TPA3116D2 Amp

We used a 27V Meanwell unit - speakers were a pair of Cabasse Galion from the early '70s - they are 8 ohm speakers and their sensitivity should be around 90-92db.

Cheers

Those speakers should be fine. Something's not right. Using a 24V supply, I did a power test on this board driving a square wave into an 8 Ohm dummy load to the onset of clipping and the the chip and inductors were not getting too hot (despite my load resistors approaching melt-down).
 
Can someone give me center to center measurements for the mounting holes on YJ blue board?

Front two holes to single rear hole: 71mm (middle of board in rear)
Left front hole to right front hole: 25mm

Following KJA 2013's lead, I decided to use Neutrik 4-pin XLR connectors for the power supply along with a DC power cable made of a twisted pair of 16 AWG Teflon-insulated stranded Copper wire. I also installed my YJ "blue" amp in a Modushop Galaxy 237 chassis I bought a few years ago when they had a sale. The chassis is large for the small Class D amp, but I'm considering input transformers for this amp in the near future, so the space will not go to waste of I go down that route.

Besides, the amp now nicely matches my Audio Research phono stage and line stage preamps.

A7-2.jpg A7-5.jpg A7-6.jpg A7-8.jpg

Maybe I'll install a muted green LED in the faceplate for a more finished look.
 
Last edited:
Good deal! :)
I think they are 26dB gain by default. The 24v laptop supplies with that kind of amps is rare - must be a big workstation or gaming laptop? I have heard ideal voltage is 21 v for best low distortion performance but 24 v should work well still.

Yes in the datasheet it recommends 21V for the 3116 IC and 24V for 3118, about the gain, any of those who has it check the colours of the resistors at the gain/slave pin... There should be 20k and 100k for 26db. Remember what saturnus said, 26db it would be a normal use gain, If using a portable source should be 32db or even 36db for iPod or iPhone use.
 
Founder of XSA-Labs
Joined 2012
Paid Member
Those speakers should be fine. Something's not right. Using a 24V supply, I did a power test on this board driving a square wave into an 8 Ohm dummy load to the onset of clipping and the the chip and inductors were not getting too hot (despite my load resistors approaching melt-down).

What RMS power were you delivering to the dummy loads? When I ran a 120 Hz sine wave test at -8dBV (from test CD out of CD player) I measured 75 watts rms into 4 ohm loads and the heatsink got hot to the touch. I did not notice the inductors getting abnormally warm. Unless you have heatsinked 100 watt resistors for dummy loads, they will get pretty hot. Think about how hot a soldering iron gets at 40 watts. 100 watt light bulbs might work if you get enough if them for 4 ohms.
 
What RMS power were you delivering to the dummy loads? When I ran a 120 Hz sine wave test at -8dBV (from test CD out of CD player) I measured 75 watts rms into 4 ohm loads and the heatsink got hot to the touch. I did not notice the inductors getting abnormally warm. Unless you have heatsinked 100 watt resistors for dummy loads, they will get pretty hot. Think about how hot a soldering iron gets at 40 watts. 100 watt light bulbs might work if you get enough if them for 4 ohms.

It was around 33 watts with a 1K sq. wave. The dummies that got pretty hot were the little aluminum finned wirewound type rated at 50W.

When I used the 100W non-inductive load resistors from Parts Express (019-020), they barely got warm.
 
this week my new red yj board arrived.
i recognize a few differences compared to the previously purchased boards:

- 680nF caps instead of 560 nF (AND blue instead of yellow!!!)
- heatsink with sharper edges
- heatsink paste leaking out under the heatsink onto the legs of the chip
- the prints on the board for R-out and L-out are permuted
- power connector sitting a bit 'sidewards' on the board
- solder joints not as good as on the previous boards (but still ok)

even if i connect a signal only to the R-channel, i still get a bit of audible noise out of the L-Output.
i cannot compare that to the older boards, because i have modded them to pbtl.

all in all the amp still sounds great for that price (10,84€ shipped to germany).
 
this week my new red yj board arrived.
i recognize a few differences compared to the previously purchased boards:

- 680nF caps instead of 560 nF (AND blue instead of yellow!!!)
- heatsink with sharper edges
- heatsink paste leaking out under the heatsink onto the legs of the chip
- the prints on the board for R-out and L-out are permuted
- power connector sitting a bit 'sidewards' on the board
- solder joints not as good as on the previous boards (but still ok)

even if i connect a signal only to the R-channel, i still get a bit of audible noise out of the L-Output.
i cannot compare that to the older boards, because i have modded them to pbtl.

all in all the amp still sounds great for that price (10,84€ shipped to germany).

Boy, looked like the price came down. 10,84€ is < 15 US$. I paid 17 US$ a few months ago. I guess the emergence of the new version (blue boards) pushed down the price of the old one. As you said, it is a heck of a deal!!

