one snubber for each pbtl....this is what i'm thinking but...but not confident to do so unless someone verified for me
this is what i have done to my pcb...so which one should i remove or cut ?
An externally hosted image should be here but it was not working when we last tested it.
(You don't need nylon standoffs with YJ's copy). BTW heatsink is also not connected to gnd, which is a little strange. I would remove a little green paint pcb screen above two resistors pointing to chip and solder the snubber directly to gnd. And remove wires and horizontal snubbers.
Thanks for the input.
I have read through some of the 784 pages in this forum☺ and also regarding the “pop” and “click” sound.
I have made a switch to operate the “mute” and “SDZ” function on the board by pulling the mute high and the SDZ low and it works great. All this according to the TI datasheet. The SDZ works great when powering down. Also what TI has written in the datasheet under section 7.3.3.
A very good turn on/off sequence is here showed from ST TDA7294 amp. That is very useful and does also work very well on the TPA3116 and removes turn on/opp pop. Please see attached picture:
I have also tested the displayed board on my B&W S604 8Ω speakers. I have tested the board @:
- 13.3 V DC car battery supply (in supply mode, NOT in charge mode)
- 24 v DC boost converter (boosted from the 13.3 v battery supply).
I had hoped for a little more power when powering the board at 24v (33 W 1% THD VS. 12w 1% THD) but it was hard to hear a big difference. The bass was a bit richer through.
Does anyone have other experience regarding powering the tpa3116 at different supply voltage?
I was very impressed that this small Digital amplifier could power the big speakers and also to a very loud level ☺
I did buy two of the above displayed boards (one of then burned the ship after 2 hours of use) and both of them had the static noise when turning the pot to adjust volume.
The diode does get hot when powering the amp using 4 Ω speakers. Is 48 °C to hot when operating at mid to high volume?
I have read through some of the 784 pages in this forum☺ and also regarding the “pop” and “click” sound.
I have made a switch to operate the “mute” and “SDZ” function on the board by pulling the mute high and the SDZ low and it works great. All this according to the TI datasheet. The SDZ works great when powering down. Also what TI has written in the datasheet under section 7.3.3.
A very good turn on/off sequence is here showed from ST TDA7294 amp. That is very useful and does also work very well on the TPA3116 and removes turn on/opp pop. Please see attached picture:
An externally hosted image should be here but it was not working when we last tested it.
I have also tested the displayed board on my B&W S604 8Ω speakers. I have tested the board @:
- 13.3 V DC car battery supply (in supply mode, NOT in charge mode)
- 24 v DC boost converter (boosted from the 13.3 v battery supply).
I had hoped for a little more power when powering the board at 24v (33 W 1% THD VS. 12w 1% THD) but it was hard to hear a big difference. The bass was a bit richer through.
Does anyone have other experience regarding powering the tpa3116 at different supply voltage?
I was very impressed that this small Digital amplifier could power the big speakers and also to a very loud level ☺
I did buy two of the above displayed boards (one of then burned the ship after 2 hours of use) and both of them had the static noise when turning the pot to adjust volume.
The diode does get hot when powering the amp using 4 Ω speakers. Is 48 °C to hot when operating at mid to high volume?
Turn on/off sequence picture:
An externally hosted image should be here but it was not working when we last tested it.
Thanks!
Hmm. heres pic i just took. I need to check those elcos between heat sink and red "wima"´s
Does this pic give more opinion for mod process?
Concerning those 5pcs elcos. They are nearest to chip. (EDIT: Those elcos are 2.2uF / 50V)
An externally hosted image should be here but it was not working when we last tested it.
Hey Oceanw! I know it's a giant thread, but have a look over at page 500-505 somewhere in that range, as member syklab and myself have done some modifications to this board.
Most of the components on the board aren't "genuine", but the inductors as mentioned before are decent in stock form. I haven't done anything to the output filters, but here are some of the mods I did.
The 10x capacitor array there, I used Panasonic FMs there, had a nice improvement right off the beginning. I swapped out the WIMAs for some 4.7uF Polyester film caps also from Panasonic. This gave some nice bass improvement.
As for the small electrolytics between the chip and the WIMAs, Buehgemeiste is correct, there's one in the middle that's used for power decoupling, where as the others are used for signal decoupling. Since space is a little tight, I think some 2.2uF are the ones that will most likely fit in there. I went with some Nichicon Fine gold in there, of the identical size as the elcos that are currently in there.
