"Do you have a suggested 10uF electro cap for the pre-TPA3255 side?"
Hi Lester: first of all I’m running SE input. I left the 4 stock bipolar UES caps that were in place when I received my board from you. Why not change them out as you and others have suggested? I would have put WIMAs there as well but space is extremely limited. But, a story first...
I also bought the updated 3e-audio TPA3255-2CH-260W (non-PFFB) earlier this year and was satisfied-to-happy with its performance but not dazzled. It supposedly came with Silmic IIs at pre-op amps and pre-3255 but they didn’t look anything like the Silmic IIs I have bought so I replaced them with new ones from Mouser. The sound was muffled even after letting them play for 50hrs or so. Then I swapped out the four pre-3255 with Nichicon KAs. The sound got a bit better but depth and soundstage wasn’t what was pleasing to me. I then swapped the 4 pre-3255 with UES bipolars and decided the soundstage on the mid-to-upper frequency range was better but not smooth and the pre-opamp Silmics still were exerting a bit of a muffled experience. At this point I just decided to try the WIMAs pre-opamp (since I had them) and the amp started to sound so much more musical, soundstage was better and I started to test op-amps to get a pleasing bottom end as eliminate any harshness on the upper end. Believe it or not 2209s or 33272s made the 3e sound really nice (on "my" system).
So I just wanted to try the WIMA arrangement on your board and I was happy with the result but really haven’t found a "best" of the op amps yet. I would have loved to use the really good oil-filled film caps pre-op amps but they are just way too big for the board and would loved to have gotten rid of the pre-3255 altogether but they are part of the PFFB loop but I didn’t want to fool that much with your wonderful design and because I’m running SE input I couldn’t bypass the whole op-amp design and go right to the 3255.
Regards,
Pete
Hi Lester: first of all I’m running SE input. I left the 4 stock bipolar UES caps that were in place when I received my board from you. Why not change them out as you and others have suggested? I would have put WIMAs there as well but space is extremely limited. But, a story first...
I also bought the updated 3e-audio TPA3255-2CH-260W (non-PFFB) earlier this year and was satisfied-to-happy with its performance but not dazzled. It supposedly came with Silmic IIs at pre-op amps and pre-3255 but they didn’t look anything like the Silmic IIs I have bought so I replaced them with new ones from Mouser. The sound was muffled even after letting them play for 50hrs or so. Then I swapped out the four pre-3255 with Nichicon KAs. The sound got a bit better but depth and soundstage wasn’t what was pleasing to me. I then swapped the 4 pre-3255 with UES bipolars and decided the soundstage on the mid-to-upper frequency range was better but not smooth and the pre-opamp Silmics still were exerting a bit of a muffled experience. At this point I just decided to try the WIMAs pre-opamp (since I had them) and the amp started to sound so much more musical, soundstage was better and I started to test op-amps to get a pleasing bottom end as eliminate any harshness on the upper end. Believe it or not 2209s or 33272s made the 3e sound really nice (on "my" system).
So I just wanted to try the WIMA arrangement on your board and I was happy with the result but really haven’t found a "best" of the op amps yet. I would have loved to use the really good oil-filled film caps pre-op amps but they are just way too big for the board and would loved to have gotten rid of the pre-3255 altogether but they are part of the PFFB loop but I didn’t want to fool that much with your wonderful design and because I’m running SE input I couldn’t bypass the whole op-amp design and go right to the 3255.
Regards,
Pete
Yes 2.2 or 3.3µF is the standard for input cap amp with 10k ohms (is a standard to)
Ok for the cap before opamps... the single supply for opamp is not the best way but can you do otherwise ?
Lester & AlexScan...thanks ..just ordered WIMA MKS 2.2uF to try them out as per your suggestions. The 4 pre-3255 caps will stay as the UES bipolars for now.
Pete
D
Deleted member 148505
Yes I think we have the same hearing perception, when I said thick cloth on Silmic II's the correct word is 'muffled'.
In fact the silmic is more muffled today than when I first soldered it. The highs now are more subdued. It is good for nearfield listening though because there are less treble.
With the Muse BP, the openness is there but not as smooth (slight distortion).
I tried my non-pffb module (inductor is same as 3eaudio module). Silmic II on pre-TPA3255, Muse BP on front end pre-opamps. Then tried NJM4580 (to simulate 3eaudio's default sound), it sounded balanced and open, very good 'HD' sound and bass. Only difference with PFFB is that it needs more smoothness in vocals. When I used all muse caps before with this non-pffb module, the opamp sounded very 'sharp'.
I think we have yet to find a good cap for the pre-TPA3255 section.
In fact the silmic is more muffled today than when I first soldered it. The highs now are more subdued. It is good for nearfield listening though because there are less treble.
With the Muse BP, the openness is there but not as smooth (slight distortion).
I tried my non-pffb module (inductor is same as 3eaudio module). Silmic II on pre-TPA3255, Muse BP on front end pre-opamps. Then tried NJM4580 (to simulate 3eaudio's default sound), it sounded balanced and open, very good 'HD' sound and bass. Only difference with PFFB is that it needs more smoothness in vocals. When I used all muse caps before with this non-pffb module, the opamp sounded very 'sharp'.
I think we have yet to find a good cap for the pre-TPA3255 section.
I forgot to mention I mounted 10uF Silmic IIs across pins 4 & 8 of the op-amp mount (on the underside of the 3e board) after some comment made by ppppp (spelling?) in some forum I can’t recall just now. I’m not going to try that on Lester’s board.
Pete
D
Deleted member 148505
We need to increase the PCB and add another switching regulator for negative rail + low noise negative post regulator.
