I purchased a pair of OPA627 already made up on an adapter from China off eBay which were not expensive. Still to arrive. They are still listing them.
Also purchased some OPA1656 chips from Mouser which I will make up on adapters if anyone in Europe is interested ?
After reading this:
Fake Audiophile Opamps Revealed | Hackaday
Fake audiophile opamps: OPA627 (AD744?!) : ZeptoBars
This is why I will never buy a cheap $ op amp from a Chinese seller for a few dollars when the real ones costs much more from Ti directly or a known distributor like Mouser.
Rob43
Fake Audiophile Opamps Revealed | Hackaday
Fake audiophile opamps: OPA627 (AD744?!) : ZeptoBars
This is why I will never buy a cheap $ op amp from a Chinese seller for a few dollars when the real ones costs much more from Ti directly or a known distributor like Mouser.
Rob43
I agree. This is why I am used to buy via Audiophonics. it is a safe bet in Europe
+++ Happy you like it !
Should be nice to get opinions with some comparisons with other amps you've owned
'Revealing'
I can say for sure that the amp and DAC combo is one of the most revealing I've ever had. I'm not the most critical listener, but I have noticed extra little details in nearly everything I've been listening to since I go this amp several days ago.
A good example is on the 'Rome' album by Danger Mouse and Daniele Luppi (with Norah Jones and Jack White). There are many little electronic percussive things throughout this album. With the Aiyima, the way these come through was a little surprising at first, as they were more distinct and clear than my recollection of the album. Also, the weird processing effects on Jack White's voice were also more apparent, almost spooky. It seemed to me this was a more direct amplification of the source material when compared to other amps I've spent time with, including IcePower 500A, Tripath TA2020/TK2050, TPA3116/8, Panasonic SA-XR45 (TAS5182), and an old (but beautiful) Hitachi HMA-6500.
I'm certainly enjoying it.
I can say for sure that the amp and DAC combo is one of the most revealing I've ever had. I'm not the most critical listener, but I have noticed extra little details in nearly everything I've been listening to since I go this amp several days ago.
A good example is on the 'Rome' album by Danger Mouse and Daniele Luppi (with Norah Jones and Jack White). There are many little electronic percussive things throughout this album. With the Aiyima, the way these come through was a little surprising at first, as they were more distinct and clear than my recollection of the album. Also, the weird processing effects on Jack White's voice were also more apparent, almost spooky. It seemed to me this was a more direct amplification of the source material when compared to other amps I've spent time with, including IcePower 500A, Tripath TA2020/TK2050, TPA3116/8, Panasonic SA-XR45 (TAS5182), and an old (but beautiful) Hitachi HMA-6500.
I'm certainly enjoying it.
Last edited:
Many thanks for this enjoyable feedback.
It is important to collect people's feelings and this allows other members to understand that sometimes a small price, can hide a beautiful object.
I have compared myself my double Mono Discret Class D Amplifier (3x300W RMS with Nichicon and Epcos Caps + 800W SMPS LLC PSU) this morning with my Aiyima TPA3251 with the Burson V5i OP amps : I keep on saying that this little Aiyima is a a jewel capable of holding the comparison with amplifiers which are worth more than ten times its price. It even sounds better than my double mono...the details are crisp and the Soundstage is awesome.
I would be really curious to know what it looks like when faced with an Hypex Ncore or a Newclass D NCDX....
I can't wait to see the changes and improvements to come on this great Aiyima TPA3251. Thanks again to all the members here to share your knowledge !
I'm using the venerable EL70 Castle Microtower speakers in my avatar picture. They have been a staple for many years.
microTower bipolar ML-TL for CHR-70 or EL70
microTower bipolar ML-TL for CHR-70 or EL70
Someone said that on aiyaima amps you can get away without the opamp preamp section if you get 4V to it.