I also added bootstrap snubbers underneath the board, there's a grounding plate underneath that is good to work with, you can easily solder some legs and such to it without having to scrape away PCB paint etc.
Overall I like working with this board, I kinda wish the gain resistors werent SMD but what can you do. Let me know if you have any questions, and if anyone else has any comments or recommendations, thatd be great to hear too!
just one 1uF for decoupling, the middle, skip all others, no need for series electrolytics after filmcaps if you increase value
Just recieved my Blue/black YJ Only! took 5 weeks 🙄
Been reading the thread (looong one too)
Seems most All the Mods are those done / tried on the Previous D amp 'craze'.. The Ta2020.
Was there, did that..ubiquitous Fake Tripath chips soon ended that though.
As result I have a supply of V nice 100mh 5 amp inductors and a Supply of Ceramic caps. Also Many input caps of the appropriate guessing ranges all as remainders from those D amp adventures.
Though in truth and with the benefit of 20/20 Hindsight 😉
Few of the mods really did... Much.. at all.
Although I was absolutely convinced otherwise at the time.. delusionally it seems.
Sooo.. Following in the above history/scenario:
What Mods 'genuinely' make a difference on these?
Surely after 700+ pages the Chaff has settled.
I have 12v, 19v and 24 v 'Genuine' laptop SMPS bricks.
Which one, or try em all ?
Been reading the thread (looong one too)
Seems most All the Mods are those done / tried on the Previous D amp 'craze'.. The Ta2020.
Was there, did that..ubiquitous Fake Tripath chips soon ended that though.
As result I have a supply of V nice 100mh 5 amp inductors and a Supply of Ceramic caps. Also Many input caps of the appropriate guessing ranges all as remainders from those D amp adventures.
Though in truth and with the benefit of 20/20 Hindsight 😉
Few of the mods really did... Much.. at all.
Although I was absolutely convinced otherwise at the time.. delusionally it seems.
Sooo.. Following in the above history/scenario:
What Mods 'genuinely' make a difference on these?
Surely after 700+ pages the Chaff has settled.
I have 12v, 19v and 24 v 'Genuine' laptop SMPS bricks.
Which one, or try em all ?
Just recieved my Blue/black YJ Only! took 5 weeks 🙄
Been reading the thread (looong one too)
Seems most All the Mods are those done / tried on the Previous D amp 'craze'.. The Ta2020.
Was there, did that..ubiquitous Fake Tripath chips soon ended that though.
As result I have a supply of V nice 100mh 5 amp inductors and a Supply of Ceramic caps. Also Many input caps of the appropriate guessing ranges all as remainders from those D amp adventures.
Though in truth and with the benefit of 20/20 Hindsight 😉
Few of the mods really did... Much.. at all.
Although I was absolutely convinced otherwise at the time.. delusionally it seems.
Sooo.. Following in the above history/scenario:
What Mods 'genuinely' make a difference on these?
Surely after 700+ pages the Chaff has settled.
I have 12v, 19v and 24 v 'Genuine' laptop SMPS bricks.
Which one, or try em all ?
Have you read the wiki? The essential mods are there.
1. Main power rail caps switch to high quality name brand ones with low ESR.y favorite is Panasonic SEPF 330uF 25V
2. Bootstrap snubber 330pF and 10R before L inputs.
3. Main input caps switch to quality film 4.7uF
4. Film caps for bootstrap caps vs old ceramics
5. High current Coilcraft inductors
6. Input trafo for balanced inputs to amp
3. and 5. might not be that useful. (3. depends on gain and desired lower x-over-frequency, 5. depends on actual inductors and desired output level)
(2. is a rule of thumb [or myth] as this highly depends on board layout and filter values)
But you won't do wrong, even if it's just to burn money. (If you do the math and some measurements, you will if it's needed. If you just believe, everything is worth the money.)
Just sayin.
(2. is a rule of thumb [or myth] as this highly depends on board layout and filter values)
But you won't do wrong, even if it's just to burn money. (If you do the math and some measurements, you will if it's needed. If you just believe, everything is worth the money.)
Just sayin.
