TPA3116D2 Amp - diyAudio
Go Back   Home > Forums > Amplifiers > Class D

Class D Switching Power Amplifiers and Power D/A conversion

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 4th June 2013, 01:08 PM   #1
diyAudio Member
 
Join Date: Jun 2011
Default TPA3116D2 Amp

Well, I somehow managed to build my first solid state amp, based on the TPA3116D2 chip. I used the schematic from the data sheet and it works and sounds great. The power supply is a 130 watt Dell laptop power brick putting out 19.5V. The chip barely gets hot driving 8 ohm speakers. Thanks to everyone who posts here for their ideas and inspiration, especially xrk971 and his great amp, that thread is what pushed me from building a chip amp to a cheaper, high efficiency class D. I scoped the output and the wave is smooth, into a resistive load you can barely see the switching on top of the sine wave. The purpose of the amp is to have something portable to take with me to the garage, basement, etc and it fits the bill quite nicely.

The only problem I ran into was with the output capacitors, I originally had cheap 50V caps in and they quickly smoked without a speaker load. I replaced them with the square 100V caps and all is well. I don't plan on running it without a load but it has to survive at least for a little while if a speaker wire pulls out or isn't connected.
Attached Images
File Type: jpg IPhone Pictures 147.jpg (923.0 KB, 8539 views)
File Type: jpg IPhone Pictures 149.jpg (751.3 KB, 7433 views)
File Type: jpg IPhone Pictures 151.jpg (586.3 KB, 6717 views)
  Reply With Quote
Old 16th June 2013, 01:59 PM   #2
ddapkus is offline ddapkus  United States
diyAudio Member
 
Join Date: Jun 2013
Hi,

I am the Audio Apps Manager at TI, and I'd like to congratulate you on a nice looking amplifier!

I suspect the reason that your output caps blew up without a load is due to a high voltage oscillation that will occur with an LC filter with no load. The load is part of the filter response, so when you remove it, you can get huge oscillations. See this app note for more info on LC filters: http://www.ti.com/lit/an/sloa119a/sloa119a.pdf

I noticed that you didn't use a heat sink... Under heavy load, this will result in thermal shutdown. You can easily add some metal to heat sink it, or you can switch to the TPA3118D2 which has a PowerPAD(tm) which allows you to solder the thermal pad on the bottom to your PCB so the PCB acts as a heat sink. But, with your design, the PCB is pretty small, so I'd probably just add an external heat sink.

Another resource you can tap into is our engineer2engineer forum at TI. You can find it here: TI E2E Community

-d2
  Reply With Quote
Old 16th June 2013, 02:02 PM   #3
ddapkus is offline ddapkus  United States
diyAudio Member
 
Join Date: Jun 2013
Oh, sorry, I forgot to add, if you want to operate with open load, you can simply use some small signal diodes to shunt the oscillations to the power and ground. Something like this:


Click the image to open in full size.

-d2
  Reply With Quote
Old 16th June 2013, 02:45 PM   #4
xrk971 is offline xrk971  United States
diyAudio Member
 
xrk971's Avatar
 
Join Date: Oct 2012
Location: Virginia
Mguntrum,
Congratulations on your amp! It sounds wonderful doesn't it? It looks like the tsop 32 breakout board worked very well for your needs. Please share with us any lessons learned from this project as I think there are lots of folks wanting to try this amp.
Regards,
X
  Reply With Quote
Old 16th June 2013, 04:22 PM   #5
xrk971 is offline xrk971  United States
diyAudio Member
 
xrk971's Avatar
 
Join Date: Oct 2012
Location: Virginia
Mguntrum,
I soldered a thin 5 mm ring of copper hacksawed from half inch copper plumbing pipe onto the thermal pad of the TPA and it is sufficient to pull off excess heat. I have also used a real finned aluminum heatsink and can barely feel it getting warm so it is overkill. If you can get a copper stub column soldered to the pad and have it make contact with the case lid, that should be plenty of heatsinking.
  Reply With Quote
Old 17th June 2013, 03:20 PM   #6
diyAudio Member
 
Join Date: Jun 2011
Thanks for the comments! I was wondering about the heat sink, the chip doesn't run hot or go into thermal shut down, even after extended use. My speakers are 8 ohm so that limits the power output. After two hours of loud running the stomp box is barely warm to the touch.

