TPA3116D2 Amp

Hopefully this post is not off-topic. I can't find a better place to put it, but just let me know if it needs to be moved. I'm just wondering, for those that have heard any of the First Watt or ncore DIY amps, do you think that any of the 3116 boards have the potential to sound as good with the right components, and if so, how much would you expect to need to spend, not including a case or power supply? I am considering building one in a few months, but this will be my first electronics project in 30+ years, so the Nelson Pass/First Watt F6, for instance, is most likely beyond my capabilities, and possibly my budget, as well.
 
Looks like you have absolute phase wrong here, see positive outputs from bridge rectifiers. But there also seems to be a via next to positive output bridge rectifier OUT2, does bottom side show anything? Or from outputpins from these bridge rectifiers signal goes straight to output terminals?

How about that new 3110, +/- output from rectifiers differs from standard tpa3110 +--+ too, so could be wrong after all? I believe I read new 3110 had grounded negatives?
432162d1407551825-tpa3116d2-amp-sure-tpa3116-output-polarity-correction.jpg

My new 3110 does not have grounded negatives, and it's labeled the same as the 3116 +--+
 
Output from TI TPA3110= L+L-R-R+. So if the MB6S bridge rectifiers are not in signalpath or signal crosses after the MB6S befor reaching 1 of the speakeroutput then speakeroutput on SureTPA3110 is correct. Looks as if they are in output signalpath, but matte black soldermask pcb doesn't make photo's very clear, maybe they are not :) No idea what their function on these pcb's would be, maybe to limit reverse current??? In powersupplies they also limit ripple and signal sinewave gets altered I think I read :) Not something desirable in speakeroutput poweramplifier???
 
Founder of XSA-Labs
Joined 2012
Paid Member
Hopefully this post is not off-topic. I can't find a better place to put it, but just let me know if it needs to be moved. I'm just wondering, for those that have heard any of the First Watt or ncore DIY amps, do you think that any of the 3116 boards have the potential to sound as good with the right components, and if so, how much would you expect to need to spend, not including a case or power supply? I am considering building one in a few months, but this will be my first electronics project in 30+ years, so the Nelson Pass/First Watt F6, for instance, is most likely beyond my capabilities, and possibly my budget, as well.

Good question. I have wondered myself and have asked this too. I think there was a member who said that he has one of the First Watt amps (can't remember which by I think an earlier one) and he said the 3116 sounds remarkably a lot like a good class A amp. The parts for and F5 (with case) will run you almost $700 and you have only 25 watts. I don't know what the F6 does but that is probably more. The difference in heat output beside the cost and weight are probably the biggest things. You could heat a room in winter with an F5.

Regarding mods on the 3116, for about $20 you can probably get all of the major mods that will make it sound excellent. There is also the input transformer that may take it to the next level and that can be expensive. A $31 option has been spec'd but not tested yet. Much more expensive $100 to $200 transformers have been used with resounding success.

I second your question and ask anyone who has an F5 etc to do an A/B test with a nicely modded 3116 and let us know.
 
Founder of XSA-Labs
Joined 2012
Paid Member
MB6S is part used, but maybe they are not in output? That could be...

As I don't have a Sure 3116 or 3110 in front of me we will have to await KJA2013 to tell us how the full wave bridge MB6S rectifier is hooked up. As I said, a full wave bridge has the neat trick of frequency doubling an AC input. Maybe it is used for the power bus to prevent reverse polarity? Odd location for that though.
 
Founder of XSA-Labs
Joined 2012
Paid Member
A bridge could be used for diode clamping of the outputs...convenient single package for four diodes.

Placing one part vs four parts is a cost savings for auto assembly...even if the part is more expensive.

That makes perfect sense. So the AC part would be connected to ground and the positive to ch1 +out and negative to ch1 -out?
 
Hopefully this post is not off-topic. I can't find a better place to put it, but just let me know if it needs to be moved. I'm just wondering, for those that have heard any of the First Watt or ncore DIY amps, do you think that any of the 3116 boards have the potential to sound as good with the right components, and if so, how much would you expect to need to spend, not including a case or power supply? I am considering building one in a few months, but this will be my first electronics project in 30+ years, so the Nelson Pass/First Watt F6, for instance, is most likely beyond my capabilities, and possibly my budget, as well.

Stereo DIY Ncore modules with shared DIY powersupply = $1350
They better be better :D And I am sure they are. Some early in this tread compared with 180 module and favoured the 3116, so at least 1 person at a certain time thought he liked 3116 better then his Hypex, but Ncore is a level higher I would assume. The Hypex's have fully load independent frequency respons and make beautifull SMPS powersupplies. Without a powersupply close to those, a good linear will be worse then Hypex SMPS's, TPA3116 isn't likely to come close I would think, can't remember PSU information in post that compared the 180 Hypex with probably redboard YJ. Caution needs to be given that YJ ampboards in early photo's show parts you will not get buying one now, all changed parts are worse...China...
 
Hopefully this post is not off-topic. I can't find a better place to put it, but just let me know if it needs to be moved. I'm just wondering, for those that have heard any of the First Watt or ncore DIY amps, do you think that any of the 3116 boards have the potential to sound as good with the right components, and if so, how much would you expect to need to spend, not including a case or power supply? I am considering building one in a few months, but this will be my first electronics project in 30+ years, so the Nelson Pass/First Watt F6, for instance, is most likely beyond my capabilities, and possibly my budget, as well.

