TPA3116D2 Amp

Shielded inductors as Rsavas pointed out should really make a big difference. Encasing the amp inside a metal shielding box that is grounded and separate from the FM board will help too. But simple use of class A/B amp, if that sounds good to you is an acceptable solution. The class D is more power efficient so for someone using solar power and batteries may be necessary.

You know what is noisy? Bluetooth transceiver units - they have residual buzz and noise that is a pain to get rid of.
I have not had much luck with switching devices. Besides the d amps, I have had a problem with a switching psu.
I converted a tabletop milling machine to cnc and powered the drives with an expensive Traco dc switching psu. The drives were picking up interference producing slow, rough and innacurate feeds. I tried to eliminate the problem with shielded cables, different earthing without success, I was tearing my hair out:mad:
I finally replaced the psu with a big old buzzing surplus linear supply and the the problem ended, my little machine just zipped around with no missed steps;)
 
Member
Joined 2008
Paid Member
Switching Amplifier (Class D) Basics | Audioholics

D amps are perhaps more problematic than most modders care to admit to, read passage on inductors. The writer is, I gather, well qualified.
I'm a music lover with neither the patience or experience to trouble shoot these inexpensive amp. The little chip amps perform flawlessly for me, but for the tweekers there is not much to work with ;-)

In all fairness, just because you have had issues with these amps doesn't mean the rest of us are somehow putting up with the same problems or tweaking to fix something that is flawed. Looking at the chip amp forum there sure seems to be a lot of modding and flawless is not a word I see thrown around there. The Fremen amp is what $1000 build based on how many years of tweaking the MyRef?

These amps don't work for you, fine. Enjoy the TDA and move on.
 
Last edited:
In all fairness, just because you have had issues with these amps doesn't mean the rest of us are somehow putting up with the same problems or tweaking to fix something that is flawed. Looking at the chip amp forum there sure seems to be a lot of modding and flawless is not a word I see thrown around there. The Fremen amp is what $1000 build based on how many years of tweaking the MyRef?

These amps don't work for you, fine. Enjoy the TDA and move on.
D-amps produce RF interference which is a well documented problem but never discussed on this forum. I have given up on tubes, vinyl and magnetic tape but I still find OTA FM current, even if I also listen to HD internet radio.
Efficiency is the only advantage I can see in using d-amps?
 
Guy's If you are having trouble with FM performance through the 3116, just listen to your station through your computer, problem fixed easy. I have not used an FM tuner in years
Fine option, if you want to listen to inferior music quality.
To me it is like comparing FM radio to a CD.
Comparing a well designed product against a cheap Asian version, is basically, you get what you pay for!! So don't complain when your cheap electronics, does not perform to your expectations.
Efficiency is the only advantage I can see in using d-amps?
Size/cost also comes to mind!!
I use a TPA3100D2 with $1000 speakers, Dynaudio Gemini's, it sounds fantastic to me. I do not need my 100W/ch Pioneer any more, unless I want to blast the tunes, so the neighbors can listen.
I design both Class "AB" and "D", they all have a place.
I bet you folks are all fine with the sound out of your new HDTV's, what do you think they use for audio amplifiers these days?
 
Fine option, if you want to listen to inferior music quality.
To me it is like comparing FM radio to a CD.
Comparing a well designed product against a cheap Asian version, is basically, you get what you pay for!! So don't complain when your cheap electronics, does not perform to your expectations.

Size/cost also comes to mind!!
I use a TPA3100D2 with $1000 speakers, Dynaudio Gemini's, it sounds fantastic to me. I do not need my 100W/ch Pioneer any more, unless I want to blast the tunes, so the neighbors can listen.
I design both Class "AB" and "D", they all have a place.
I bet you folks are all fine with the sound out of your new HDTV's, what do you think they use for audio amplifiers these days?
I have auditioned all the D and chip amps I have purchased on a pair of vintage Tannoy Yorks that I recently refused an offer of $7000!
All the T amps sound great but they do not do FM. Chip amps do, at least the ones I have purchased do.
As for price, these amps are inexpensive because the magic is in the chips and they are almost free.
I have a pair of chip amp powered active speakers connected to the audio output of my TV, it's probably the speakers and not the amps that make the difference.
Internet radio is fine if you only listen to the few that broadcast in hi res.
 