Regards,
 
I am the one with the "hot rod" blue board Wolfram mentioned above. First of all I have to say, that I was not prepared to get this kind of sound from an amp for 20€ shipped. Very, very good already from the start, improving the first hours and then more the last days.
Call me impressed. Up to this one every Class D amp or the like had very good bass, sometimes good midrange, almost never good highs and sounded clearly less "natural" than my tube amps.
This one is the first doing it all and sounding very natural, dynamic and balanced even bone stock. Total steal in my book, and I don't even have the itch to swap parts, as I normally have very soon after the first tunes are played.
Unfortunately it is getting really hot with more than 20 Volt, I cannot touch the inductors at all without burning my finger, so 15 is highest. But it sounds better with some more volts. And when I leave the inputs unconnected, there is a loud noise from the speakers, the same appears when the preamp is connected, but switched off. All this doesn't seem to be normal, so I wrote to the vendor and asked what to do now. Will report back when something happens....

cheers,
Juergen
 
Founder of XSA-Labs
Joined 2012
Paid Member
I am the one with the "hot rod" blue board Wolfram mentioned above. First of all I have to say, that I was not prepared to get this kind of sound from an amp for 20€ shipped. Very, very good already from the start, improving the first hours and then more the last days.
Call me impressed. Up to this one every Class D amp or the like had very good bass, sometimes good midrange, almost never good highs and sounded clearly less "natural" than my tube amps.
This one is the first doing it all and sounding very natural, dynamic and balanced even bone stock. Total steal in my book, and I don't even have the itch to swap parts, as I normally have very soon after the first tunes are played.
Unfortunately it is getting really hot with more than 20 Volt, I cannot touch the inductors at all without burning my finger, so 15 is highest. But it sounds better with some more volts. And when I leave the inputs unconnected, there is a loud noise from the speakers, the same appears when the preamp is connected, but switched off. All this doesn't seem to be normal, so I wrote to the vendor and asked what to do now. Will report back when something happens....

cheers,
Juergen

I am glad they sound good I may have to buy these myself. My order of the two green Ybdz boards still have not arrived - almost a month now crazy slow service. How loud are you running them? It is unusual that the inductors are getting that hot.
 
The heat is there in any case, also without signal. And yes, it sounds very good indeed.
BTW I am not convinced that SMT designs sound better. Maybe they have advantages in theory, but through hole parts, for example caps, sound much better than SMT parts in my experience. So I always look for the through hole boards, exept high freq switching stuff in MHZ regions.
One month is too long, if you ask me. I had my blue board in 6 days from HK to Germany.

Ah, and thanks to Danzz for the nice design:cheers:...
 
I am the one with the "hot rod" blue board Wolfram mentioned above. First of all I have to say, that I was not prepared to get this kind of sound from an amp for 20€ shipped. Very, very good already from the start, improving the first hours and then more the last days.
Call me impressed. Up to this one every Class D amp or the like had very good bass, sometimes good midrange, almost never good highs and sounded clearly less "natural" than my tube amps.
This one is the first doing it all and sounding very natural, dynamic and balanced even bone stock. Total steal in my book, and I don't even have the itch to swap parts, as I normally have very soon after the first tunes are played.
Unfortunately it is getting really hot with more than 20 Volt, I cannot touch the inductors at all without burning my finger, so 15 is highest. But it sounds better with some more volts. And when I leave the inputs unconnected, there is a loud noise from the speakers, the same appears when the preamp is connected, but switched off. All this doesn't seem to be normal, so I wrote to the vendor and asked what to do now. Will report back when something happens....

cheers,
Juergen

Like you, I am very impressed with the YJ blue amp too. This is one of the best-sounding Class D amps I've heard paired with my tube preamps--maybe the best to date. I've only replaced the stock Nichicon HD caps with Panasonic FM caps and I rolled my own shielded input cables to replace the cable that came with the amp.

A7-9.jpg

I reported that there was a cold solder joint with one of the input caps on my YJ blue amp. It's possible that there might be something amiss with the worksmanship on your amp, and that might be causing the temperature to run away at higher Vcc. I touched up all the thru-hole solder joints on my amp using the lead-containing Cardas solder I use for all my projects. It might not be the solution, but it would be too much to do.
 
Nice to read all these good reports on YJ blue board. But unfortunately Danzz didn't care to insert any sort of bump/pop protection :rolleyes:, rather than the blue one from Chengzhi has. Is it an issue?

I've reported before that my Danzz-designed YJ blue amp doesn't have a turn-on thump. On turn-off, I hear a very slight thump, but I see absolutely no driver excursion on my speakers, which have 98dB sensitivity. I experienced the same with Sure Electronics TPA3110D2 amp. Danzz followed the Texas Instruments recommended schematic. There's absolutely nothing wrong with that. Thanks again to Danzz for creating a great layout.
 
Founder of XSA-Labs
Joined 2012
Paid Member
Could the Danzz YJ 2.0 Blue Board be the best sounding amplifier value EVER? We have several people report that the 3116 amp sounds very good and even gives their tube amps the run for the money. Can someone who has a nice class A amp like the F5 or similar comment on the comparison of sound quality between say the hot rodded Danzz 2.0 board and their class A amp? Something in the 25 watt to 75 watt/ch range for comparison and the type of speaker used for the comparo should be mentioned. If someone even has a nice stereo digital recorder like a Zoom mic can make a sound clip recording to post that would even be better. I think the problem may be that people who build home-heating class A amps may never give this amp a try... I know some folks who have nice class A amps, I may have to just give them one to try. :) Might ruin it for them and they won't ever look at their class A the same again? But then again, I may not know what I don't know... which is why I am asking for listening impressions and comparisons.
Thanks,
X
 
Last edited:
what do you guys think of putting this little boy into such a cheap enclosure?

I found this on ebay at 12,99 $ with free shipping from china and has even the RCA inputs that can be connected, has the power input, speaker terminals, etc

it would need a quick and dirty job to wipe it's guts out though :)


An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.
 
Last edited: