Hey guys just wondering what your opinion is on either of them and if anyone ever did an ab comparison between the two. I find the 49600 to be very good in my composite design, almost reference level, but wonder if 6120 can be even better
I've just bought some LME49600 's . Care to show your composite design here ?
Since the wire : "The Wire" Ultra-High Performance Headphone Amplifier - PCB's
and :
O2 headamp output booster PCB
not many interesting designs.
Since the wire : "The Wire" Ultra-High Performance Headphone Amplifier - PCB's
and :
O2 headamp output booster PCB
not many interesting designs.
You will still need driver chips for the TPA6120.
Very specific requirements for solder mask thickness, power pad etc. See the last pages of the datasheet.
TI introduced a new version of the BUF634 a few years back -- BUF634A. Pretty impressive.
TPA6120A2 is just the best value by far. There are some other pretty good CFA line drivers, but they are at least 5 times the price in low quantities. Same thing goes for the pure (diamond) buffer ICs like the LME49600.
I just looked at the BUF634A datasheet once again and noticed that it actually contains distortion figures now (in contrast to the original BUF634 datasheet or the LME49600 one where they use it in a composite amplifier configuration for the distortion measurements). LMH6321 used to be the only one of the buffer ICs with actual open loop distortion measurements in the datasheet, so that is a welcome change.
I just looked at the BUF634A datasheet once again and noticed that it actually contains distortion figures now (in contrast to the original BUF634 datasheet or the LME49600 one where they use it in a composite amplifier configuration for the distortion measurements). LMH6321 used to be the only one of the buffer ICs with actual open loop distortion measurements in the datasheet, so that is a welcome change.
From a builder's standpoint tpa6120 is such an ugly driver with many restrictions imposed by its pinouts. What if i wanna go dual mono with seperate grounding? That means 2 chips now but one side disabled on each because the paralleling is out of the question. And when disabling do i just unconnect one side or ground their input pins?
It just seems like an unconventional and unfriendly chip to me. I wouldve chosen it without hesitation if it was packaged in a more traditional form.
And my target design being a dual mono amp, having to waste half a chip per channel is a big letdown
It just seems like an unconventional and unfriendly chip to me. I wouldve chosen it without hesitation if it was packaged in a more traditional form.
And my target design being a dual mono amp, having to waste half a chip per channel is a big letdown
Oh it was just a vanilla datasheet iteration. Except i used 5534 and for that once again datasheet compensation cap of 22pf between pin 5 and 8
Why is paralleling not an option for you? Or differential configuration?
THS6182 has slightly nicer to work with pinout in my opinion, but as I said 5 times more expensive.
THS6182 has slightly nicer to work with pinout in my opinion, but as I said 5 times more expensive.
You mean this one :
https://www.diyaudio.com/forums/hea...600-reference-design-project.html#post2523423
Reading the TPA6120 's datasheet ... a lot about instability and what is the 30 something ohm at its output ? That is a big no no.
Cooling is a bit better than de 1mm2 of a BUF634A , but nowhere near the LME49600 , that is if you value your component NOT running hot.
LME49600, LMH6321 & BUF634 are all variations of a monolithic diamond buffers with no NFB.
TPA6120 is not much more a unity-gain-stable (under specific conditions) opamp.
We are comparing apples with oranges.
Patrick
TPA6120 is not much more a unity-gain-stable (under specific conditions) opamp.
We are comparing apples with oranges.
Patrick
Its a valid comparison under my discussion because what im saying is i wanna make fruit salad ��
Hmm on a quick glance paralleling doesnt seem like it will be optimal wiring at all. The two outputs have to meet around the chip which means it will have to go diagnoally across the psu trace somewhere. Not to mention that will also involve the feedback resistor doing the same