Last edited:
Please could you be more specific for all the rest of the members? For example about mod nr. 3. Thank you!3. and 5. might not be that useful. (3. depends on gain and desired lower x-over-frequency, 5. depends on actual inductors and desired output level)
(2. is a rule of thumb [or myth] as this highly depends on board layout and filter values)
But you won't do wrong, even if it's just to burn money. (If you do the math and some measurements, you will if it's needed. If you just believe, everything is worth the money.)
Just sayin.
Please could you be more specific for all the rest of the members? For example about mod nr. 3. Thank you!
Well. Different gain settings produce different input impedances. So a fixed value input decoupling cap, say 4.7uf, would produce different HPF -3db cut off freq with each gain setting. That's also how it would effect x-over.
Input cap values, as well as the coils, and power decoupling caps, are a matter of personal choice when it comes to it.
I do happen to use 4.7uf with 26db gain as that's the -3db fc I need, well near enough as ideally I would need 5uf.
I use 470uf 35V power decoupling caps, not the 330uf suggested., nor Panasonic's. Anythings better than the oversized stock 1000uf. lol.
Last edited:
Well. Different gain settings produce different input impedances. So a fixed value input decoupling cap, say 4.7uf, would produce different HPF -3db cut off freq with each gain setting. That's also how it would effect x-over.
Input cap values, as well as the coils, and power decoupling caps, are a matter of personal choice when it comes to it.
I do happen to use 4.7uf with 26db gain as that's the -3db fc I need, well near enough as ideally I would need 5uf.
I use 470uf 35V power decoupling caps, not the 330uf suggested., nor Panasonic's. Anythings better than the oversized stock 1000uf. lol.
Thank you for your quick reply. I understand again that I don't know much in this field 🙂
So I will use the board with I guess 20 db (100k resistor removed) with some 3 way open baffle with passive XO at 400hz and 3k. What are the values that you suggest for them?
Thank you for your quick reply. I understand again that I don't know much in this field 🙂
So I will use the board with I guess 20 db (100k resistor removed) with some 3 way open baffle with passive XO at 400hz and 3k. What are the values that you suggest for them?
For 20db gain setting your input impedances is 60kohm. Value of your -3db cut off of the input RC depends on how low a bass freq signal you want unimpeded!
Using the 60kohm, try out this calculator:
Coupling Capacitor Calculator by V-Cap
Using the 60k the ideal value for 20 hz optimum response would be 1.3uf. Obviously needing a 1uf or 1.2uf near as can get. It will all depend on how low your bass needs are.
For example
Personally, I use the 26db gain(not suggesting any better) which has 30kohm impedance. I use 4.7uf to have optimum response as low as 11hz. But that's because I am using a pllxo before amp, as I don't like any xo between amp and speakers. So my pllxo makes the real decision as to what my freq slopes look like entering amp.
As my OB's are 2ways, the full range driver goes down to around 40hz comfortably at 1/2 volume (rated @ 20W 1/2 volume). Over 1/2 volume and the 8" drivers start to weaken in performance. So I use the pllxo to keep under 40db suppressed slightly, so I don't need the amps input RC to shape the freq response of the input at all!
It doesn't hurt to have bigger input caps even if you are at lower gain. The quality may be important here. However there are also arguments that using a big cap may act as an antenna for RF pickup (vs tiny SMT caps). I have not had an issue with RF pickup with my 3.2uF Panasonic film caps.
The first 2 mods are probably the most bang for the buck and result in immediate audible improvements.
The first 2 mods are probably the most bang for the buck and result in immediate audible improvements.
It doesn't hurt to have bigger input caps even if you are at lower gain. The quality may be important here. However there are also arguments that using a big cap may act as an antenna for RF pickup (vs tiny SMT caps). I have not had an issue with RF pickup with my 3.2uF Panasonic film caps.
The first 2 mods are probably the most bang for the buck and result in immediate audible improvements.
Agreed. Just pointing out that size choice of input caps can also be to suited to system requirements of desired LF response.
See, if I was to forget my pllxo and just use a hf xo between amp and tweeter I could rely solely on an amp input RC for the 8" driver. Using only a 1.2 or 1.6 uF with the 30kohm to set the LF at suitable level for the driver, would probably save some energy that way too!
All I can say is its easier to find the smaller value caps at low cost compared to bigger values.
And use balanced inputs.
