Can't say exactly, something happened in the 80W region before the supply foldback.
Most likely yes. I only plotted up to the first yellow line.
Pretty much nope, not even at 1W. Beside this, noone really measured "all other Ebay/Aliexpress boards".
But if it sounds good to you, go for it.
Attachments
Last edited:
Fine. Whatever the reason is the boards are apparently quite good. The overload of posts and information makes it difficult to discern. This thread has seen all corners of the spectrum 🙂
Questions:
1. Is the original Sanwu TPA3118 board better than or equal to the unbranded ones ?
2. Is the suspicion of fake chips proven ?
Questions:
1. Is the original Sanwu TPA3118 board better than or equal to the unbranded ones ?
2. Is the suspicion of fake chips proven ?
Last edited:
You'll have to conclude for yourself.
As noted a couple of times, this is an genuine TI chip i soldered to the board. 🙂
Not tested.
Not tested (due to the fact i don't have the "fake" ic handy - it's in the lab)
Hmm, seen this?
Last edited:
It is very clear to you but not to me and I think I am not the only one as it is a doctormord/irribeo affair the last pages. I see you except anyone reading the thread to interpret the measurements and to reread all pages ?
Is there any difference between the original one and the suspected one ? Is the suspected one a different lower specced chip ?
Is there any difference between the original one and the suspected one ? Is the suspected one a different lower specced chip ?
Last edited:
OK so the suspected chip is not tested (at all) but an original one is soldered to an unbranded board that had a suspected chip. Not a comparison but simply replacing the possible rebranded chip for a known good one. What is the meaning of NOT comparing ? It i like questioning the witness but let the suspect slip away 🙂
I see a diagram that measurements are worse than factory measurements but irribeo says the boards measure better than EVM ...
So I can not conclude a thing except that I think I read that the coils are underrated. That might be the reason for the last diagram showing non data sheet results.
* This could still mean "quite good boards" but it could also show the need for better inductors.
I see a diagram that measurements are worse than factory measurements but irribeo says the boards measure better than EVM ...
So I can not conclude a thing except that I think I read that the coils are underrated. That might be the reason for the last diagram showing non data sheet results.
* This could still mean "quite good boards" but it could also show the need for better inductors.
Last edited:
Standard Sanwu board, musicsample, 15 or 16V (14A at 15cm distance) psu, load electrostatic panel parallel connected to subwoofer, didn't trip protection at 11V output, gmarsh/dugs both mono too trip protection around 5V output into that load. I was 1 meter away, I didn't want to hurt myself or possibly speakers by getting over 11V with standard Sanwu. With better inductors I only very briefly checked if 11V still was no problem, sound was a lot nicer, obviously, but still uncomfortably loud, resistor is helpfull with that test 🙂
Sanwu inductors are poor.
Sanwu inductors are poor.
Correct, i send the suspected chip to the lab for a decap. I don't understand your last question, i did not wanted to do a comparison, but wanted to see how the board is performing.
(Assuming the "fake" is (like) "original" and/or from same silicon, they should perform exactly the same.)
You decide. From my point of view, and the diagram shows it, the boards performs not close to EVM. I have no problems with that, if the measured performance fit the needs.
Normally the majority of members here at DIYA gives a sh*t on measurements if the sound is good. So i'm showing only the diagrams and you'll have to judge for yourself. The primary question is "will i hear it?".
- Will i hear the difference between 0.01% and 0.03% with speakers not able to resolve this low THD? -> No
- Will i hear the difference between 0.1% and 0.9%? Possibly yes. With poor speakers, most likely not.
- Costs for a decrease of 0.03% to 0.01%, i.e. $1, worth it? Most likely not.
- Costs for a decrease of 0.9% to 0.1%, i.e. $1, worth it? Most likely yes.
- Costs for a decrease of 0.9% to 0.1%, i.e. $10, worth it? Most likely not.
Conclusion? 🙂
Correct, that's a strong indictation.
Correct, this would nee further investigation and/or measurements.
Last edited:
Staccato style communication (@ irrribeo) but I understand it.
* Preliminary conclusion
The original Sanwu TPA3118 mono boards :
- are quite good PCBs
- need gain reduction to 20 dB for best results.
- don't trip "early" like any other TPA3118 board (even well designed known good ones)
- sound very good but sound improves considerably with better inductors (which ones ? JP)
- sound with different inductors improves but measurements should prove this
- need remeasuring when supposedly better inductors are used for final conclusion
To do:
- make list of possible inductor replacements
- remeasuring with replacement inductors
- write final conclusion
* Preliminary conclusion
The original Sanwu TPA3118 mono boards :
- are quite good PCBs
- need gain reduction to 20 dB for best results.
