OK, I put the curves for a better understanding ->
ORIGINAL
UPGRADED
Looking at the distortion at very low power, one can conclude that it is just the use of a different power supply AND NOTHING ELSE.
(A MEANWELL SE-450-48 will give comparable results)
I really doubt that this difference can be heard... (0.05 @ 10mW and 0.0002 @ 10W) -> are you a bat-man ?
Otherwise, i am very skeptical about the reality of the upgraded curve, however, with its scale which is not respected as I pointed out above.
I doubt that the investment for this upgrade (of which we still don't know anything) is 'profitable' for the price involved...
I would be very curious to see these curves with the original PCB (without upgrade) just with the OPAMP replaced as I indicated previously:
I am sure that the result would be at least as good as with the result obtained by this upgrade: pretend I didn't say anything
@ TNT: be more attentive because it is very clear that you do not know how to interpret what you are reading here, you are confusing several things and this distorts your conclusions
ORIGINAL
UPGRADED
Looking at the distortion at very low power, one can conclude that it is just the use of a different power supply AND NOTHING ELSE.
(A MEANWELL SE-450-48 will give comparable results)
I really doubt that this difference can be heard... (0.05 @ 10mW and 0.0002 @ 10W) -> are you a bat-man ?
Otherwise, i am very skeptical about the reality of the upgraded curve, however, with its scale which is not respected as I pointed out above.
I doubt that the investment for this upgrade (of which we still don't know anything) is 'profitable' for the price involved...
I would be very curious to see these curves with the original PCB (without upgrade) just with the OPAMP replaced as I indicated previously:
I am sure that the result would be at least as good as with the result obtained by this upgrade: pretend I didn't say anything
@ TNT: be more attentive because it is very clear that you do not know how to interpret what you are reading here, you are confusing several things and this distorts your conclusions
Last edited:
Hi @muducu@ TNT ->
where do you see this improvement ?
Unless I misunderstood you, the darkest curve on the graph indicates the results obtained with an 8 Ohm load, the other lighter curves are the results with a 4 Ohm load...
@ maciekw ->
No worries with the PCB from ALIEXPRESS, it's almost perfect, I received one a week ago (not tested yet as I'm waiting for my new power supply
-> I will present 'here' my amplifier equipped with this PCB as soon as it is fully finished, I think towards the end of the month
In conclusion, the only regret is the use of OPA1612 OPAMPS which are also soldered on the PCB without the possibility of adding others... 👈
Indeed, although this OPAMP has interesting characteristics in terms of THD (0.000015%), I find for my part that it is far from being the most musical (it cannot be measured, we have not yet invented device for this).
My preference is for the OPAMP OPA 828 which uses FET technology, the OPA 1612 being a BIPOLAR.
On this subject 3eAudio does not communicate, they just point out that replacing this OPAMP would 'disrupt' the rendering in terms of sound quality...
I'm not sure because I don't have the input circuit diagram (yet) but just considering the fact that the PCB input impedance is 10K (unbalanced) and 20K (balanced), the OPAMP used is not at all the most suitable since its THD increases sharply from about 3K whereas an OPAMP in FET technology (like the OPA 828) from this value is no longer dependent on that -this is its THD will be at the maximum that of the resistance at the input of the circuit.
Here is an illustration chart from TEXAS INSTRUMENTS for better understanding ->
View attachment 1165375
We can see very clearly on this example that in this case, the THD of the bipolar OPAMP is much more important than with the OPAMP FET at 10K and even more at 20K: it's not me who says it is the manufacturer (TEXAS INSTRUMENTS) of these OPAMPS 👈
This last AOP (828) has also been studied to show only a maximum of even harmonics while reducing the most the odd harmonics which are all below the level of the even harmonics, which generally makes listening much more pleasant for listeners
The choice of the OPAMP OPA 1612 is not the best either since the OPA 2211 (available in double mounting on DIP8 adapter because it's a single SOIC8 OPAMP) is more efficient in terms of bandwidth, open loop gain and offset voltage at its inputs but it would have been necessary to be able to replace the OPAMPS, which is not the case on this PCB since it does not offer the possibility of OPA ROLLING since the PCB is not equipped with a DIP8 'tulip' support
Otherwise and for simplicity, I invite 3eAudio to carry out tests with double OPAMPS in FET technology in SOIC8 format so as not to modify the PCB, namely:
- OPA 1642: 'neutral' tone
-OPA 1656: 'dynamic' tone but a slight 'rising' tendency in the treble.