Last edited:
It doesn't hurt to have bigger input caps even if you are at lower gain. The quality may be important here. However there are also arguments that using a big cap may act as an antenna for RF pickup (vs tiny SMT caps). I have not had an issue with RF pickup with my 3.2uF Panasonic film caps.
The first 2 mods are probably the most bang for the buck and result in immediate audible improvements.
Xrk.... Number 4..... Question.
Do you know whether polycarbonates have been tried as bootstraps? If so, how did they fare?
Hey Oceanw! I know it's a giant thread, but have a look over at page 500-505 somewhere in that range, as member syklab and myself have done some modifications to this board.
Hi jk44, and thank you for your suggestions!
Below some questions.
I swapped out the WIMAs for some 4.7uF Polyester film caps also from Panasonic. This gave some nice bass improvement.
1.) I have 10uf/100V Panasonic (ECQ-E1106JF) polyester caps. I think they could fit fysically but do you think they are usable in this place?
As for the small electrolytics between the chip and the WIMAs, Buehgemeiste is correct, there's one in the middle that's used for power decoupling, where as the others are used for signal decoupling. Since space is a little tight, I think some 2.2uF are the ones that will most likely fit in there. I went with some Nichicon Fine gold in there, of the identical size as the elcos that are currently in there.
2.) Pana 10uf/100 meantioned above, how about its usability in here?
You changed all 5 pcs to Nichicon FG ?
I also added bootstrap snubbers underneath the board, there's a grounding plate underneath that is good to work with, you can easily solder some legs and such to it without having to scrape away PCB paint etc.
3.) I must confess that idont get bootstep snubber thing. If i take pic from underside of board could you address soldering spots?
---
I checked past posts you addressed, member syklab had soldered 2.2uf oscons directly to both ends of chip. That option tempts me also. Can i use again in here my 10uf/100V panas or oes it have to be elco and preferably 2.2uf?
Thanks again for help!
3. and 5. might not be that useful. (3. depends on gain and desired lower x-over-frequency, 5. depends on actual inductors and desired output level)
(2. is a rule of thumb [or myth] as this highly depends on board layout and filter values)
But you won't do wrong, even if it's just to burn money. (If you do the math and some measurements, you will if it's needed. If you just believe, everything is worth the money.)
Just sayin.
Thank you sir 🙂 Interesting informations
Yess I've read the Wiki
But based on experiences with the Ta Wiki I'm reluctant to treat it as Gospel 😉
Clearly I've reaquired the 'Fiddle with it' Bug.
But this time I'm trying to be circumspect/selective in my foolishness.
Agreed: fitting Foil Coils is erm.. Too odd, not gonna happen.
Varying input uf did make differences in Ta amp boards, but results were varied and subject to personal preferences/perception.. will fiddle though.
Input Cinemags Could work.. But C'mon? this is a 12$ Chi Fi board, Not a Firstwatt build
Today's problem is a passable container/mounting so I can fire the rascal up.. without risk of shorting anything.
Hi jk44, and thank you for your suggestions!
Below some questions.
1.) I have 10uf/100V Panasonic (ECQ-E1106JF) polyester caps. I think they could fit fysically but do you think they are usable in this place?
2.) Pana 10uf/100 meantioned above, how about its usability in here?
You changed all 5 pcs to Nichicon FG ?
3.) I must confess that idont get bootstep snubber thing. If i take pic from underside of board could you address soldering spots?
---
I checked past posts you addressed, member syklab had soldered 2.2uf oscons directly to both ends of chip. That option tempts me also. Can i use again in here my 10uf/100V panas or oes it have to be elco and preferably 2.2uf?
Thanks again for help!
Hi Oceanw,
Yes I have used the 10uf Panasonic in the same spot before on another board, however the fit may be a bit tight. The yellow board that we have has some rather small holes for capacitor legs, so just be sure get the sizes right.
If you read irribeo's post above, the 2.2uf caps (elcos) between the big input caps (currently WIMA) and the chip won't be necessary, if you use the 10uF caps at that spot. You will need to keep the single 2.2uf there, as I have seen other members do.
As far as the oscon's at the ends of the chips, you can't really use the panasonic film caps there, this should be done with electrolytics or the aluminum-polymer caps (like the oscon). Since these are your power decoupling aps
- Home
- Amplifiers
- Class D
- TPA3116D2 Amp