Most sensible way is to create a separate board for buffer opamps.
We already have dedicated bypass electro for each opamp so I think if you will add bypass cap, you can place it on top of the opamp adapter board, just like in aiyima modification thread.
Yes Lester! Before I changed to the WIMAs on the 3e board, the op-amp I liked best (out of the 15 I roll through) was the 4580. Looking back on my notes I actually had 10uF UES bipolars on the front end and 10uF KTs (like KAs) pre-3255 and 4580 sounded natural, no harshness, an airy quality was present & bass response was good.
Pete
D
Deleted member 148505
Hmm have you tried LME49860? If you want 'airy' sound you'll like it. (but muffled when paired with Elna Silmic II compared with JRC4580) You can also disable PFFB on JLE board so that you can compare the sound with or without PFFB using the same opamp.
The PFFB pushes back the vocals and the instruments.
The PFFB pushes back the vocals and the instruments.
Last edited by a moderator:
D
Deleted member 148505
Guys, I read all your posts about Silmics... I have told you that Silmic dont belong to the signal path 
Electrolytes in the signal path are used only in cheap products...
Silmics make the harsh sound more rounded in positive way. But when you already have good sound, with Silmics it can for sure end up too veiled or muffled. Anyway, dont use them in signal path.
If you can't go with "no cap", the second best option is to use MKP (foil polypropylene).
As per Wimas (MKP), lots of folks uses them, lots dont like them. I belong to the second group But if you want to use Wima, use MKP not MKS.
Electrolytes in the signal path are used only in cheap products... Silmics make the harsh sound more rounded in positive way. But when you already have good sound, with Silmics it can for sure end up too veiled or muffled. Anyway, dont use them in signal path.
If you can't go with "no cap", the second best option is to use MKP (foil polypropylene).
As per Wimas (MKP), lots of folks uses them, lots dont like them. I belong to the second group
But if you want to use Wima, use MKP not MKS.JLesterP said:
Omg we have 65kHz switching SMPS?
So, in the end, isn't it better to use 10uH? I see farnell has 103KL in stock. I know, it is less linear etc. But... 10uH + 1uF has 50kHz freq.
10uH + 1,5uF has 41kHz freq.
Both with usable Q.
JLesterP said:
I have shielding tape (thin metal foil with adhesive layer, 5cm in width on a 5m long roll) for purposes like that. To create something like half farraday cage over AC parts. Or some 5x5cm cuprextit (proto PCB with full Cu layer without holes) placed between yellow TENTA and Nichicon and grounded. But perhaps it will be useless without full farraday cage. And without oscilloscope or spectral analyser it is only guessing.
Perhaps some zinc-plated steel conduit, hammered into shape, would work. In North America the readily available (and cheap) stuff is called EMT, and with its additional wall thickness vs tape I suspect it would be more effective at 50/60 Hz.
Steel Conduit Holds Steady | Electrical Contractor Magazine
As long as it's grounded, of course.
Steel Conduit Holds Steady | Electrical Contractor Magazine
As long as it's grounded, of course.
Last edited:
Agreed.. If they could have fit i would have used them but the cap area and lead spacing excluded MKP use ppppp. I tried shorting the pre-3255 caps on the 3e board but the fault LED lit up.
Pete
It's funny how I have always thought that the worst part about the SMPS, that needs shielding, is the SMPS itself, or at least its switching part. And now we are fiddling about how to protect the good SMPS from the bad AC
BTW, those "small but good input caps" are hard to find, I know... This is how it ended on my small TPA3116 amp
BTW, those "small but good input caps" are hard to find, I know... This is how it ended on my small TPA3116 amp
Attachments
Thank you sir
Off topic aside, I've heard somewhere that small film cap KEMET SMR is worth a try. 1,2uF has 5mm lead spacing and 7x7mm footprint, it could fit. Bigger values have bigger size, but we could place two of them on the top side of PCB and two on the bottom. From its datasheet: "The capacitor offers excellent sound quality in audio applications."
That and the green Muses are the only hopes I have for Lester's 3255 board, as there is really not much space. I also hoped that there will be possibility to use input transformers instead of opamps+caps, but if it will lit error LED, that option is lost...
But to make it a little more exotic, there is still good amount of space on the output LP side, for, lets say, 4x Mundorf Mcap EVO. If the zobels goes under the board
Off topic aside, I've heard somewhere that small film cap KEMET SMR is worth a try. 1,2uF has 5mm lead spacing and 7x7mm footprint, it could fit. Bigger values have bigger size, but we could place two of them on the top side of PCB and two on the bottom. From its datasheet: "The capacitor offers excellent sound quality in audio applications."
That and the green Muses are the only hopes I have for Lester's 3255 board, as there is really not much space. I also hoped that there will be possibility to use input transformers instead of opamps+caps, but if it will lit error LED, that option is lost...
But to make it a little more exotic, there is still good amount of space on the output LP side, for, lets say, 4x Mundorf Mcap EVO. If the zobels goes under the board
ppppp: I tried looking for the 1.2 uF Kemets but none of my usual suppliers has them in normal stock (Mouser, Digikey or Newark Electronics) but thanks for the suggestion. Dang!
I will try the low level Mundorf MKPs on the 3e audio board though. I use you mounting picture as inspiration
...and those big Mundorfs at the LC side of Lester’s board...me thinking that’s a stretch....LOL
Pete
I will try the low level Mundorf MKPs on the 3e audio board though. I use you mounting picture as inspiration
...and those big Mundorfs at the LC side of Lester’s board...me thinking that’s a stretch....LOL
Pete
- Status
- Not open for further replies.