Measurements in terms of PCB noise with these OPAMPS could confirm or invalidate my comments if 3eAudio is 'ready' for this exercise which could be very interesting
Cordially.
we test OPA1656,OPA1602,OPA1678 during development, most of a common op-amp can be used, the reason we select OPA1612 is for getting the lowest noise (in term of performance).
Hi 3eAudio
A little reminder on the data from the manufacturers' sheets:
the THD is indicated for a signal at 1kHz with an RL of 2K for the following two AOPs ->
OPA1612 -> THD = 0,000015% with eN = 1.1 nV/√Hz and iN = 1700 fa/√Hz
OPA1656 -> THD = 0,000029 % with eN = 4.3 nV/√Hz and iN = 1.2 fa/√Hz
Now we can start
As you have understood, this RL is much larger on your PCB so, by extension, the THD which will be measured with the OPA1612 (bipolar) will be much larger whereas with the OPA1656 (FET) it will not change not !
Here is a small excerpt from the net for explanation ->
"...Bipolar AOPs have a large current-to-noise density, which the source resistance converts to density voltage-noise.
FET AOPs have a major advantage in this respect over bipolar amplifiers because their components have a extremely low current noise density (in fA).
This is in contrast to bipolar AOPs because it is 1000 times greater than that of FET AOPs (therefore in pA) !
To compare the noise contributors on the two AOPs, voltage and current noise at 1 kHz can be used as a reference.
In other words, the voltage-to-noise density is calculated based on the resistance of the source with densities of voltage and current noise taken at 1 kHz.
Ignoring the intrinsic voltage-to-noise density of a source with 16k (for example) resistor has little impact on the total noise density because the amplifier starts to be affected only by current-to-noise density and thermal noise source resistance density.
Beyond 16 k, the total voltage-to-noise density begins to converge with i R N S ×.
Bipolar AOP provides the lowest noise of the two low source impedance amplifiers, starting at around 660R and below.
The CMOS amplifier exhibits the least high-resistance noise at the source.
The THD is when the thermal noise of the source resistor is 4.3 nV/√Hz, which is 2k for the FET and only 1.1 nV/√Hz for the bioplar...
Beyond this point (~@3K), the output noise begins to converge with the thermal noise of the source resistance because the CMOS (FET) amplifier has negligible current noise.
The THD+N in bipolar amplifiers gets worse with increasing source resistance, while the CMOS (FET) amplifier has negligible current noise to increase the total distortion.
In a low noise, low distortion application with a large source resistance, a CMOS amplifier would be a better choice..."
I submit my 'painting' (TEXAS INSTRUMENTS source) to illustrate ->
I therefore maintain that it is not necessary to stick only to the manufacturer's data, even less not to interpret them completely and that it is necessary to measure to control the final THD level with bipolar OPAMPS and FET OPAMPS to wrap up what needs to be...
Without saying that the technical performances 👁️🗨️are not important, I think that the musical performances 🔊 are to be taken with more consideration since it is the purpose of this PCB normally.
Regards 👀
A little reminder on the data from the manufacturers' sheets:
the THD is indicated for a signal at 1kHz with an RL of 2K for the following two AOPs ->
OPA1612 -> THD = 0,000015% with eN = 1.1 nV/√Hz and iN = 1700 fa/√Hz
OPA1656 -> THD = 0,000029 % with eN = 4.3 nV/√Hz and iN = 1.2 fa/√Hz
Now we can start
As you have understood, this RL is much larger on your PCB so, by extension, the THD which will be measured with the OPA1612 (bipolar) will be much larger whereas with the OPA1656 (FET) it will not change not !