I wasn't comfortable soldering directly to the chip and because I was worried the heat required to get the solder to melt to a heat sink might damage the chip. I was also worried the heat would melt the contacts with the board.

Due to the tight layout and didn't see a good way to clamp a heat sink to the chip. In retrospect it would have been easier to solder the heat sink on the chip first, then mount it. The thing to do would be as you suggested to mount an aluminum or copper piece to the bottom plate that would contact the top of the chip. I like your idea of using copper pipe section. I'll get right on that.


The funny thing is, soldering the chip to the board was by far the easiest part of the build. The board came pre tinned, all I had to do was put some jell flux on the legs and a half second touch with a chisel tip soldering iron, it was over in 10 minutes. The board was very reasonably priced.

TSSOP to DIP adapter 32-pin 0.65mm pitch



The hardest part of the build was getting everything to fit into a stomp box case. I was looking for the lowest cost and got most of my components from Tayda Electronics.

I did go with upgraded inductors. That seemed like a place you didn't want to scrimp.

2101-V-RC Bourns | Inductor Leaded 2101 V RC | Onlinecomponents.com

I really like the way it sounds, tight, fast, and accurate but non fatiguing is the best way I can describe it. This is the first amp I've used with no eq, straight from the player, and have been happy with the sound. Now I have to build better speakers!

I did read through the forum on the TI website, it was very helpfull, especially the schematic for the eval board.

When I was working on the output section I was going to add a C-RC filter, I added a 1nF snubb to the 680nF on the LC and liked what it did to the output. I stopped there because I'm happy with the sound.

I was afraid the cheap laptop power supply would be saggy so I added the biggest capacitors that would fit. At 100 Hz full power I was seeing almost no effect on the power supply voltage. The bass thumps, again tight, so I stopped there.

My background is vacuum tube guitar amps, this is my first solid state and I couldn't be happier with the TPA3116. I'm blown away with how efficient the little chip is, how few components it takes to run, and how good it sounds. I was all set to build a chip amp but for less than the cost of a good power supply I built the whole amp that I'm very happy with. You would swear after you hear how loud it is the little chip is magic.
  Reply With Quote
Old 17th June 2013, 06:23 PM   #7
diyAudio Member
 
Join Date: Jun 2011
I do have a few questions.

Should I install an inline fuse in the amp? I was thinking a 5 amp slow blow.

The ground is connected to the chassis at one point, I believe this is correct but wanted to check.

On the output, would a resistor across the speaker outputs help to control oscillations under a no load condition? Right now I have the 100V caps and have run it one minute with no load and nothing bad happened. Maybe 100 ohms? Something large enough it would draw minimal power.

My next amp is going to be a 4.1 for computer duty, with the sub amp integrated into the sub, so I would need two, two channel amps together. Should the amps be tied together so the switching is synchronous?

For my computer amp I only need 10-15 watts per channel so I was thinking to try a lower gain (20), is there anything I should look out for? Maybe just an internal L pad to drop the input instead with the standard gain (26) would work better.

Thanks
  Reply With Quote
Old 17th June 2013, 09:22 PM   #8
DUG is online now DUG  Canada
diyAudio Member
 
Join Date: Jan 2003
Location: mississauga ontario canada
Quote:
Originally Posted by mguntrum View Post
I do have a few questions.

...

On the output, would a resistor across the speaker outputs help to control oscillations under a no load condition? Right now I have the 100V caps and have run it one minute with no load and nothing bad happened. Maybe 100 ohms? Something large enough it would draw minimal power.

My next amp is going to be a 4.1 for computer duty, with the sub amp integrated into the sub, so I would need two, two channel amps together. Should the amps be tied together so the switching is synchronous?

For my computer amp I only need 10-15 watts per channel so I was thinking to try a lower gain (20), is there anything I should look out for? Maybe just an internal L pad to drop the input instead with the standard gain (26) would work better.