I have purchased on a pair of vintage Tannoy Yorks that I recently refused an offer of $7000!
And I thought that I spent enough $ on speakers. I find mfg speakers costs vs performance is thing of diminishing returns. To each ther eown it seems.
There is a big price difference comparing mfg speakers vs ones that you assembly from a kit, as I have done.
My first experience was some Avant 6A's that me and the bro bought at the same time we bought the Pioneer SX-950. I paid $739 for the Pioneer and bro spent $700 for the speakers. We were kids in high schoool, it was our summer savings. We blew those speakers up in no time. As I found out, they used cheap Philips drivers, ecaps and very small inductors in the x-overs. So the quest to upgrade went on for a few years after warranty ran out. So I ended up putting in a new 15" woofer, a Morel MDM55 for a mid and MDT29 for the tweeter and a high quality Philips x-over, using large air core wound inductors and 200V film caps. I have never had to do anymore. I said I'll blow my friends ears before I blow up anymore drivers. I saw today Madisound is having a sale on those same Morel drivers, which I installed years ago, so what does that say?
All the T amps sound great but they do not do FM.
Pls explain, as I am confused by this statement.

I do get a chuckle about talking about a cheap Class "D" amp and $1000-7000 speakers. Some how it just does not seem to be a fair comparison for the class "D" amp. But in the end it sound just fine to me.

Cheers
Rick
 
High efficiency speaker systems (Fostex in my case) are a wonderful match with the cheapest D class amp I could find, specifically the PE TPA3110. And this is for my HDTV video plus fm nearfield system. The midrange trumped my Trends 10.1/ Welborne ps setup. The T amp has a somewhat recessed midrange quality that gets me a little lost with the vocals because I listen at probably lower volumes than most
D class amps just seem much more alive.
Tubes make up my main and headphone systems. A 500+ lp vinyl collection keeps me from converting whole hog to digital, however I thoroughly enjoy my Sony PS 1, (output modified), and small cd collection.
I was about to assemble a Jerry's Elec. TPA 3122D2 with Bourns inductors and Panasonic caps but am holding off for now. Small form factor kit bashes are a great thing for me.
Through you folks posting articles about my specific problem of interference I have narrowed it down to the speaker cables, which though under 4 ft. are not adequate in containing the rfi. The cables by necessity are draped near my beloved Rotel RT950BX tuner (again modded quite heavily). So my quest begins to find a solution to the speaker cable dilemna. Maybe woven copper sheathing with a ground scheme.
 
Last edited:
Founder of XSA-Labs
Joined 2012
Paid Member
Through you folks posting articles about my specific problem of interference I have narrowed it down to the speaker cables, which though under 4 ft. are not adequate in containing the rfi. The cables by necessity are draped near my beloved Rotel RT950BX tuner (again modded quite heavily).

Try looping your speaker cables a couple of times through a ferrite donut EMI suppressor as close to the amp output as possible. Also, does your class D amp employ the EMI snubber circuit as recommended by TI?

Shielding a cable is a very tricky thing to do properly - if you have a small gap or leak (as you will after the cables between the speaker banana posts and leading to the drivers themselves inside the cabinet), it will do no good.
 
Member
Joined 2008
Paid Member


I do get a chuckle about talking about a cheap Class "D" amp and $1000-7000 speakers. Some how it just does not seem to be a fair comparison for the class "D" amp. But in the end it sound just fine to me.


My speakers are diy but would certainly be worth ~$2000 retail and these TPAs are as good as any amp I've tried with them including SimpleSE and Onix SP3 tube amps, IR2092, Tripaths, Pioneer chip amp receiver (Geddes), and various solid states.

I've mentioned the $10 PE TPA3110 at other audiophile forums but can't get anyone to try them. These chips are so cheap that I think people just refuse to take them seriously or just think they are going to be like Tripaths. It's only by random luck that the TBI Millenia amp (TPA3100) has become an audiophile darling - it was available for 4 years before anyone noticed. Now people are comparing it to four and five figure amps, including Ncore. The slick enclosure goes a long way I imagine.

Does anyone think the Yuan Jin boards can be improved upon (aside from component modding)? These amps are so good I'd like a bona fide high quality board, provided there's room for improvement.
 
Last edited:
Try looping your speaker cables a couple of times through a ferrite donut EMI suppressor as close to the amp output as possible. Also, does your class D amp employ the EMI snubber circuit as recommended by TI?

Shielding a cable is a very tricky thing to do properly - if you have a small gap or leak (as you will after the cables between the speaker banana posts and leading to the drivers themselves inside the cabinet), it will do no good.


Would the snubber circuit be the four smd caps directly adjacent to the output terminals and I imagine connected with them?


Otherwise I'm not too up on the schematic of this Sure Electroncs 2X8 watt TPA3110 board sold by Parts Express.


Thanks for the suggestion of looping through ferrite donuts.
 
