Ya, these 330pF 10R thing.. anyone ever scoped the effects? Anyway, it's good to have them (I use 200p 10R), but they are mainly for EMI reasons.
Bigger input caps, hmm, depends also on your set gain and input impedance. 1uF at 20dB will mostly be sufficient. You "might" hear a difference between a ceramic and a film cap, if the first one creates more distortion at the lower end. (depends on quality) This is measured by a friend of mine and other. (it were ~0.2% at 300hz downwards) So if you go for something like WIMA MKT, you'll be fine.
Changing the bulk caps to OSCON, why not. (I personally don't like rolling caps without measurements)
In the end, this all won't increase performance if your PSU is a bad one but from my testing, you have to be very unlucky to find a bad one, nearly all laptop bricks (genue brand stuff) do good.
Personally, i use Liion batteries (stand alone 3 cells or with an additional stepup converter) or SLA.
At moderate levels the only "improvement" you might hear is a good film input cap if the actual one is really bad.
And if you notice some hiss, AVCC needs some filtering (10R + 1uF).
All said above is my personal experience and/or based on measurements.
(You might want to have a look on my website)
Regards.
Bigger input caps, hmm, depends also on your set gain and input impedance. 1uF at 20dB will mostly be sufficient. You "might" hear a difference between a ceramic and a film cap, if the first one creates more distortion at the lower end. (depends on quality) This is measured by a friend of mine and other. (it were ~0.2% at 300hz downwards) So if you go for something like WIMA MKT, you'll be fine.
Changing the bulk caps to OSCON, why not. (I personally don't like rolling caps without measurements)
In the end, this all won't increase performance if your PSU is a bad one but from my testing, you have to be very unlucky to find a bad one, nearly all laptop bricks (genue brand stuff) do good.
Personally, i use Liion batteries (stand alone 3 cells or with an additional stepup converter) or SLA.
At moderate levels the only "improvement" you might hear is a good film input cap if the actual one is really bad.
And if you notice some hiss, AVCC needs some filtering (10R + 1uF).
All said above is my personal experience and/or based on measurements.
(You might want to have a look on my website)
Regards.
I don't know what type of cap your 820uF/35V is but I noticed a huge difference when switching from 470uF/35V Panasonic FM to Panasonic 330uF/25V SEPF OSCON's. One of the biggest value bang-for-the-buck mods you can make to improve sound quality. Pricey (relatively to other electrolytic caps but nothing compared to boutique oil and paper hand rolled caps) at nearly $2 ea but worth it.
As DUG said, the main power cap mod is low ESR is good because it provides a higher transient current ability for bass and sharp transients like percussion.
The bootstrap snubber is simply following TI's evaluation board schematic and helps to remove the high frequency harshness. It makes an audible difference in smoothness and is a very cheap fix.
The larger input caps extend the low frequency cutoff of the amp to a solid 20Hz vs 40Hz (depending on gain setting). Better caps also improve sound quality for highs. There is some debate whether or not tiny 10uF SMT ceramic caps are better because they don't act as antennas and pick up RF like large metal foil polyester caps. I have used both and like both.
Changing a real Panasonic FM/FR 820uF for a fake Oscon 220uF will indeed improve cymbal performance of 3110 and 3116 ampboards.
A real Oscon will improve even a fake panasonic FM/FR.
(Don't worry most FM/FR sold in Germany also are fake I am told by very reliable source, so your FM/FR cymbals could sound okísh. Best order them from China to be certain. But never ever order them from Digikey/Mouser/Elfa, they sell the real, bad ones).
If you replace a real, low esr eg Rubycon or NCC/UCC 820uF/35V for a Oscon you are wasting money imo.
You really need to purchase 2 amps and mod one and perform a double blind test.
So you say I need Low ESR OSCON to have sharp transients (for percussion) but also need snubbers to not have sharp transients to prevent high frequency harshness?
@Irribeo, so you say I need to buy fake caps to get better sound?
----------
So to get this right:
Changing 820uF/35V Panasonic FM/FR to 220uF/35V Panasonic OSCON -> lower ESR -> sharp transients == good
Good sharp transients == bad because high frequency harshness -> place Snubber-RC -> limit sharp transients --> good
For the swap:
Real Panasonic FM/FR 820uF -> Fake OSCON 220uF == Improvement
Fake Panasonic FM/FR 820uF -> Real OSCON 220uF == Improvement
Fake Panasonic FM/FR 820uF -> "could sound okísh" == not need to "improve"
So changing real/fake panasonic FM/FR for real/fake OSCON is always an improvement, but fake panasonic FM/FR could sound okay.
Real Panasonic FM/FR 820uF == "the bad ones"
Fake Panasonic FM/FR 820uF == the good ones
Fake OSCON 220uF == the good ones
Real OSCON 220uF == the good ones
Real "real, low esr eg Rubycon or NCC/UCC 820uF/35V" == the good ones
Conclusion:
Real Panasonic FM/FR need to be blamed, but not Rubycon or "real low esr", because Panasonic FM/FR are not "real low esr".
Buying Panasonic FM/FR from reliable distributor is bad, because they sell real parts, better to buy fakes from china.
Got it?
@Irribeo, so you say I need to buy fake caps to get better sound?
----------
So to get this right:
Changing 820uF/35V Panasonic FM/FR to 220uF/35V Panasonic OSCON -> lower ESR -> sharp transients == good
Good sharp transients == bad because high frequency harshness -> place Snubber-RC -> limit sharp transients --> good
For the swap:
Real Panasonic FM/FR 820uF -> Fake OSCON 220uF == Improvement
Fake Panasonic FM/FR 820uF -> Real OSCON 220uF == Improvement
Fake Panasonic FM/FR 820uF -> "could sound okísh" == not need to "improve"
So changing real/fake panasonic FM/FR for real/fake OSCON is always an improvement, but fake panasonic FM/FR could sound okay.
Real Panasonic FM/FR 820uF == "the bad ones"
Fake Panasonic FM/FR 820uF == the good ones
Fake OSCON 220uF == the good ones
Real OSCON 220uF == the good ones
Real "real, low esr eg Rubycon or NCC/UCC 820uF/35V" == the good ones
Conclusion:
Real Panasonic FM/FR need to be blamed, but not Rubycon or "real low esr", because Panasonic FM/FR are not "real low esr".
Buying Panasonic FM/FR from reliable distributor is bad, because they sell real parts, better to buy fakes from china.
Got it?
Last edited:
Which are generally better, the small shielded or the big red inductors on these boards? Also, what are all the small electrolytics on this board for?
An externally hosted image should be here but it was not working when we last tested it.
Very closeSharp transients do not sound harsh.
And do not test good sounding German Panasonics with rated voltage while not wearing glasses. Seriously.
This all must be a joke!
I only buy legit real stuff at Farnell/Digikey/Mouser. Why i ever should buy faked stuff like Panasonic Caps? To support product piracy?
I'm sorry, but this really is big nonsens to me.
I'm out of this.
Yeah, if you give a crap about what you do as an electronics designer, you don't use fake parts.
Especially in an application like this. The TPA draws a fair bit of ripple current from PVCC, if your fake "FM" or "OSCON" cap happens to be a repackaged general purpose capacitor with a half amp of ripple current rating, it's not going to be reliable. And if the ESR sucks, it's not going to do a good job decoupling either.
Especially in an application like this. The TPA draws a fair bit of ripple current from PVCC, if your fake "FM" or "OSCON" cap happens to be a repackaged general purpose capacitor with a half amp of ripple current rating, it's not going to be reliable. And if the ESR sucks, it's not going to do a good job decoupling either.