Since that would probably be anyone with an active preamp my question would be how to do that. I would cut the board traces after the input connectors, twist a pair of wires from there to the other end of the bipolar caps I would remove which would be the coupling caps after the gain stage. Which are those? This would bypass the pot and opamps.
Since that would probably be anyone with an active preamp my question would be how to do that. I would cut the board traces after the input connectors, twist a pair of wires from there to the other end of the bipolar caps I would remove which would be the coupling caps after the gain stage. Which are those? This would bypass the pot and opamps.
I had a quick look a while go at the pix here and also at the recommended schematics by TI.
So it could well be this manufacturer simply implemented the recommended TI schematic, perhaps though with some variations re part's values and of course the volume control potentiometer. At least if counting the caps etc. it appears to be quite consistant with the TI doc, but of course only people owning the unit and inspecting it can tell.
Anyway, if that's the case it should indeed be pretty easy to bypass the preamp section, either at the DC blocking caps Power amp entry or -even easier and without having to modify the unit- at the Op amps output as these are socketed. That would mean taking the existing op amps out and having a small pluggable adaptation board fitted that would link whatever source input you wish directly to the power amp section while keeping the DC blocking caps though (and thinking of course about keeping mass continuity with the input somewhere). To do so permanently you could if happy either tweak the unit's input RCA to get there directly, or fit a switch to commute the RCA between original path and your modified path, presumably running line cables on the right hand side.
Be warned though... the 4V you are refering to are probably (?) just what you think is worth having to enable driving the power amp section to your (ouput /LS) power needs. OK, you need 'a given voltage' due to the gain in the power amp section - howmuch voltage at the input depends a lot on your needs (LS and room and listening level). BUT you also want to be able to drive the power amp section, so you really want a quite low impedance source output and short quality cables etc.
Last but not least, some people find that the Class D chip's sound is somewhat sterile. I can't comment on that, but it certainly is on some Class D and therefore manufacturers are "tweaking" the sound of their input stage / preamp section in order to get the sonic signature you want. I feel this is a plus on that unit: the possibility to play very easily with the input section to get the sound you like. Of course, once you feed directly the power amp section, that possibility is gone, and you may or not prefer the direct sound... I guess it also depends a lot on your source. But my guess is that part of the sonic quality of this unit are due to the layout and some other part to the input section signature. As people reported here, swapping op amps makes quite a difference to the sound. Note in that regard that OPA1656 is quite neutral, very transparent / low distorsion, while still being musical (I know, sounds contradictory), flowing, helping a natural sound. So you don't lose anything but you may help making Class D more listenable (flow, 3D, natural sound, note endings). Something some listeners are sensitive to...
Good luck either way
Claude
So it could well be this manufacturer simply implemented the recommended TI schematic, perhaps though with some variations re part's values and of course the volume control potentiometer. At least if counting the caps etc. it appears to be quite consistant with the TI doc, but of course only people owning the unit and inspecting it can tell.
Anyway, if that's the case it should indeed be pretty easy to bypass the preamp section, either at the DC blocking caps Power amp entry or -even easier and without having to modify the unit- at the Op amps output as these are socketed. That would mean taking the existing op amps out and having a small pluggable adaptation board fitted that would link whatever source input you wish directly to the power amp section while keeping the DC blocking caps though (and thinking of course about keeping mass continuity with the input somewhere). To do so permanently you could if happy either tweak the unit's input RCA to get there directly, or fit a switch to commute the RCA between original path and your modified path, presumably running line cables on the right hand side.
Be warned though... the 4V you are refering to are probably (?) just what you think is worth having to enable driving the power amp section to your (ouput /LS) power needs. OK, you need 'a given voltage' due to the gain in the power amp section - howmuch voltage at the input depends a lot on your needs (LS and room and listening level). BUT you also want to be able to drive the power amp section, so you really want a quite low impedance source output and short quality cables etc.