Pls explain, as I am confused by this statement.

I do get a chuckle about talking about a cheap Class "D" amp and $1000-7000 speakers. Some how it just does not seem to be a fair comparison for the class "D" amp. But in the end it sound just fine to me.


I have purchased a number of D amps over the last year or so, 6 I believe, and they all radiate RF interference to a greater or lesser degree. I have tried ferrite beads on the speaker wire, replaced the switching psu with a linear, even tried a mains isolating transformer.
My speakers are 3 sided, designed to fit in the room corners, so I cannot shorten the wires.
My best sonic results, to my ears anyway, is a pisswilly little TDA7297 costing $6 including postage from China, unbelievable. I purchased it originally to put inside one of a pair of RS Minimus 7 speakers and just for giggles I connected it to my Tannoys. I love, I get no rf interference and it beats the pants off a McIntosh SS intergrated amp it replaced.
I did mention this on the Tannoy Forum but it was not discussed. I guess if you have invested a wad on tube gear.........
 
Founder of XSA-Labs
Joined 2012
Paid Member
I've mentioned the $10 PE TPA3110 at other audiophile forums but can't get anyone to try them. These chips are so cheap that I think people just refuse to take them seriously or just think they are going to be like Tripaths.

I recently tried the $10 Sure TPA3110D2 and I agree it sounds unbelievable - for the size and the price. I like the screw terminals and option for hard soldering line inputs and speaker outputs. I think it sounds almost as good as the 3116 just lower power. If you drive it with 19v and use 4 ohm speakers the distortion is even lower and power higher. I bought some Panasonic 470uF power caps to upgrade but haven't had time yet as they sound so good stock.

Everyone should have at least a couple of these on hand for DIY projects. Pair this with a 99dB pro audio driver and it is unbelievably loud and clean.
 
I recently tried the $10 Sure TPA3110D2 and I agree it sounds unbelievable - for the size and the price. I like the screw terminals and option for hard soldering line inputs and speaker outputs. I think it sounds almost as good as the 3116 just lower power. If you drive it with 19v and use 4 ohm speakers the distortion is even lower and power higher. I bought some Panasonic 470uF power caps to upgrade but haven't had time yet as they sound so good stock.

Everyone should have at least a couple of these on hand for DIY projects. Pair this with a 99dB pro audio driver and it is unbelievably loud and clean.

I have one of them now and I hope to give it a good listen in the next few days.
 
these TPAs are as good as any amp I've tried with them including SimpleSE

This reminds me of a currently running Toyota commercial on American TV where other cars are compared to the Camry...."it's as good as a Camry"

Obviously I have a bias to my old friend the SSE which has been connected into my studio/lab system for almost 8 years. The system has evolved over time, and dozens of amps have come and gone, but SSE number 1 (the first one ever made) has remained. I have a Technics turntable that I bought new in 1978, a Toshiba DVD player that also plays CD's, and a Haswell based PC that is a media center, recording studio, and DAW (Digital Audio Workstation) and a small Samson mixer to tie it all together and provide inputs for guitars and mics.

The Yamaha studio monitors have been here for 10+ years. There is a cheap Polk powered subwoofer (10 inch 50 watts) behind the computer monitor on the left pointed backwards at the corner of the room. This is done primarily to avoid rattling the monitor to death and the cheap sub seems to sound better this way. Loud EDM will cause the sub to distort before the SSE. The Yamahas die at about 80 Hz, so the sub is set to go from 100 Hz down to "32 Hz". I am in the process of rearranging the studio/lab so the FH3 horn speakers are not present here yet, nor are the DIY "loudboxes" that I use for playing guitar.

I use a mediocre laptop for a mobile DAW, along with a Lepai 2020 T amp and the pair of loudboxes all run from a boat battery. Two weeks ago I had this system set up at an outdoor ham radio show connected to a Roland keyboard. I have plans to build a portable DJ/DAW "boombox" that runs on battery power for these situations. A friend from work came by and told me that he had several TI evaluation boards that I could play with so I went to see him. I got to test a TI 3116 board, and saw it make big power in the lab at work, but didn't hear it. I had to give that one back, but I did get a 3100 board to take home.

It is exactly the same as the board shown in post #1 of this thread where it is also compared to the SSE:

http://www.diyaudio.com/forums/class-d/217903-ti-tpa3100d2-amazing-class-d-amp.html

So what should I do????? Well I think I should compare it to the SSE!

I disconnected the SSE and wired the TI 3100 board in its place leaving everything else the same. I had to flip the "phase" switch on the sub to make it work. I tried powering the board from a Sorensen bench power supply that can be adjusted from 0 to 25 volts and will put out up to 5 amps. I also tried the 19 volt laptop brick from the mediocre HP laptop rated for 3.5 amps.

