I, too, am lazy, I just put in the 10's and left it like that.
It is in a box and I have not got around to opening it up to play with it.
Winter is coming...
I wish I added some 10uF poly caps from my last order with Digikey. What do you guys think of this Panasonic unit for $3.49 ea - is that about the going price for these things?
Panasonic part ECQ-E1106KF - metalized polyester, 10 uF 100V, 10%.
Panasonic part ECQ-E1106KF - metalized polyester, 10 uF 100V, 10%.
An externally hosted image should be here but it was not working when we last tested it.
Worth a try. I used the SMDs that Dug linked to earlier from Digikey and they sound very good.
The two large 1u capacitors at the input appear not to be really necessary as they are followed by a 3u3 capacitor in series for the sub amplifier part of the circuit. If anything this 3u3 should be replaced with a film capacitor. Also the sub. TPA3116D2 has a series connected pair of 1u capacitors on each input.....why?
It looks as if four capacitors could be removed from this part of the circuit with a possible improvement in performance.
Would someone care to verify my findings before getting the soldering iron out.
It looks as if four capacitors could be removed from this part of the circuit with a possible improvement in performance.
Would someone care to verify my findings before getting the soldering iron out.
Ummm, are you referring to your 2.1 amp?
Keep in mind this is a single supply amp and has DC on the input pins.
Last edited:
Yes, I realise that, but there is a 3u3 capacitor after the two 1u capacitors and before the op amp. The two in series are before the sub amplifier so removing the 2 large 1u capacitors still leaves the op amp and the sub amplifier DC blocked. The left and right channel amp inputs are also DC blocked with 1u capacitors.
I think we might need a wikispace just like lepai's: Amplifier:Lepai T Amp - diyAudio
Listing all the mods, results, etc.
Listing all the mods, results, etc.
Sorry for the confusion, yes I am referring to the 2.1 amplifier that I mentioned the issue with the sync in my previous post.
I have not seen the 2 channel board so I don't know what the circuit is like around the pre-amp part. If it is the same as the Lepai one then there are no "spare" capacitors.
I am not convinced that the 22µH inductors in the output are wrong. Their value depends to a large extent on the speaker impedance and 22µH is about right for 8Ω. 10µ is the right value for 4Ω speakers. According to some the distortion is slightly lower with a larger value so I think it is worth trying different values to suit the speakers that are connected.
The sub channel may benefit from a reduced value but it doesn't matter that the cut-off is lower as it is only handling low frequencies.
It is interesting to note that the data sheet for TPA3116 shows only 10uH for 4-8 ohms and even in full BTL for 2 ohms also shows 10uH.
But the TI eval module uses a 22uH and notes a triple footprint and lists 10uH and 22uH options and in the general eval module specs suggests a load of 3.2-8 ohms.
But the TI eval module uses a 22uH and notes a triple footprint and lists 10uH and 22uH options and in the general eval module specs suggests a load of 3.2-8 ohms.
It is interesting to note that the data sheet for TPA3116 shows only 10uH for 4-8 ohms and even in full BTL for 2 ohms also shows 10uH.
But the TI eval module uses a 22uH and notes a triple footprint and lists 10uH and 22uH options and in the general eval module specs suggests a load of 3.2-8 ohms.
DUG,
Have a read of this:-
http://www.ti.com/lit/an/sloa119a/sloa119a.pdf
It is an interesting read about Class D filters.
It is an interesting read...will have to read it completely later.
From what I glanced at it should be read by anyone using Class D amps.
Whether I or anyone else gets anything out of the math is another thing.
There is still stuff in there that is good.
Thanks for posting the link.
suggest to try a Stacked Film Capacitor,Panasoinc ECQV(L)(M). We use these in Pioneer SX-x50 restorations. Next step up is polypropylene (PP) and cost. But really for a class "D"?
667-ECQ-V1H225JL
667-ECQ-V1H225JL
For a TPA3100D2 amp that I designed, I am using a 1u/1210/50V/X7R and sounds just fine to me. I have gain set to min, so it is eff. the same as a 10u, if the gain is max. So, I know, a stacked film poly cap should be better, but I can't hear it. I know that the 1u/1210/50V/X7R cap distortion, is way less than the amp, so does it make any diff? not to me or my ears.
I can really tell the diff in amp performance based on the speakers. If I use 4 ohm, this amp works way better than 8 ohm. I selected the inductors based on a 4ohm load and used ferrite beads as well.
I hardly ever use my Pioneer any more, that and I have a real tuner, not an ancient gang vcap tuner.
Cheers
Rick
I can really tell the diff in amp performance based on the speakers. If I use 4 ohm, this amp works way better than 8 ohm. I selected the inductors based on a 4ohm load and used ferrite beads as well.
I hardly ever use my Pioneer any more, that and I have a real tuner, not an ancient gang vcap tuner.
Cheers
Rick
Copied from ST application note
"Snubber network: the key function of a snubber network is to absorb energy from the inductive component in the power circuit (the output coils and the speaker). The purpose of the snubber RC network is to dissipate the unnecessary high pulse energy, such as a high voltage spike, in the power circuit which is dangerous to the system."
So yes it is necessary if you want your amplifier protected against the spikes due the fast switching edges causing overshoot.