBL Coaxial drivers CS762)
I paired it with 3 types of power supply, while keeping volume at source to 70% :
  1. 12V3A wall adapter- using this, going beyond 11 O'clock on volume knob distorted sound unbearably (6 O'clock being Zero)
  2. 65W laptop charger(19.5V, 3.34A )- this "opened" up the amplifier. Now I was able to push upto 3 O'clock before sound distorted
  3. 90W laptop charger(19.5V4.62A)- this became the power supply for the amplifier. Marginal power increase was transferred to level control as well as now the peak was at almost 4 O'clock, 5 O'clock being the last position
As increasing power increased output level, I tried running it with 24V5A SMPS too. But in this case, the gain wasn't noticeable while the amplifier got hot more than ever before. As I needed a 2.1 solution, the setup of speakers running on Nobsound amp and subwoofer through car amp(powered by Computer PSU) was too much of a hassle, I switched a 2.1 amplifier, still TPA3116 based. It's specs looked promising and it was cheap enough to be experimented with.
With this amplifier there were 2 issues:
1. It's "Treble" knob at it's minimum setting and it's "Bass" knob at maximum setting used to cut audio from the speakers and rather a sound used to come out of the speakers, kinda of like when you are tuning AM radio stations. After lot of reading around, found out the problem lies in the board design. The "knobs" were 50K Pots and fix suggested using 10K Pots.
2. When bass heavy track was played like 'Bass I love You', as soon as Bass would kick in at volume levels greater than 60%, the signal would cut out and resume momentarily when the bass part was over. It was like as if the amplifier was going into "Protect" mode. My experience from Car audio suggested me it's a power issue as the amp was hooked on to a 90W laptop charger. Thus I got the 24V5A PSU. Still no dice.
After researching on this issue, eventually found out that output stage capacitors weren't powerful to drive the signal home. Swap of capacitors was suggested.

But I didn't want to mess around with SMD boards and had other priorities(Studies) so I packed it up.
Finally I after reading around and discussing on DIY forums I got this. It's a TPA3255 based board and I'm powering it with a Meanwell PSU This ensured that this amplifier isn't going to be starved of power at any moment. Audio output, connectivity, channel separation, everything is amazing. It drives those DIY speakers(2x45W@4ohms) and DIY Subwoofer(2x180W@8ohms, wired in parallel so effective impedance that amplifier bass channel sees is 4ohms) effortlessly. Even after running for 8-10 hours continuously, it gets just warm. There is a static hiss when nothing is playing but that's audible when you actually put your ear to the drivers.
But it too has it's fault and that is the Speaker Pop when switched On. It's considerable on L-R channel, so much so that even if volume/gain pot turned all the way down, it's almost 50dB Pop. If by mistake the subwoofer gain pot is anywhere above 0, it's like a gunshot going off. Solution to this is a speaker protection circuit but my jugaad involves using 2 switches, one each for each channel on the Positive line. The speakers are turned "off" using the switch and after powering On the amplifier, the speakers are switched "on". It's not the most elegant solution but works for the moment.

While power delivery especially in terms of bass has been good with these Class D amplifiers, the Highs have a plastic feel to them, a kind of shrillness. It's not fatiguing but the soundstage is hollow. The Mids are almost absent. Maybe it's my drivers, maybe it's the amplifier but I know what I have to do, move on from Class D.

Hi @Rud31 ,
1. Since I am not going for Sub, Laptop power supply of Output Voltage/Amps: 19V 4.74A, Output Watt: 90W should be good enough, right? Please confirm.
2. I am planning to connect 2.0 speakers only. But my Amp has L/R/Sub outputs. Having Sub out in Amp, will I lose all LOW in L/R speakers? (My speakers response starts at 50Hz). One of the comment I read in amazon for this product worries me and stops me from investing 1K to 1.5K on powersupply. Here is it -
"It's not a true 2.1. The subwoofer crossover is set too high. It's close to 1khz!!!. I can hear Male vocals and saxophone sounds through my subwoofer. In a true 2.1 configuration one should not be able to locate the sound of the subwoofer. Ideal cut off frequency should be 100Hz."
Please advice.
 