The SSE runs triode wired KT88's for about 8 WPC. It weighs about 20 pounds and draws about 120 watts to make 16 watts of audio. It burns those 120 watts all the time, even when no music is playing. The TI board makes more power if you turn up the B+ voltage. I liked it best with the power supply maxed at 25 volts. It idles at less than a watt of power consumption, and consumes about 10% more power than it puts out. The SSE measures at about 1% distortion at 5 watts. The TI board measures 0.11%.

Listening tests at low volume levels didn't reveal a clear winner. The amps sounded quite similar, with an edge to the 3100 in the dynamics, and an edge to the SSE in detail and realism. When the volume was cranked the 3100 was quite obvious when it clipped. The SSE is subtle. You can see the clipping on the scope before you can hear it. Even though the 3100 has more than twice the power of the SSE, they both become too distorted to listen to at about the same acoustical level. For most music, I would chose the SSE, for EDM, techno, and some loud rock, the chip amp might be a batter choice.

For now, I am going to leave the 3100 in place, because I have two different YuanJing 3116 boards waiting for a suitable power supply. They will get tested next.

There is no contest for a battery powered portable "BoomStation." Some sort of chip amp will get used. So far I like what I hear with these little chips.
 

Attachments

  • Nexta.jpg
    Nexta.jpg
    235.8 KB · Views: 1,118
  • Setup_2a.jpg
    Setup_2a.jpg
    197.5 KB · Views: 1,080
  • Setup_1a.jpg
    Setup_1a.jpg
    195.6 KB · Views: 1,064
  • SSEa.jpg
    SSEa.jpg
    220.1 KB · Views: 1,069
I ran out of time last night, but I was going to say that the 3100 board sounds much better, and distorts less when fed by the linear power supply, even when it is set to the 19.2 volts that the laptop brick puts out.

The laptop brick has hard current limiting, and possibly foldback, so that when you ask it for over 3.2 amps even for a millisecond, its voltage output drops. The recovery time once the overload is removed can be 10's of milliseconds. Most SMPS's do this by design to protect themselves. The big old linear supply has a fat capacitor across it's output, so for short intervals it can put out even more than it's 5 amp rating.

I have a MeanWell 156 watt unit on order for the 3116 boards. The 3116 board can put out 200 watts, so it will need at least 225 watts to operate at full power. This can be acomplished by putting a large cap across it's output, since you don't need full power continuously. Some SMPS's will refuse to start up into a large cap. Some experimentation will be needed.

The "BoomStation" that I want to build will never need this much power when used on battery power, or with its internal speakers, but it will be capable of playing through external speakers on AC power, hence the need for small efficient audio amps.

George and thanks for being objective even in subjective tests.

I always tell it like I see it. I will always have my SSE in the system, but I can switch in other amps rather easilly, and its always good to have another weapon in your arsenal. I made several chip amps in the late 90's since the National Semiconductor sales rep was always giving me free chips. I wasn't too keen on these since I could blow them up by playing my guitar through them. I fixed several Peavey amps with blown chips too. It's too soon to tell how the TI chips can eat overdriven guitar, but I did get some TAS5630 chips for a BIG amp.
 
Pls explain, as I am confused by this statement.

I do get a chuckle about talking about a cheap Class "D" amp and $1000-7000 speakers. Some how it just does not seem to be a fair comparison for the class "D" amp. But in the end it sound just fine to me.

radiosmuck
What I mean is some/most? people who own $1000-7000 speakers do not consider a class "D amp worthy of driving their expensive speakers.
As a case in point, as wushuliu has stated
I've mentioned the $10 PE TPA3110 at other audiophile forums but can't get anyone to try them.
I call them audiophile snob's. Like a kid that won't eat something and has never even tasted it.
I do like the comparison of a class "A" tube amp with a class "D" amp. I wonder what Nelson Pass has to say about Class "D" amps? People have to learn to put their biases aside and look at things objectively. Akin to thinking with your head and not your heart.
I hope I have clarified my point.

Tubelab.com, your feedback is great information to read about testing of these TI parts. I have not done the extensive testing that you are doing so it valuable info for me.
I am skeptical of powering my TPA3100D2 design using a switch-mode supply, but since the premise of using class "D" is for efficiency gains, it does make sense to use one. I will try using my HP laptop brick and see how it performs with the amp itself and the tuner etc to see if it makes any differences that I can determine.

Cheers
Rick
 
Last edited: