The SMSL SA-36A Pro has what appears to be ceramic SMD caps on the "-" differential inputs connected to Ground. I'll check the capacitance of those two caps tomorrow. My guess is that the power-off pop is due to a combination of the 3.3uF input capacitance and the slow discharge the 8x Panasonic M electrolytic caps used for power supply filtering. Ever since I put the Panasonic OSCON caps on power supply decoupling positions, the amp has been sounding great as a standalone integrated amp driving my Klipsch Forte II speakers.
Don't forget the subject of this tread, it is known to disrupt FMtuners, DABtuners, cdplayers, ipods, wireless mouse and keyboards and probably a lot more 🙂
I used the Wimas to replace the two larger input caps and for the smaller caps I used Panasonic ECWFD caps, as they were small and fit well.
They sound very good with the Wima.
ECFWD
Thanks for the tip on the Panasonic ECWFD MKP caps. The lead spacing for the 2.2uF/250V Wima MKP10 caps is larger than the 22.5mm thru-hole spacing on the Yuan Jing blue amp. How did you connect the caps to the board? Can you post a photo?
I believe you are right about a combination of both imbalance input cap values and slow discharge power supply caps. I left my PS output cap at 330uF.
Ceramic SMD may not be as large as 3.3uF (correct me if I'm wrong as I have not seen one with this big value).
Do you recommend go straight to OSCON ? Are they much better than Panasonic FM of same value ? The pic of the OSCON in digikey doesn't look like what's in the pics you post ?
- AC
Last edited:
No, I won't 🙂
But hey, guess what - with power filter and all stuff I have mentioned before there is no such issue by me. Despite the fact my computer/wi-fi/powerline sitts next to my audio gear.
Well, not now, but who knows what might come out tomorrow when I finally fix that 3116 amp (from the thread) 😉
Last edited:
The 27.5mm spacing on the cap worked with the MKP10 leads (even though they are very short), with careful use of needle nose pliers. If I took a photo, you would not see anything as the leads are completely covered by the cap.
Okay. I understand how you mounted the caps now--great idea. I'll have to order some caps from Mouser and give it a try. I really like the Wima MKP10 Polypropylene caps, and the 2.2uF capacitance and size might be the best solution for me.
I'll need to measure the ceramic caps on my SMSL SA-36A Pro amp. I've been up early this morning after a busy work week spinning vinyl. It just sounds so good with my tube phono stage that I don't want to shut it off.
With the 330uF/25V Panasonic SEPF OSCON caps, the amp is very musical and the soundstage is significantly larger than with the stock 47uF no-name hybrid or organic semiconductor caps (not sure what they were). The presentation is detailed and three dimensional. Listening to an original copy of The Rolling Stones' "Sticky Fingers" right now and the amp is jamming with Keith Richards' guitar. Earlier, I listened to the new Analog Productions reissue of the RCA Living Stereo LP of Fritz Reiner and the Chicago Symphony performing Resphigi's "Pines of Rome" and the new 200g reissue of King Crimson's "Court of the Crimson King." When I can enjoy a wide range of music I feel pretty good about the system.
http://industrial.panasonic.com/www-data/pdf/ABD0000/ABD0000PE141.pdf
I have a question about the ecwfd caps mentioned. Is the impedance of filmcaps always so high on lower frequencies, is that the way they block DC?? 100hz looks high too, does that matter on input?
I have a question about the ecwfd caps mentioned. Is the impedance of filmcaps always so high on lower frequencies, is that the way they block DC?? 100hz looks high too, does that matter on input?
Just picked up the Atlantic 180g reissues of Led Zeppelin I and Led Zeppelin II. Sounds great through the SMSL SA-36A Pro amp and the Klipsch Forte II's. Absolutely no shortage of bass and slam, especially on "Whole Lotta Love."
Is there enough interest in a TPA3116 / 3118 PBTL bare board for a vendor sale?
I do not want to hijack this thread.
Reply only by PM to me...have room for about 60.
Don't ask about features...that will take place in another thread.
The board will have some options for those that wish to play with different components.
It will be SIMILAR to JPG attached.
Doug
I do not want to hijack this thread.
Reply only by PM to me...have room for about 60.
Don't ask about features...that will take place in another thread.
The board will have some options for those that wish to play with different components.
It will be SIMILAR to JPG attached.
Doug
Attachments
Hey guys, i just ordered a board from yj and i had a question about upgrading the dc coupling caps. Does voltage rating matter at all? Im thinking of getting 2.2uf wima caps; mouser has the big ones at 100v and the small ones at 50v. Would that be fine?
Also, if i up the gain to 32db, would i need a different pot than the panasonic 50 kohm one the wiki recommends?
Also, if i up the gain to 32db, would i need a different pot than the panasonic 50 kohm one the wiki recommends?
Well I have always believed the rule of thumb is that the input impedance of an amplifier should be minimum 4 times greater than the output impedance of the preceding stage (the pot in this case). Given the input impedance is 9k at 36db gain that would be a rather small pot! Personally I would go for a 5k pot assuming the output impedance of the DAC or whatever you are feeding it is low enough. Typically a DAC output impedance is around 500 ohms or less so 5k should be OK.
Of course matching impedances is not quite that simple, varies with volume setting etc..
Those inductors look to be wound with fairly large gauge wire = lossy at HF. I wonder what the quiescent current is?
If the imput should be 4x greater, then why does the wiki recommend a 50k pot (recommended by rhing). Im assuming it would be for the stock gain of 26db which would put the imput at only 30k.
The 50k Panasonic EVJ has worked well in my TPA3116 and TPA3110 amps. You can try 20k. I just happened to have a bunch of these from Digikey.
wiki also quotes rhing from audiokarma with an unfortunate mixup about outputfilter inductors, the blueboard 10uH and 680nF has Q = 1.47512711317 for 8 ohm, that is not considered limited peaking by TI themselves in their application report. The redboard 22uH and 680uH filter has Q = 0.994530496813 for 8 ohm, that is considered limited peaking.
The 10u 680uF filter has Q = 0.737563556583 for 4 ohm, that is considered ideal.
The 10u 680uF filter has Q = 0.737563556583 for 4 ohm, that is considered ideal.