3e Audio A7 [TPA3255 PFFB Stereo] Finished Amplifier - User Impressions | Owner's Thread

[JeremyFife]

I won't argue about this fantastical phenomenon; we know that the TPA32xx input contains capacitors through which DC current cannot flow.

In any case, the question wasn't whether the respected audio engineers present on this forum consider the presence of electrolytic capacitors in the audio path beneficial or detrimental to sound, but rather how to implement a bipolar power supply for the op amps in this amplifier to eliminate these capacitors.
And that's the second question.

But the first question is: will it be possible to HEAR the difference between the A5 and A5ce amplifiers on 8-ohm speakers, assuming the same op amps are used in the preamp stage?

To your last question: speaker impedance varies with frequency, it's highly unlikely that yours are a constant 8 Ohm. Without knowing their impedance profile its hard to give an answer ... but it highly unlikely that there will be any difference between amps.
Caveat 1: if your speakers are an unusually difficult load, there might be a difference.
Caveat 2: if you change anything (literally anything) then you could easily hear a difference just because of the cognitive biases that we are all subject to. That's why there is such a focus on measurements and proper listening controls here.

Short version: very (very) unlikely to be a difference.

 

[r-36m2]

To your last question: speaker impedance varies with frequency, it's highly unlikely that yours are a constant 8 Ohm. Without knowing their impedance profile its hard to give an answer ... but it highly unlikely that there will be any difference between amps.
Caveat 1: if your speakers are an unusually difficult load, there might be a difference.
Caveat 2: if you change anything (literally anything) then you could easily hear a difference just because of the cognitive biases that we are all subject to. That's why there is such a focus on measurements and proper listening controls here.

Short version: very (very) unlikely to be a difference.

Thanks for the detailed answer!
The speaker system this amplifier will be paired with is based on simple 8-ohm Visaton BG8 and SC10 speakers with a simple crossover. This design is old, and many people have used and tested it successfully. In my opinion, this is a fairly simple load for the amplifier.

 

[VientoB]

Bit of a gimmick allowing the opamps to be changed but I guess it helps them sell more.

 

[r-36m2]

Bit of a gimmick allowing the opamps to be changed but I guess it helps them sell more.

I replaced the standard NE5532 op amps in their previous amplifier with the same old and cheap MC33272, and the sound improved.
We all (hopefully) understand that operational amplifiers shouldn't be changed indiscriminately, without regard to the circuit and without taking their wiring into account. Indeed, there are microcircuits that transmit sound better, all other things being equal, but they will only be effective if they are in optimal conditions for proper operation.
I don't know how much of a gimmick this is with the op amps, but I'm most interested in the question: how much better does the A5 sound than the A5SE, and what's the point of the PBTL, other than the low-impedance load?

 

[TurtlePaul]

In any case, the question wasn't whether the respected audio engineers present on this forum consider the presence of electrolytic capacitors in the audio path beneficial or detrimental to sound, but rather how to implement a bipolar power supply for the op amps in this amplifier to eliminate these capacitors.

There are a total of eight ground pins on the TPA325x. Are some of them power supply grounds and some grounds tied to output of the gate driver? TI doesn't share and assumes that you will tie them all together to the ground plane if your board follows the example implementation.

Similarly, a lot of these manufacturers have had to wrestle with the chip to implement post-filter feedback. TI provides a general schematic for a post-filter feedback implementation, but it only picks up 5-6 dB vs. the reference implementation. The chip contains an internal feedback loop and TI does not share the details of this loop, so it is non-trivial to calculate the stability of the amp in this configuration.

There are some mysteries of the TPA325x that Texas Instruments doesn't share. The TPA3255 even has voltage regulators within the chip, which is does not expose to any external pins. To some degree the designers of these amps tinker with the chip to find performance gains but it is non-trivial. To some degree, you need to accept that the TPA325x is a high performance black box.

I think that it is in TI's interest to just show the best implementations but not show how their self-oscillating high performance architecture works internally.

 

[r-36m2]

There are a total of eight ground pins on the TPA325x. Are some of them power supply grounds and some grounds tied to output of the gate driver? TI doesn't share and assumes that you will tie them all together to the ground plane if your board follows the example implementation.

Similarly, a lot of these manufacturers have had to wrestle with the chip to implement post-filter feedback. TI provides a general schematic for a post-filter feedback implementation, but it only picks up 5-6 dB vs. the reference implementation. The chip contains an internal feedback loop and TI does not share the details of this loop, so it is non-trivial to calculate the stability of the amp in this configuration.

There are some mysteries of the TPA325x that Texas Instruments doesn't share. The TPA3255 even has voltage regulators within the chip, which is does not expose to any external pins. To some degree the designers of these amps tinker with the chip to find performance gains but it is non-trivial. To some degree, you need to accept that the TPA325x is a high performance black box.

I think that it is in TI's interest to just show the best implementations but not show how their self-oscillating high performance architecture works internally.

Colleague, thank you, but my question wasn't about the secrets of the TPA32XX. I was asking about the bipolar power supply for the preamplifier stages, and also about how audible the difference between BTL and PBTL operation can be with an 8-ohm load, because the price difference between the A5 and A5se (F7 and A7se) is significant.

 

[VientoB]

I replaced the standard NE5532 op amps in their previous amplifier with the same old and cheap MC33272, and the sound improved.
We all (hopefully) understand that operational amplifiers shouldn't be changed indiscriminately, without regard to the circuit and without taking their wiring into account. Indeed, there are microcircuits that transmit sound better, all other things being equal, but they will only be effective if they are in optimal conditions for proper operation.
I don't know how much of a gimmick this is with the op amps, but I'm most interested in the question: how much better does the A5 sound than the A5SE, and what's the point of the PBTL, other than the low-impedance load?