Here is a small excerpt from the net for explanation ->
"...Bipolar AOPs have a large current-to-noise density, which the source resistance converts to density voltage-noise.
FET AOPs have a major advantage in this respect over bipolar amplifiers because their components have a extremely low current noise density (in fA).
This is in contrast to bipolar AOPs because it is 1000 times greater than that of FET AOPs (therefore in pA) !
To compare the noise contributors on the two AOPs, voltage and current noise at 1 kHz can be used as a reference.
In other words, the voltage-to-noise density is calculated based on the resistance of the source with densities of voltage and current noise taken at 1 kHz.
Ignoring the intrinsic voltage-to-noise density of a source with 16k (for example) resistor has little impact on the total noise density because the amplifier starts to be affected only by current-to-noise density and thermal noise source resistance density.
Beyond 16 k, the total voltage-to-noise density begins to converge with i R N S ×.
Bipolar AOP provides the lowest noise of the two low source impedance amplifiers, starting at around 660R and below.
The CMOS amplifier exhibits the least high-resistance noise at the source.
The THD is when the thermal noise of the source resistor is 4.3 nV/√Hz, which is 2k for the FET and only 1.1 nV/√Hz for the bioplar...
Beyond this point (~@3K), the output noise begins to converge with the thermal noise of the source resistance because the CMOS (FET) amplifier has negligible current noise.
The THD+N in bipolar amplifiers gets worse with increasing source resistance, while the CMOS (FET) amplifier has negligible current noise to increase the total distortion.
In a low noise, low distortion application with a large source resistance, a CMOS amplifier would be a better choice..."
I submit my 'painting' (TEXAS INSTRUMENTS source) to illustrate ->
I therefore maintain that it is not necessary to stick only to the manufacturer's data, even less not to interpret them completely and that it is necessary to measure to control the final THD level with bipolar OPAMPS and FET OPAMPS to wrap up what needs to be...
Without saying that the technical performances 👁️🗨️are not important, I think that the musical performances 🔊 are to be taken with more consideration since it is the purpose of this PCB normally.
Regards 👀
Good morning
I don't know where you got this 'tinkered' graphic from...
Moreover from the start, it is not indicated anywhere which PCB it is, nor the power supply used, etc...
You say anything by writing if it's N or THD: it means nothing and proves once again that you have decidedly understood nothing.
It's about THD+N: document yourself before writing down what goes through your head, please
https://www.ap.com/blog/thd-and-thdn-similar-but-not-the-same/
The FFT will allow you to visualize the harmonics especially.
To find out what it can be used for, look here ->
https://www.nti-audio.com/en/support/know-how/fast-fourier-transform-fft
Have a good day
I don't know where you got this 'tinkered' graphic from...
Moreover from the start, it is not indicated anywhere which PCB it is, nor the power supply used, etc...
You say anything by writing if it's N or THD: it means nothing and proves once again that you have decidedly understood nothing.
It's about THD+N: document yourself before writing down what goes through your head, please
https://www.ap.com/blog/thd-and-thdn-similar-but-not-the-same/
The FFT will allow you to visualize the harmonics especially.
To find out what it can be used for, look here ->
https://www.nti-audio.com/en/support/know-how/fast-fourier-transform-fft
Have a good day
Based on some advice from fellow DIYers, I've changed some part of my amplifier.
1. changed the transformer to DRV134PA based module (single to balance)
2. shortened the signal cables from DRV134PA to the TPA3255 board
3. added the dc noise filter (PO8936ZC) suggested by a user in the PO89ZB forum
4. changed the speaker cables to canare 2s7fg
At least no complaints found so far, period...