- don't trip "early" like any other TPA3118 board (even well designed known good ones)
- sound very good but sound improves considerably with better inductors (which ones ? JP)
- sound with different inductors improves but measurements should prove this
- need remeasuring when supposedly better inductors are used for final conclusion
To do:
- make list of possible inductor replacements
- remeasuring with replacement inductors
- write final conclusion
Last edited:
Additional Note to my previous reply.
Normally the majority of members here at DIYA gives a sh*t on measurements if the sound is good. So i'm showing only the diagrams and you'll have to judge for yourself. The primary question is "will i hear it?".
- Will i hear the difference between 0.01% and 0.03% with speakers not able to resolve this low THD? -> No
- Will i hear the difference between 0.1% and 0.9%? Possibly yes. With poor speakers, most likely not.
- Costs for a decrease of 0.03% to 0.01%, i.e. $1, worth it? Most likely not.
- Costs for a decrease of 0.9% to 0.1%, i.e. $1, worth it? Most likely yes.
- Costs for a decrease of 0.9% to 0.1%, i.e. $10, worth it? Most likely not.
Conclusion?
Normally the majority of members here at DIYA gives a sh*t on measurements if the sound is good. So i'm showing only the diagrams and you'll have to judge for yourself. The primary question is "will i hear it?".
- Will i hear the difference between 0.01% and 0.03% with speakers not able to resolve this low THD? -> No
- Will i hear the difference between 0.1% and 0.9%? Possibly yes. With poor speakers, most likely not.
- Costs for a decrease of 0.03% to 0.01%, i.e. $1, worth it? Most likely not.
- Costs for a decrease of 0.9% to 0.1%, i.e. $1, worth it? Most likely yes.
- Costs for a decrease of 0.9% to 0.1%, i.e. $10, worth it? Most likely not.
Conclusion?
The way you communicate is hard to follow, sorry. On one hand you want to be neutral and leave the reader to think for himself and on the other hand you give an opinion (sometimes between the lines).
With regards to costs to improve the boards: difficult to tell what is too expensive and what not but my personal opinion is that 2 x 7 $ makes 14 $ initial costs for a stereo setup which is peanuts compared to the rest of the parts needed to make it a complete amp. If the boards are really that good an investment of let's say 25 $ to improve them and squeeze the maximum out of them would be OK. I would even think a total of 50 $ is not a problem when assembling an amplifier that outperforms other amps I have with a margin at a lower cost.
One needs a case and PSU anyway for a serious amplifier (we don't stay in the "piece of wood with some boards" stadium do we ?). A minority will use these powerful boards mobile, most will use the boards either in stereo or multichannel setup. 50 $ for 2 channels maxed out for quality is a healthy investment it seems.
With regards to costs to improve the boards: difficult to tell what is too expensive and what not but my personal opinion is that 2 x 7 $ makes 14 $ initial costs for a stereo setup which is peanuts compared to the rest of the parts needed to make it a complete amp. If the boards are really that good an investment of let's say 25 $ to improve them and squeeze the maximum out of them would be OK. I would even think a total of 50 $ is not a problem when assembling an amplifier that outperforms other amps I have with a margin at a lower cost.
One needs a case and PSU anyway for a serious amplifier (we don't stay in the "piece of wood with some boards" stadium do we ?). A minority will use these powerful boards mobile, most will use the boards either in stereo or multichannel setup. 50 $ for 2 channels maxed out for quality is a healthy investment it seems.
Last edited:
I just tried to answer your questions. 🙂
Having an own opinion does make me "unneutral". 🙂
My opinion isn't meant to bias someone.
(Would i spend another $10 bucks in this board? Maybe once, for measurements.)
Language barrier seem to have influence in this conversation to. 😉
Having an own opinion does make me "unneutral". 🙂
My opinion isn't meant to bias someone.
(Would i spend another $10 bucks in this board? Maybe once, for measurements.)
Language barrier seem to have influence in this conversation to. 😉
Let's say that we all want these boards to perform the absolute best possible way (that is why many are in this hobby). We found very cheap boards with a lot of potential. Their price is a fraction of what we would spend for quality when buying a ready made amplifier or even when building a comparable DIY amplifier. We have spent a small ,even laughable, amount of money and want to max out results. What would we do to make a stereo or multichannel setup the best possible way for still acceptable costs ?
Board costs are 6.99 $. We count costs for just one board to make things simpler.