Sure thing.

 

[oldloder]

... but I'm most interested in the question: how much better does the A5 sound than the A5SE ...

I wonder how much better the A5 will sound than the A5SE ... to you. I hope you'll get both and tell us your "User Impressions" ... complete with what you determine to be the cause for the difference.

 

[r-36m2]

Sure thing.

Sorry, what "Sure thing"?

I wonder how much better the A5 will sound than the A5SE ... to you. I hope you'll get both and tell us your "User Impressions" ... complete with what you determine to be the cause for the difference.

To be honest, I was hoping that someone had already done this))))

 

[melomane13]

I have made abx test with A5 and A5se on speaker impedency of 4 ohm minimum. No différences

 

[melomane13]

I have sold A5, buy aiyima a70, make abx test with A5se and a70. No differences. I have sold a5se...

 

[melomane13]

Sorry, what "Sure thing"?

"I replaced the standard NE5532 op amps in their previous amplifier with the same old and cheap MC33272, and the sound improved."

Sure than "Sure thing" is refers to the part in bold of your syllogism. It could be either a confirmation of what you're saying or irony. I don't know.

 

[Wolf11Man]

Îrony, sure thing.

 

[r-36m2]

"I replaced the standard NE5532 op amps in their previous amplifier with the same old and cheap MC33272, and the sound improved."

Sure than "Sure thing" is refers to the part in bold of your syllogism. It could be either a confirmation of what you're saying or irony. I don't know.

Thanks for the reply! If there's no difference between the 5 and the A5ce, I'll probably go with the A5ce.

Regarding the improved sound due to the op amp change—that's not ironic. I've tried different op amps. I have some very old Philips NE5532. And the sound with them was noticeably better than with the stock NE5532. Plus, for the sake of experimentation, I bought new 5532s at a store. The sound with them was simply awful, so I returned them to the store. Perhaps they were counterfeits, I don't know.
I didn't measure the THD and IMD, but the sound changes were clearly audible. Again, I'm not one to listen to wires and capacitors, but the sound changes were very noticeable.

Ultimately, I settled on the MC33272 op amp. I like the amplifier's sound best with them; it's a little smoother, more musical, and at the same time, not muddy.
Generally speaking, the difference between op amps in the preamp is sometimes audible, but not very noticeable. I heard the biggest difference when replacing the op amp in the DAC's low-pass filter. But that's another story.

 

[r-36m2]

I have sold A5, buy aiyima a70, make abx test with A5se and a70. No differences. I have sold a5se...

This is very interesting. Are you saying the A5se doesn't sound any better than the Aiyima A70?
I wonder what the Aiyima A80 sounds like?

 

[VientoB]

Guys there are threads on opamp changes showing they do very little indeed.

Does Op-amp Rolling Work?

When I post my last video on review of Douk A5 amplifier, I mentioned that rolling (changing) op-amp ICs in there is fruitless. I got have a dozen comments under that video in youtube asking why so I thought I experiment again with the A5. Note that I have done the same testing with both DACs...

www.audiosciencereview.com

Sparkos SS3602 Opamp Rolling In Fosi P4

This is a detailed measurements of effect of swapping the stock NE5532P op-amp in Fosi P4 preamplifier with Sparkos SS3602 "discrete" opamp. The SS3602 costs US $79.80 and the NE5532P, US $0.47. Swapping opamps in P4 is more challenging due to back connectors/switches that need to be...

www.audiosciencereview.com

 

[melomane13]

This is very interesting. Are you saying the A5se doesn't sound any better than the Aiyima A70?
I wonder what the Aiyima A80 sounds like?

No, what I'm saying is that,
under the following conditions:
1) ABX comparison
2) with a load that doesn't go below 4 ohms and with a phase shift of a maximum of ±60 degrees,
all the amplifiers in question—A5, A5SE, A70—are indistinguishable.

Since I don't know the load you're using, I can't tell you more about the A80. Show me the impedance and phase curve of your speaker, like this for example:

 

[melomane13]

Tthat's not ironic. I've tried different op amps.

My native language isn't English, but Italian, and I'm bilingual in French.
I understand English, especially written English.
To be sure of what I'm conveying, I double-check my comprehension by switching between the three languages, making sure I have the same meaning in all three...
I'm certain you're missing some nuances, like the term "irony," in its original context.
What is your native language?
I'd like to try to explain what many here have been trying to do in numerous posts.

 

[r-36m2]

What is your native language?
I'd like to try to explain what many here have been trying to do in numerous posts.

Russian.

Yes, I understand there are some misunderstandings here due to language differences and translation inaccuracies. Let's leave it at that, friend. This isn't a linguistic forum. Let's talk about amplifiers instead.

Unfortunately, I can't provide you with the phase response of my speaker systems, as they are homemade, using 8-ohm Visaton BG8 and SC10 speakers in 60-liter closed boxes.

I thought the A5/A7(CE) amplifiers were among the highest-quality implementations of the TPA325X, as well as some of the best amplifiers in their price category. If their sound quality is comparable to amplifiers with more extensive features, such as the Aima A80, O-NOORUS D3 (PRO), Douk Audio Ampapa D1, LOXJIE A40 (based on a different chip), and others, then, given their similar price range, it makes sense to choose devices with a more extensive feature set.

I'm familiar with the work of a competent engineer who has also very successfully implemented the capabilities of the TPA3255, but he doesn't sell assembled amplifiers, only printed circuit boards, and I no longer want to spend evenings with a soldering iron.

 

[melomane13]

You can either measure the actual impedance with REW and a sound card, or simulate it with Visaton Boxsim (a free program you can download).
Without this data, it's just hot air.
Without this data, buy A5 , you will be certain that it will be able to drive any load, within its power envelope. if you want more power , buy A7