Thanks a lot to you DIYers, and the developer of this great board, 3e audio!
** Better DIYing than BUYing **
1. changed the transformer to DRV134PA based module (single to balance)
2. shortened the signal cables from DRV134PA to the TPA3255 board
3. added the dc noise filter (PO8936ZC) suggested by a user in the PO89ZB forum
4. changed the speaker cables to canare 2s7fg
At least no complaints found so far, period...
Thanks a lot to you DIYers, and the developer of this great board, 3e audio!
** Better DIYing than BUYing **
Last edited:
Good morning
It's going in the right direction BUT:
you have plenty of room to put a balanced volume potentiometer and remove all the 'horrors' on your input signal...
... you just risk gaining a LOT in SQ
If your source really does not have balanced outputs then use a cable that allows RCA to XLR: you will certainly not have 4 volts at the input but 2 volts, however at least you will have removed a whole 'heap' of circuits that are sources of failure, of 'noise' and so on and ESPECIALLY the transformer which feeds all this 'bazaar'
You will thank me later
Regards.
PS: given the layout of your circuit, it would be in your best interest to wire the output of your DC power supply DIRECTLY to your amplifier PCB.
This would have the effect of shortening the link as much as possible, which is much more beneficial and would 'capture or broadcast' less noise or interference once again
To further improve this, you can twist the wires very tight and make a 'shield' over them as I have shown in previous posts.
I say that but I said nothing...
It's going in the right direction BUT:
you have plenty of room to put a balanced volume potentiometer and remove all the 'horrors' on your input signal...
... you just risk gaining a LOT in SQ
If your source really does not have balanced outputs then use a cable that allows RCA to XLR: you will certainly not have 4 volts at the input but 2 volts, however at least you will have removed a whole 'heap' of circuits that are sources of failure, of 'noise' and so on and ESPECIALLY the transformer which feeds all this 'bazaar'
You will thank me later
Regards.
PS: given the layout of your circuit, it would be in your best interest to wire the output of your DC power supply DIRECTLY to your amplifier PCB.
This would have the effect of shortening the link as much as possible, which is much more beneficial and would 'capture or broadcast' less noise or interference once again
To further improve this, you can twist the wires very tight and make a 'shield' over them as I have shown in previous posts.
I say that but I said nothing...
you know which released first as a baseline.... maybe they were 'inspired' by the 'topology' of the last TOPPING PA7 but they don't want to say it
most likely they are based on ti's app note(there some other designer like XKR,JLE,dormoor... in diy forum) i believed, depend on how you customized into your design.
i just see the result post on ASR, and i guess the plus version is 2x TPA3255 in PBTL mode and not-plus version is 1x TPA3255 in BTL mode.
but the price ($549) is quite surprise to me....
https://www.audiosciencereview.com/forum/index.php?threads/topping-pa7-plus-amplifier-review.43932/
People hi 
Hi 3eaudio
Thin !!!
I hope the same thing will not happen with mine ->
I haven't tried it yet because I haven't finished connecting the two balanced plugs
And then there's a bunch of connections inside that I don't even know how to plug in
I'll tell you that later... 👈
Kisses 💋
Hi 3eaudio
Thin !!!
I hope the same thing will not happen with mine ->
I haven't tried it yet because I haven't finished connecting the two balanced plugs
And then there's a bunch of connections inside that I don't even know how to plug in
I'll tell you that later... 👈
Kisses 💋
Hello everyone
@ 3eaudio,
I'm glad you like it
1> Regarding the chassis, they come from ALIEXPRESS ->
https://fr.aliexpress.com/item/1005...st_main.8.21ef5e5bXl2qt2&gatewayAdapt=glo2fra
There are several models and finishes that will suit most of us as you will see.
This seller is super nice, very professional, takes time to fully understand the expectations of his customers and his prices are the best I have found on this sales site (yet I spend a lot of time looking for the slightest saving).