* Original Sanwu TPA3118 mono modification cost list:
1. What inductor would we need ?
2. What other parts are of unknown or mediocre quality and need replacing ?
3. Indicate all part costs
4. make total cost bill
5. leave board owner to decide whether investment is sound
* We could predefine a certain amount of maximum 25 $ for 1 board with all possible improvements that the board will allow.
I am not so sure about that. You seem a person that wants to back up acoustic results by measurements and that is a good way to understand matters. No layman can debate that this is a "wrong" method.
Board costs are 6.99 $. We count costs for just one board to make things simpler.
* Original Sanwu TPA3118 mono modification cost list:
1. What inductor would we need ?
2. What other parts are of unknown or mediocre quality and need replacing ?
3. Indicate all part costs
4. make total cost bill
5. leave board owner to decide whether investment is sound
* We could predefine a certain amount of maximum 25 $ for 1 board with all possible improvements that the board will allow.
I am not so sure about that. You seem a person that wants to back up acoustic results by measurements and that is a good way to understand matters. No layman can debate that this is a "wrong" method.
Last edited:
Again for any of the simple class D amps, look at outputfilter and load, your speakers aren't like 4 ohm resistors at 40kHz most of times. Replacing for best inductor to go with 4 ohm resistor isn't going to give best results often. Design postfilter feedback for 3118 and it will be easier 🙂
Show best inductor/filter for 4 Ohm, 6 Ohm and 8 ohm situation for original Sanwu mono TPA board. Define cost. Make drawing. Show pic of mod.
(I adapt to communication style).
BTW it would be handy if one would mention which inductor must be used. Please note that all this can be written on 1 A4 page/PDF. Saves a lot of reading, rereading, reading between lines and interpreting measurements. At least that is when one has limited time and needs to do other stuff in life. Would possibly kill the thread though 😀
Joined effort of members experiences and measurements combined is what makes this forum so powerful. If information is combined then it seems best to add it together and make it one information sheet for readers. If one of you has measured and drawn conclusion and feels obliged to post all this here I don't see a reason to repeat that. Conclusion is to use a better inductor ? OK suggest a better one and remeasure. I could use another one and measure. OK we choose best out of those 2. Saves each of us a lot of time. Reader reads PDF, picks best choice, solders that one to his board and is happy. Personally I don't see any other reason to fill hundreds of pages with incomplete or garbled info and going back and forth to other chips etc. No-one can keep up with information this way.
(I adapt to communication style).
BTW it would be handy if one would mention which inductor must be used. Please note that all this can be written on 1 A4 page/PDF. Saves a lot of reading, rereading, reading between lines and interpreting measurements. At least that is when one has limited time and needs to do other stuff in life. Would possibly kill the thread though 😀
Joined effort of members experiences and measurements combined is what makes this forum so powerful. If information is combined then it seems best to add it together and make it one information sheet for readers. If one of you has measured and drawn conclusion and feels obliged to post all this here I don't see a reason to repeat that. Conclusion is to use a better inductor ? OK suggest a better one and remeasure. I could use another one and measure. OK we choose best out of those 2. Saves each of us a lot of time. Reader reads PDF, picks best choice, solders that one to his board and is happy. Personally I don't see any other reason to fill hundreds of pages with incomplete or garbled info and going back and forth to other chips etc. No-one can keep up with information this way.
Last edited:
I am not the one filling pages full of information. My part is that I try to keep up as a normal reader and IMO it is hard to follow so I indicate that it could be done differently. I contribute in various threads and have done some practical matters too. In this thread I am a spectator trying to decipher information, possibly in the interest of other readers. Communication is hard to follow and questions are seldomly completely answered making it hard to extract real useful information. A reader that has these same boards but lacks experience might just want to follow instructions how to make his/hers boards like you both have pictured it and measured it. Those that have experience will join your discussions automatically.
As you can see I am trying to motivate both you and irribeo to join forces/information and make it a concrete way of modding these specific boards to best level for your readers. Maybe you haven't realised but you both are chit chatting on a international public forum with many readers 🙂
As you can see I am trying to motivate both you and irribeo to join forces/information and make it a concrete way of modding these specific boards to best level for your readers. Maybe you haven't realised but you both are chit chatting on a international public forum with many readers 🙂
Last edited:
Cheers. 🙂
I'll do another measurement set with "better/known" inductors, but will not invest much more time into it if things don't get better much.
There's actually enough other workload on my bench. 🙂
I'll do another measurement set with "better/known" inductors, but will not invest much more time into it if things don't get better much.
There's actually enough other workload on my bench. 🙂