Hi @Rud31 ,
1. Since I am not going for Sub, Laptop power supply of Output Voltage/Amps: 19V 4.74A, Output Watt: 90W should be good enough, right? Please confirm.
2. I am planning to connect 2.0 speakers only. But my Amp has L/R/Sub outputs. Having Sub out in Amp, will I lose all LOW in L/R speakers? (My speakers response starts at 50Hz). One of the comment I read in amazon for this product worries me and stops me from investing 1K to 1.5K on powersupply. Here is it -

Please advice.
TL;DR
  1. Yes
  2. You will but check what the High Pass filter is set at to see how much.
Detailed version
Do not look at power ratings of amplifiers and power supplies as is. Remember the demon lurking in the shadow called Loss, be it thermal or transmission. Using 90W power supply won't equate to 90W output(45W per channel) even if the amp is advertised as 50W per channel for L-R. Even though class D amps are efficient but they are not 100% efficient. That's Holy Grail of modern electronics. Moreover consider the fact that it's a Chinese implementation of class D design with aim of mass production so quality and efficiency are the last items on their agenda, if they are even there. 90W laptop power supply will suffice depending upon:
1. Efficiency/Sensitivity of speakers. If the speakers are efficient, even low amount of power would be enough to drive them to higher sound levels. Taking the worst case scenario with some hypothetical value range, let's say you get speakers with sensitivity of 86dB and the amp actually delivers 25Watts per channel, then you can theoretically drive them to almost 98-99dB level but this would also mean that distortion also increases(which it will in case of these cheap amps) Add to that the fact that audio level drop in an acoustically treated room is also almost 3dB per meter. So your listening position will also matter
2. As high pitched frequencies give the impression of loudness with less power input, these cheap amps focus more on Highs, to make them sound "loud and powerful". So the crossover is usually set high and the frequency roll off is also sloppy. To find out what's the scenario in the case of your amplifier, do the following:-
  1. Connect a pair of speakers, preferably full range drivers to your amplifier.
  2. Connect the amplifier to a source(laptop or phone)
  3. Set source volume level to your average listening level
  4. Set amplifier gain to the level where you know beyond which it starts distorting/clipping
  5. Run Tone generator at Source and do a frequency sweep from 2000-3000Hz to below.
  6. Note the frequency at which the sound from speakers stops being audible
If the crossover is set up high, you will see audio levels start to dip around 300Hz. A good implementation would have them dip around 100Hz. If the full range signal is being driven through the 2 channels, you will continue to see output in lower frequency spectrum as well but that's highly unlikely in the case of your amplifier. In case of unavailability of full range drivers, use some computer 2.1 system's satellites. They ought to suffice.
 
TL;DR
  1. Yes
  2. You will but check what the High Pass filter is set at to see how much.
Detailed version
Do not look at power ratings of amplifiers and power supplies as is. Remember the demon lurking in the shadow called Loss, be it thermal or transmission. Using 90W power supply won't equate to 90W output(45W per channel) even if the amp is advertised as 50W per channel for L-R. Even though class D amps are efficient but they are not 100% efficient. That's Holy Grail of modern electronics. Moreover consider the fact that it's a Chinese implementation of class D design with aim of mass production so quality and efficiency are the last items on their agenda, if they are even there. 90W laptop power supply will suffice depending upon:
1. Efficiency/Sensitivity of speakers. If the speakers are efficient, even low amount of power would be enough to drive them to higher sound levels. Taking the worst case scenario with some hypothetical value range, let's say you get speakers with sensitivity of 86dB and the amp actually delivers 25Watts per channel, then you can theoretically drive them to almost 98-99dB level but this would also mean that distortion also increases(which it will in case of these cheap amps) Add to that the fact that audio level drop in an acoustically treated room is also almost 3dB per meter. So your listening position will also matter
2. As high pitched frequencies give the impression of loudness with less power input, these cheap amps focus more on Highs, to make them sound "loud and powerful". So the crossover is usually set high and the frequency roll off is also sloppy. To find out what's the scenario in the case of your amplifier, do the following:-
  1. Connect a pair of speakers, preferably full range drivers to your amplifier.
  2. Connect the amplifier to a source(laptop or phone)
  3. Set source volume level to your average listening level
  4. Set amplifier gain to the level where you know beyond which it starts distorting/clipping
  5. Run Tone generator at Source and do a frequency sweep from 2000-3000Hz to below.
  6. Note the frequency at which the sound from speakers stops being audible
If the crossover is set up high, you will see audio levels start to dip around 300Hz. A good implementation would have them dip around 100Hz. If the full range signal is being driven through the 2 channels, you will continue to see output in lower frequency spectrum as well but that's highly unlikely in the case of your amplifier. In case of unavailability of full range drivers, use some computer 2.1 system's satellites. They ought to suffice.
@Rud31 you have nailed it with core tech. I have ordered Wharfedale Diamond 220 bookshelf speakers (spec shown below) and trying to pair this Amp with it.