Delivery times are a little longer than usual but the advantage is that they can do the frame drilling as you wish by providing them with a plan
For my copy, I only asked them for the front hole (diameter 22mm) because it is 8mm thick, which is quite difficult to achieve if you do not have the appropriate tools and knowledge to do it properly 👈
I hadn't asked them for the holes in the back plate because I wasn't yet sure of the exact layout of my PCBs having not received them yet (including yours on which I replaced the 'thermal paste' by ARTIC MX6 also used to put it in contact with the chassis).
2> As for the serigraphs on the device, these are 'transfers' but are really very solid and even difficult to remove. I have several in advance (20) ->
You can ask the supplier for almost anything you want, including colors such as chrome with rainbow reflections, as I did for Lecter's SYLPH D200 MKII.
The advantage with this type of screen printing is to be able to place them as desired and that if the effect obtained is not pleasing, there are others to redo them differently as desired again.
I had ordered 21 'stickers', I only used one because I knew exactly what I wanted from the start but it's a matter of taste.
3> For this new amplifier that I made, I wanted to obtain a device with the 'minimum' required in terms of its presentation but also of these circuits while being as efficient as possible: often simplicity offers the best results , that's often what I've seen in my little DIY and upgrade experience.
I'm going to put the extra effort into getting the wiring done thanks to my first experience on the SYLPH D200 MKII which was already more than acceptable in my opinion and as some seem to have indicated (including Lester).
You will have to be a little patient for me to show you the 'innards' of this amp with photos and some explanations, reasons for my choices, etc...
If my courage is there but especially if my schedule allows it, I could perhaps post them by this evening (it's 8:30 in the morning here) but I prefer not to make promises that I wouldn't keep then a bit of patience...
See you soon
@ 3eaudio,
I'm glad you like it
1> Regarding the chassis, they come from ALIEXPRESS ->
https://fr.aliexpress.com/item/1005...st_main.8.21ef5e5bXl2qt2&gatewayAdapt=glo2fra
There are several models and finishes that will suit most of us as you will see.
This seller is super nice, very professional, takes time to fully understand the expectations of his customers and his prices are the best I have found on this sales site (yet I spend a lot of time looking for the slightest saving
).Delivery times are a little longer than usual but the advantage is that they can do the frame drilling as you wish by providing them with a plan
For my copy, I only asked them for the front hole (diameter 22mm) because it is 8mm thick, which is quite difficult to achieve if you do not have the appropriate tools and knowledge to do it properly 👈
I hadn't asked them for the holes in the back plate because I wasn't yet sure of the exact layout of my PCBs having not received them yet (including yours on which I replaced the 'thermal paste'
by ARTIC MX6 also used to put it in contact with the chassis).2> As for the serigraphs on the device, these are 'transfers' but are really very solid and even difficult to remove. I have several in advance (20) ->
You can ask the supplier for almost anything you want, including colors such as chrome with rainbow reflections, as I did for Lecter's SYLPH D200 MKII.
The advantage with this type of screen printing is to be able to place them as desired and that if the effect obtained is not pleasing, there are others to redo them differently as desired again.
I had ordered 21 'stickers', I only used one because I knew exactly what I wanted from the start but it's a matter of taste.
3> For this new amplifier that I made, I wanted to obtain a device with the 'minimum' required in terms of its presentation but also of these circuits while being as efficient as possible: often simplicity offers the best results , that's often what I've seen in my little DIY and upgrade experience.
I'm going to put the extra effort into getting the wiring done thanks to my first experience on the SYLPH D200 MKII which was already more than acceptable in my opinion and as some seem to have indicated (including Lester).
You will have to be a little patient for me to show you the 'innards' of this amp with photos and some explanations, reasons for my choices, etc...
If my courage is there but especially if my schedule allows it, I could perhaps post them by this evening (it's 8:30 in the morning here) but I prefer not to make promises that I wouldn't keep then a bit of patience...
See you soon
