TPA3116D2 Amp

Jump to Latest
#3,421
hajj said:
You meant 680nF (0.68uf) right?
Click to expand...
yeah lol, getting too tired of all those incorrect Q posts here and now even by TI employee 😀

BTW Same TI employee I think also has a tricky post about how to PBTL board I feel, he says it is best to not have Linp and Linn directly to ground, but better through a cap. Now how similar are those caps, does the chip still sees a signal on pins?? His answer seems directly opposite to datasheet and earlier answer by Don Dapkus, but who knows he might even have tried it🙂
 
#3,422
xrk971: does the breeze2.1 have a master and slave tpa3116? Are they sync'd correctly? John8 had a post around #340 I believe about value cap to look for, he also asked same question on TIforum and got confirmed that datasheet was incorrect here and that datasheet value used on YJ red 2.1 did not work.
 
#3,423
irribeo said:
xrk971: does the breeze2.1 have a master and slave tpa3116? Are they sync'd correctly? John8 had a post around #340 I believe about value cap to look for, he also asked same question on TIforum and got confirmed that datasheet was incorrect here and that datasheet value used on YJ red 2.1 did not work.
Click to expand...

I just checked the board and pin 16 of the stereo master TPA3116D2 IC is connected to pin 16 of the slave IC via a 103 (10k) SMT resistor and there is a 102 (1 nF) SMT cap connected from pin 16 of the slave TPA3116D2 IC to ground. This matches the TI recommended schematic in Fig 27 of the datasheet. I do not hear any beat frequency issues in listening to it so imagine that the master/slave circuit was implemented correctly.
 
#3,424
47pfd/50V
4.70K

These are the values EVM uses and that do work, john8 has posted about other values that also work, TI forum too. Problem with the mentioned 10k 1nF datasheet values is the slave doesn't get sync'd correctly, john8 had a frequency counter, slave chip works on 100khz with datasheet values and correctly at 400khz (if that is what master uses) with EVM values and a couple of other values, it is not that critical, just that datasheet values do not work LOL 10k resistor can work I remember, just capacitor needs to be changed
 
Last edited:
#3,426
xrk971 said:
How does one know if it is not working properly? Can I hear it or need an O-scope?
Click to expand...

Even a frequency counter I think, not sure. Slave chip doesn't follow masterclock now, don't know if it will have a steady frequency of 100khz or that it will vary very much, he doesn't see external clock, this is TIforum answer, they promised to correct it, but they didn;t LOL



TPA3116D2 in 2.1 master/slave mode


This question is answered



Posted by John Hill on Sep 18 2013 01:56 AM


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

Prodigy20 points

I believe there is an error in the figure 27 schematic for a typical application of the TPA3116D" used in 2.1 master/slave mode.
U2 is in slave mode, the sync output from U1 fed via R73, 10k to the sync input with a 1nF capacitor C41 as a slew rate limiter. This reduces the swing on the sync input to around 1V peak/peak which in insufficient to sync U2.
In order to get U2 to sync I suggest C41 is reduced to 100pF to allow a full swing and the correct slaving of U2 to sync the two amplifiers.



TPA3116D2 in 2.1 master/slave mode

Report Abuse






You have posted to a forum that requires a moderator to approve posts before they are publicly available.






All Replies


  • Posted by Paul C. Chen on Sep 18 2013 02:09 AM
    Verified Answer
    Verified by Paul C. Chen


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

    Mastermind31135 points

    Hi John,
    Yes the RC values need to be reduced. 4.7K Ohms and 47pF would work well. This is already noted for the next DS revision. Thanks.

    reg,
    Paul.




    Report Abuse





    You have posted to a forum that requires a moderator to approve posts before they are publicly available.