Wharfedale Diamond 220 speaker:
Sensitivity(2.83V @ 1m)86dB
Recommended amplifier25-100W

Peak SPL95dB
Nominal impedance8Ω Compatible
Minimum impedance4.1Ω
Frequency response (+/-3dB)56Hz - 20kHz
Bass extension (-6dB45Hz
Crossover frequency2.2kHz

Honestly speaking, being a noob in audio world I can understand first point, but not second. But one thing looks clear, this speaker does deserve better amp than this. I am right?
 
The second part involves some bit of Testing and Evaluation so do it when you get the time. These Wharfedales are good and with Sensitivity of 86dB, theoretically they'll easily reach their peak SPL of 95dB with just 8 Watts of power! Depending upon how much the amp colors (read distorts) signal, you should get a decent audio experience without pushing the amp too much. But this depends upon your listening position too. If you sit even 2 meters away from these speakers, then going by aforementioned formula, you'll need 32Watts per channel to get to 95dB and I'm pretty sure that trying to do that, the amp will probably distort the signal.
In short, if your listening position is going to be close by, you'll do fine. But if you plan to sit further back, you'll need a bigger, better amplifier.
 
Last edited:
While power delivery especially in terms of bass has been good with these Class D amplifiers, the Highs have a plastic feel to them, a kind of shrillness. It's not fatiguing but the soundstage is hollow. The Mids are almost absent. Maybe it's my drivers, maybe it's the amplifier but I know what I have to do, move on from Class D.

All I'll say is you are giving up too soon.
The only common factor I could see between the boards you used is the presence of tone controls - and you can't expect that to be good on ChiFi amps.
If possible, try get a basic board that's as close to the reference TI design as possible


@Rud31

Honestly speaking, being a noob in audio world I can understand first point, but not second. But one thing looks clear, this speaker does deserve better amp than this. I am right?
I have use class D Amps with $1200 , $700 and $250 speakers
I also use a $1500 AVR and a $1500 Integrated Amp for the same purpose.
I have rotated Amps and speakers in all sorts of permutations and combinations - All I can say is this:

$ to $ (or rupee to rupee), there is absolutely no comparison - These amp blow the more expensive counterparts straight outta the water.
Minor cons aside (the biggest of them being just a little bit of white noise if you hold your ear next to tweeter when idling) , these amps (if setup and used well) redefine cheap & cheerful
 
I haven't given up on them. Their time will come one day. Right now I just don't have time for getting into DIY.
The amp/chip made by TI isn't bad, it's the reference board design/layout "adaptation" using cheap components that ruins them. Class D is the future. Be it car audio where manufacturers like Rockford Fosgate, which is going for class D even for their multichannel amps, not just monoblocks or PA amplifier manufacturers such as Crown (XLS Drivecore series )and Behringer(iNuke series), everyone is getting on the bandwagon. Even "audiophile" companies like Bang and Olufsen and PS Audio(BHK series, Stellar series mono blocks) are using class D topology.
 
The Marantz PM7000N offers big, spacious and insightful sound, class-leading clarity and a solid streaming platform in a award winning package.
Back
Top