Thanks
I've run my TPA3116 on two different speakers...no issues

IMHO you should connect the sync lines... or use separate supplies.

I run my 3116 on 20 dB and an ipod will push it very well. Both with Yamaha 849's and Signet SL-16's.(my computer speakers)

I also run on a laptop supply at 19.5V but with the additional of tone control board ccts had to add some filtering from some parts I had laying around. (I did not notice the noise issues with only the amp.

I was fortunate that I designed a pcb for the 3116 and it had a good ground plane under it. I know that xrk971 built his first one "in air" and got good results with it but IMHO class d needs a good ground plane. Just a thought for your oscillation issue.

They do sound good though.

Happy listening.

__________________
Doug We are all learning...we can all help
"You can't stop the signal, Mal. Everything goes somewhere..."
  Reply With Quote
Old 17th June 2013, 11:02 PM   #9
diyAudio Member
 
radiosmuck's Avatar
 
Join Date: Jul 2012
Location: Montreal
Quote:
Originally Posted by ddapkus View Post
Hi,

I am the Audio Apps Manager at TI, and I'd like to congratulate you on a nice looking amplifier!

I suspect the reason that your output caps blew up without a load is due to a high voltage oscillation that will occur with an LC filter with no load. The load is part of the filter response, so when you remove it, you can get huge oscillations. See this app note for more info on LC filters: http://www.ti.com/lit/an/sloa119a/sloa119a.pdf

I noticed that you didn't use a heat sink... Under heavy load, this will result in thermal shutdown. You can easily add some metal to heat sink it, or you can switch to the TPA3118D2 which has a PowerPAD(tm) which allows you to solder the thermal pad on the bottom to your PCB so the
PCB acts as a heat sink. But, with your design, the PCB is pretty small, so I'd probably just add an external heat sink.

Another resource you can tap into is our engineer2engineer forum at TI. You can find it here: TI E2E Community

-d2
My Favourite speakers are 15 Ohm and they sound great on a tpa3123d2, tda7297 and tda7294 amps. There is never any mention in any of the documentation for the amps and their use with 15 Ohm speakers, why would that be?
All my amps were purchased as ready built boards.
  Reply With Quote
Old 17th June 2013, 11:15 PM   #10
diyAudio Member
 
theAnonymous1's Avatar
 
Join Date: Feb 2004
Location: Anonymityville
Quote:
Originally Posted by mguntrum View Post
I wasn't comfortable soldering directly to the chip and because I was worried the heat required to get the solder to melt to a heat sink might damage the chip. I was also worried the heat would melt the contacts with the board.
I've been using J-B Weld to glue on heatsinks for years. It works really well. I couldn't tell you what its thermal conductivity is, but it definitely works better than nothing!

Quote:
Originally Posted by radiosmuck View Post
My Favourite speakers are 15 Ohm and they sound great on a tpa3123d2, tda7297 and tda7294 amps. There is never any mention in any of the documentation for the amps and their use with 15 Ohm speakers, why would that be?
All my amps were purchased as ready built boards.
Probably because that's outside of the consumer norm for speaker impedance. With a 15-Ohm load there will be some peaking at the top end of the frequency response (with the class-d amps). If your speakers are fullrangers, maybe you don't mind that little extra boost up top.

Last edited by theAnonymous1; 17th June 2013 at 11:17 PM.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
FS- Twisted Pear Buffalo 32S Board & System New & Unused secretriches Swap Meet 1 10th August 2009 02:17 PM
Modify & build out (2) Western Electric 300B / 2A3 Hammond monoblock amps & UTC LS-5 quality sound Tubes / Valves 0 9th May 2009 03:59 PM
Krell KSA 80 bias & dc offset check & adjust procedure Clouseaupte Solid State 3 9th August 2007 09:39 AM
DIY-AMP <ZAS> Low THD & IM high bandwith murphy Solid State 18 30th December 2005 10:28 AM


New To Site? Need Help?

All times are GMT. The time now is 10:13 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2