  • Posted by John Hill on Sep 18 2013 02:18 AM


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

    Prodigy20 points

    Thanks Paul, my question was answered in minutes.
    What is the purpose of the capacitor? I assume it reduces noise pickup.
    Regards,
    John



 
#3,428
I believe john8 saw distortion on scope and then verified with a frequency counter. Date question was september 2013, you can search this tread to find how he noticed. You might be the only person on this thread not terrified of SMD, you even did a build befor boards were available🙂
 
#3,429
I measured the audio distortion (with mic) from my sub before attempting to swap out the cap, then swapped it out with a 47pF 50 volt NP0 SMT cap that I had on-hand. Easy operation, despite the fact that the board was still in the case with just the lid removed and is surrounded by other components. The magnifying goggles and good tweezers were a must have. I then measured the distortion from my sub after the fact and within the levels of distortion that my speakers have (circa 2-4% THD in the range of interest) - I saw no difference. So either the different cap had no effect, or the improved distortion performance is much smaller than typical distortion levels that subs typically have. 🙂

Oh well... I guess I will rest better knowing that I have the right value in place and if called upon, the amp will have optimal behavior. I guess it is good to know that the amps are now sync'd. I guess unless you put a freq counter to it, you will never know.
 
#3,431
I wonder if switching the sync resistor from 100k to 47k is also required? I only swapped the cap from 1nF to 47pF at this point. I have to say that in listening to it it sounds fine so not sure if I should keep pursuing this. Anyone else have any experience fixing the sync line RC circuit error that TI had on Fig 27? It must be subtle because so many manufacturers built this circuit and did not catch it when listening to it. I suppose I could try probing the sync voltage at pin 16 of the slave amp? 2.0 volts peak-peak at 400 kHz is about 1.4 volts RMS on an AC volt meter? Wish I kept my O-scope. 🙁 Sold it a long time ago because thought I would hardly ever need it anymore.
 
Last edited:
#3,432
100pf with 10k had full swing according to john8, you can send him PM if he tried 47pf and 10k, I do believe he wrote exact value wasn't important as long as it was smaller, he just happen to have 100pf I believe.

there were some reports of distorted sound from 2.1, don't know if it was this long forumthread or another, don't know if they used 12v or 24v, if it was only on loud levels, no idea if switching frequency could be stable without a (clear) (master)clock, but seems something undesirable to have 🙂

you think they really care about how the board sounds? first 50 sold they care maybe, YJ even had same inductors on red board TI used, abrax was amazed TI used the same as YJ 🙂 are they still the same? they used nichicons, do they still? people ordered 1 of the wieners, liked it, ordered another and that one sounds different, surprise? they put fake oscons on the cheapest boards, do they care? maybe if they receive many complaints about sound 2.1 they look at it and probably change the gainresistors, switch to 2 masters, easier, less complaints, sync line will still be there because they ordered xxxx pcb's 🙂

what if they produced together 100.000 pcb's with parts mentioned by TI in datasheet, and TI changes datasheet because TI made a mistake, all of a sudden all pcb's produced are not according to datasheet, buyers have more reason to feel something wrong and maybe better, TI might be to blame and who pays costs of changing 60.000 pcb's if TI is responsible...? Revisions like this can take time just for legal reasons.
 
#3,433
Irrebeo,
I hear ya... 🙂 if it sounds decent to the masses (and I must admit that for $13 you could do worse for an amp that sounds this good) then good enough to keep making money. Incidentally, a friend of mine in China says the Danzz blue/black board sells there for 60 Yuan or about $11 USD! How do you build that amp for that much money? I will do a simple mod of slapping a second 100k resistor on top of current one to get 50k in parallel and see if that changes anything.
 
#3,437
Help!,Can I make this 2.1 into mono,I want to use it as a sub woofer amp,,this is the first one I bought and never liked the sound so I figured if I can get enough power from it maybe I can use it on my TDA7492 project,Any one know if this is possible,oh and the sub is 2 ohm 2 10"

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.
 
#3,438
#3,439
My YJ Danzz "Blue/Black" 2.0 board just arrived today from HK. Hooked it up to my Nautaloss II system (http://www.diyaudio.com/forums/full-range/247598-nautaloss-ref-monitor.html) for the tops via a miniDSP for the XO, and the TPA3116 board designed by Dug for the subwoofer. Just put on a few test tracks, very happy with the sound so far. Very detailed and great soundstage. I will listen to it in stock form for a few days before modding it. I have the 330uF OSCON's and some 3.3 uF input caps planned as upgrades.

Just wanted to thank Danzz for a fantastic design!
 
You must log in or register to reply here.
Top Bottom