I am attending the California Audio Show here in the San Francisco Bay Area next Friday. I plan to listen to some First Watt and Pass Labs gear. Unfortunately, the dealers and manufacturers who sponsor the show would frown on me bringing a DIY TPA3116 amp to compare to their very expensive amplifiers. Still, I hope to glean enough from listening sessions to formulate my own opinion on whether the more expensive Class A amps are significantly better than the TI Class D amps.

Unfortunately, there will not be a Burning Amp Festival in San Francisco this Fall where comparing some PassDIY (i.e., First Watt derivatives) and other DIY amps like mine would be possible. Maybe I'll just have to build my own F3 or Aleph J to make that comparison. I have the deepest respect for Nelson Pass who has contributed so much to the DIY community. He told me about Elna Silmic II caps a few years ago at a Burning Amp Fest, and in certain applications, these caps are special.

At last year's BAF, I had a chance to listen extensively to the Amp Camp Amps, and they were very nice indeed. The PassDIY amps didn't do anything for me, but I chalk that up to the choice of speakers, which did not match up well with the room acoustics at the Fort Mason Center. My opinion is that a well implemented TPA31xx amp with the right speakers can compete with more expensive amps. I've been interested in Class D amps since the Sonic Impact T-amp caused a big stir, and I've had a chance to work with the really expensive and excellent-sounding Lab Gruppen Class D pro audio amps. To my ears, these TPA31xx amps are great, especially when paired with tube preamps, for modestly-powered systems or systems with very efficient speakers. Can't wait to get my CineMag transformers and run my YJ TPA3116 on my friend's Altec Lansing Model 19's.

Here's a photo of my last gathering where my modified YJ TPA3116 amp gave my modified Dynakit Stereo 35 amp (a legend in its own right). We only spun vinyl that day using my ARC PH5 phono stage and ARC LS7 line stage tube preamps. My friends couldn't believe how well the Class D amp was producing a rich sound with even a some bloom (sorry, but tube amps rule in this area).

image.jpg
 
Here is my NEW 4 Channels Remote Volume Controller and hooking up with YJ Blue board.

I use the Voltage converter from Single 24V to +/- 12V for PGA power supply. Now sharing the one SMPS(24V) for TPA3116 and PGA2311.

The Codes supporting the Apple remote and TPA3116 mute, S/D controls.

TPA3116_PGA2311_A1.jpg


TPA3116_PGA2311_A2.jpg


TPA3116_PGA2311_A3.jpg

Uses all Nichicon FG caps

How do you get the apple remote to work with this setup?
 
Dear TPA3116D2 experts,

From the TPA3116D2 Wiki, on output filter capacitors I am finding reference to Wushuliu's post #658 on the following link:

TPA3110 and TPS3116 amps

He mentions: "Wima FKP, MKP, and MKS for .68uf,10nf, and 1nf output filter caps"

Ok to go with only one type of cap (e.g. MKP) or I need 3 types for each capacitance?

Also any guidance on the minimum voltage?
 
Last edited:
I have the deepest respect for Nelson Pass
As do I. If I could afford a J2, or even an M2, then I would probably not be posting on this topic.

To my ears, these TPA31xx amps are great, especially when paired with tube preamps, for modestly-powered systems or systems with very efficient speakers.
That is what I am planning to do if I build a TPA-based amp.

Here's a photo of my last gathering where my modified YJ TPA3116 amp gave my modified Dynakit Stereo 35 amp (a legend in its own right).
Looks like you forgot to finish your sentence. I have considered refurbishing a Dynakit ST 70, or buying one of the kits based on the design.

I look forward to reading your thoughts as you make further mods. I'm hoping that the 3116 can be made into a real giant killer for not more than a few hundred dollars. That would be great news.
 
That's the problem with a limit on editing posts. What I was trying to say was that the YJ amp gave my Dynaco a good run. My friends who all have very nice systems were very impressed with the YJ amp. As I mentioned before, one of my friends ended up buying the unit from me, so now I'm modding a second unit.
 
As I don't have a Sure 3116 or 3110 in front of me we will have to await KJA2013 to tell us how the full wave bridge MB6S rectifier is hooked up.

These boards are murder to see the signal paths on.

The pair of inductors for each channel are connected to the AC side of one of the two MB6S Bridge Rectifiers

The DC side goes to the bank of decoupling capacitors.


Here's a simple diagram showing it.
 

Attachments

  • Implementation of MB6S in Sure TPA3116.jpg
    Implementation of MB6S in Sure TPA3116.jpg
    126.7 KB · Views: 653
These boards are murder to see the signal paths on.

The pair of inductors for each channel are connected to the AC side of one of the two MB6S Bridge Rectifiers

The DC side goes to the bank of decoupling capacitors.


Here's a simple diagram showing it.

??? so your 12A DC supply (mine) is connected to output to your speakers? separated by a device that has 0.5A peak ability? maybe I read datasheet wrong?
 
Founder of XSA-Labs
Joined 2012
Paid Member
These boards are murder to see the signal paths on.

The pair of inductors for each channel are connected to the AC side of one of the two MB6S Bridge Rectifiers

The DC side goes to the bank of decoupling capacitors.


Here's a simple diagram showing it.

Yes, but where does the +/- of the rectifier bridge go?
 
Founder of XSA-Labs
Joined 2012
Paid Member
It's stated on the diagram.

Pin 1 is +
Pin 2 is -

There's probably much more to it, but this much is what I measured.

Sorry, did not see the little 1 and 2 from my phone's screen. As they go to the power rail and ground, they are clamp diodes just like Dug suggested. But why is this needed or helpful? Seems like it may add some sort of diode switching noise to the system at the crossing level of 0.6VDC?