thanks 🙂
so there is really something about the Volt+ board.
Their THD+N culminates at 0,03% for 1W in 4ohm and 0,02% in 8ohm.
To compares to the above measurements on the sanwu PBTL with the best inductor going beyond 0,1% .
Usual disclaimer: Difference can come from the device measured, the measurement process or both !
so there is really something about the Volt+ board.
Their THD+N culminates at 0,03% for 1W in 4ohm and 0,02% in 8ohm.
To compares to the above measurements on the sanwu PBTL with the best inductor going beyond 0,1% .
Usual disclaimer: Difference can come from the device measured, the measurement process or both !
Yeah Volt+ is good. Layout sucks big time though with regards to where the connectors are located. When I think of it that may be the reason it performs so well 😉
Last edited:
Aight, some aspects are not known for the AP measurements.
load (pure resistive or resistive+inductive)
wiring (load, power supply)
influence of AES17 filter
Thermals?:
I remember thermals have a greater effect than most imagine.
Possibly any pcb difference on thermals? I personally do not use them.
Yes, way more harder to assemble boards.
When assigned to CE duties, I had to remove thermals on some boards for them to pass assessments.
Just a crazy thought.
bruce
Attachments
Thermals , yes certainly. But I think you have to look at the whole and not the silver bullet. Its everything ..
We were not aware of such of a difference with the inductor though , we will do some testing on our boards and publish the results.
I think that TPA3118 can indeed become even better , I mean better THD+N etc on testing. I have a few new ideas . How about having as design goal : The best sounding TPA3118 ? What would you do different.
The best sound isn't always the best measurement. And "sound" is so much subjective, hard to judge/rate.
Design goal: best measured/speced amp with the possibility to add some k2 with just 1-2 changed parts. 😀
Design goal: best measured/speced amp with the possibility to add some k2 with just 1-2 changed parts. 😀
True ! Should have said that when I mentioned it to be good. No measurement equipment was hurt when stating this.
Last edited:
I stand by what I said...Design goal :Best sounding Tpa3118 (and yes that means that THD+N should be extremely good but not necessarily an end in itself...just a confirmation if you want)
Sound quality is very subjective, agreed. However in analog design there are ways to improve the SQ (while not necessary the thd) ...old school analog.
Seconded Doctormord, very interesting test. Thanks for sharing the results.
I'm really curious about something...
... having had great success with air-cored inductors giving a lovely sweet, natural sound I wonder whether you have considered THD testing an air-cored one?
I wound my own:
http://www.diyaudio.com/forums/class-d/289354-diy-air-cored-inductors-how-i-made-mine.html
here and they get very slightly warm, but cooler than I'd thought they'd be. Even on a 32V 7498.
I'd be fascinated to know if the greater resistance of the coil outweighed the benefits of the linearity of air. I suspect the amp THD should be more independent of voltage too but only the numbers would say for sure.
We end up with K2 domination anyway, unless when we use ribbon tweeter. AD modulation (tpa32xx) brings (tiny little bit) heavier K2 domination. But for excellent IMD we probably need lowest possible harmonic distortion ?
Thats one of the things I had in mind...recommended to me by a friend with "golden ears". I doubt you will see any thd improvements but apparently sound wise, it makes a difference.
Got some samples from Wurth to test.
Now we talking!
Quick test I have hard to tell Volt+ apart from Sanwu!
Have only made the 20db mod running it with 24v SMPS.
Very low noise, in same level as Volt+.
Adding a bad pic.
Now we talking!
Quick test I have hard to tell Volt+ apart from Sanwu!
Have only made the 20db mod running it with 24v SMPS.
Very low noise, in same level as Volt+.
Adding a bad pic.
I didn't know those green connectors existed with 2.5mm pitch.
That's a great fit for those tight boards.
I wrecked several by soldering cables on and off.
That's a great fit for those tight boards.
I wrecked several by soldering cables on and off.
Those green could connectors i use are 5.08mm which only fit with slight bending. It's more like 3.81mm for the output.
(2.5mm are the inputs)
(2.5mm are the inputs)
What do you think of the quality of these coils for the 3118? There are also other values.
ZML86 Lot of 4pcs Vishay Dale IH03BQ100K High Current Filter Inductor 10uH 10% | eBay
ZML86 Lot of 4pcs Vishay Dale IH03BQ100K High Current Filter Inductor 10uH 10% | eBay
The screw headers I used are 3.5mm.
If I did it again I'd use the 5mm ones or use 3.5mm plug headers. Reattaching wires to the 3.5mm is too fiddly and time consuming. The holes aren't round and having to carefully smooth and reattach the wires whenever I want to remove one of the amp boards is irritating.
Something like this....
https://www.rapidonline.com/camdenboss-ctb93hd-pluggable-side-entry-closed-headers-3-5mm-pitch-74990
J.
Reclock to 1200Khz made sharpness of sound MUCH better vs 400KHz. Temp rising is +10degree celcius max. Just try.
Last edited:
I can't attach pic to prev post. Next try: https://drive.google.com/open?id=0B3WNAAYdkPdYMU9veXkwU1BKSGc