Quote:
I have added the snubber directly on the outputs as in the ap note. The R is 1206 and the C is 0805.
Features were put there for experimentation or personal preferences.
Pre-LC filter snubber: Directly on the IC outputs...same node as bootstrap caps and inductors connect.
see post 73 for board view: http://www.diyaudio.com/forums/group-buys/257996-gb-tpa3116-8-pbtl-bare-pcb-8.html
I have added the snubber directly on the outputs as in the ap note. The R is 1206 and the C is 0805.
Features were put there for experimentation or personal preferences.
Pre-LC filter snubber: Directly on the IC outputs...same node as bootstrap caps and inductors connect.
see post 73 for board view: http://www.diyaudio.com/forums/group-buys/257996-gb-tpa3116-8-pbtl-bare-pcb-8.html
Last edited:
Please keep us posted on the results of your planned mods.
Years ago, I compared the sound of using all balanced connections from my CD player to preamp to amp with all single ended and really did not find much discernible differences. Granted the IC connecting the components are different (between the SE and BAL)and the output stages of the CD players (balanced and SE) are different too. So the comparison might not be totally valid. The main purpose of using balance connection is to reduce noise due to long IC (critical in pro-audio applications). In home system, the IC connectionis relatively short.
By the way, my preamp is a TVC that has both SE and balanced connection and I am using the SE connection with my Audiobah board at this point. Sounds really good. My next plan is to mod the YJ Blue/black board. Interested to see what are the "most effective mod" from everyone's experiments.
Regards,
T
Sorry, I did not get them from Mouser. You can try order from Coilcraft directly at its website.
Regards,
Happy to help Rich. I've learnt pretty much everything I know from others on the forum. Here is a sketch showing the all of the LL1540 and audiobah AC1308 amp board connections:
Attachments
Will do, parts for mods 1, 2, and 3 have been ordered... Thanks for sharing your experience with balanced inputs. I think it may have more to do with some inherent 'filtering' that the small signal transformer provides. The removal of the existing inductors is tricky as I do not have a dual prong "tweezer" style soldering iron and will have to resort to prying and applying pressure while heating one side.
Today I soldered the smd input and bootstrap capacitors within 3-5mm on the YJ black board chip and that removed the rest of the ringing. I added back the large 1000uF silmics and and ringing were still gone 🙂 I shortened down the input traces and soldered on the input wire on underside of the board. I was a bit sceptical abaout using smd input caps, but it seems that 1uF 50V X7R plays pretty decent. I still use a cheapo rca cable and a bad psu in my test setup, but it now sounds like a decent amp!
Changed the 2 Panasonic 330uF 35V power cap
with 2 old Solen 22uF I had for 10 years.
Need to listen more tomorrow.
with 2 old Solen 22uF I had for 10 years.
Need to listen more tomorrow.
It could be. I always find that when I use the TVC as preamp, the background is quieter or "darker" and the music becomes a bit more vivid.
Yeah, desoldering the existing inductors could be tricky. I think someone reported that too. I was hoping one of those "dual prong tweezer" style of desoldering iron (specifically for SMDs) you described would make life easier. Apparently not, based on your experience. I am always worry about destroying the copper trace on the PCB. Look like Hiamp's board is the easy way out, albeit at higher cost.
Regards,
I didn't get around to installing my toroidal ferrite-core inductors as planned, but I did discover a separate issue with my board; heat. With an ambient room temperature of 30C yesterday and the board installed in a (passively-cooled) chassis with SMPS, the heat sink on the TPA3116 was painfully hot, and the stock inductors were almost painfully hot.
I've remedied this by installing active cooling in the chassis, so now I am right back to where I was with the actively-cooled Sure STA508. The whole point of any of these amps, for me, is to have a quiet (passively-cooled) "summer" amp that doesn't dump heat into the room like my tube amp does. Apparently, I will need to engineer a proper passive cooling solution, regardless of which class-D chip I choose. Forecast is showing typical Kamloops summer heat of 35-37C over the next few days. 🙂
I've remedied this by installing active cooling in the chassis, so now I am right back to where I was with the actively-cooled Sure STA508. The whole point of any of these amps, for me, is to have a quiet (passively-cooled) "summer" amp that doesn't dump heat into the room like my tube amp does. Apparently, I will need to engineer a proper passive cooling solution, regardless of which class-D chip I choose. Forecast is showing typical Kamloops summer heat of 35-37C over the next few days. 🙂
Last edited:
Cogitech,
How loud are you running your amp? They should not be getting that hot. Mine is barely warm at normal listening levels. It may be the YZ blue/black and certain combinations of caps/inductors that makes it hot. On my Ybdz board, it never even gets warm - you can't feel any excess heat.
X
How loud are you running your amp? They should not be getting that hot. Mine is barely warm at normal listening levels. It may be the YZ blue/black and certain combinations of caps/inductors that makes it hot. On my Ybdz board, it never even gets warm - you can't feel any excess heat.
X
Not very loud. When I checked it yesterday, it had been playing basically all day long at quite low volume. I couldn't keep my finger on the heat sink more than 1 second and then I said a curse word. The coils weren't quite as hot, but still much hotter than they should be, IMO.
Last edited:
YJblue gets warmer then all other 3116's, without music inductors get frightingly warm even. But even on YJblue I never felt a hot heatsink, I did read about that once befor in this tread. Finally it turned out the gainsetting resistors were wrong values I believe, better check.
Thanks. I will check those resistors for sure. That is an easier solution than installing heat pipes and/or water cooling. 😀
Befor adding additional snubber to YJblue, check YJblue version of snubber on pcb, it isn't same values as in TI datasheet....or maybe they ran out of those parts too when they built my board. Wiener board snubber is same as in datasheet, pt2313 3116 remote amp same.
De-soldering the inductors was easier than expected - but that was using one of those desoldering hot air stations used for smt parts.
Warm heatsink/inductors means something is wrong - they are dead cool with my amp - no matter if idling or playing music.
Warm heatsink/inductors means something is wrong - they are dead cool with my amp - no matter if idling or playing music.
All 3116 and 3110 are cool, except YJblue. Problematic parts everywhere I fear.
How long do you have YJblue? What brand of electrolytic caps did it have originally?