Last but not least, some people find that the Class D chip's sound is somewhat sterile. I can't comment on that, but it certainly is on some Class D and therefore manufacturers are "tweaking" the sound of their input stage / preamp section in order to get the sonic signature you want. I feel this is a plus on that unit: the possibility to play very easily with the input section to get the sound you like. Of course, once you feed directly the power amp section, that possibility is gone, and you may or not prefer the direct sound... I guess it also depends a lot on your source. But my guess is that part of the sonic quality of this unit are due to the layout and some other part to the input section signature. As people reported here, swapping op amps makes quite a difference to the sound. Note in that regard that OPA1656 is quite neutral, very transparent / low distorsion, while still being musical (I know, sounds contradictory), flowing, helping a natural sound. So you don't lose anything but you may help making Class D more listenable (flow, 3D, natural sound, note endings). Something some listeners are sensitive to...
Good luck either way
Claude
Just as adition to help your calculation...
My (limited, I haven't studied this deeply and won't do it)) understanding is:
1- voltage gain of the TI chip alone is 20dB, so your input voltage x10.
4V input would voltage would mean 40V at the output. That's 400W at the output per channel into 4R LS load, more than what the unit can deliver, and still 200W into 8R speakers. Not sure why you want 4V, if you have them already, fine, if using a normal source with say 1 to 2V output in real life with real record's forte, that may already help you testing...
2- Input impedance of the chip seem to be 24k, so you are quite safe on that side unless using silly source output stages with very high impedance.
Good luck
Claude
My (limited, I haven't studied this deeply and won't do it)) understanding is:
1- voltage gain of the TI chip alone is 20dB, so your input voltage x10.
4V input would voltage would mean 40V at the output. That's 400W at the output per channel into 4R LS load, more than what the unit can deliver, and still 200W into 8R speakers. Not sure why you want 4V, if you have them already, fine, if using a normal source with say 1 to 2V output in real life with real record's forte, that may already help you testing...
2- Input impedance of the chip seem to be 24k, so you are quite safe on that side unless using silly source output stages with very high impedance.
Good luck
Claude
Hi amigos,
I am finaly back with my DIY Amplifier based on the "non public" Zero Zone TPA3255 board with Nover caps. This is gonna be my second attempt....
As reminder : months ago, my first attempt was a disaster because the LM2575 IC burnt with my 28V AC Linear PSU at first power ON. We ignored the reason : deffective board, deffective caps ? etc
December 2019 : first attempt :
Before I start the second attempt amigos (I am so scared lol) :
April 2020 : second attempt
1) Can I try without any speakers connected ? I guess TPA32XX are self protected
2) Would you try before with a 22V Toroidal PSU ?
3) Any other precautions you would think of?
Cross fingers now
I am finaly back with my DIY Amplifier based on the "non public" Zero Zone TPA3255 board with Nover caps. This is gonna be my second attempt....
As reminder : months ago, my first attempt was a disaster because the LM2575 IC burnt with my 28V AC Linear PSU at first power ON. We ignored the reason : deffective board, deffective caps ? etc
December 2019 : first attempt :
Before I start the second attempt amigos (I am so scared lol) :
April 2020 : second attempt
1) Can I try without any speakers connected ? I guess TPA32XX are self protected
2) Would you try before with a 22V Toroidal PSU ?
3) Any other precautions you would think of?
Cross fingers now
Hi Danuboun.
Before you start this amp up again with any power I seem to remember that the problem with the Chip that burnt was probably not with the chip but most likely that the Inductor that is part of this Buck circuit is far too small.
From the photos, it does not appear that you have replaced the Inductor with one of a higher value so it is most likely just going to burn out again !!
Just a thought
James.
PS. All my OPA1656 chips arrived from Mouser today so I will get them assembled onto DIL adapters ready to use in the various amps. Anyone else want any ?
Before you start this amp up again with any power I seem to remember that the problem with the Chip that burnt was probably not with the chip but most likely that the Inductor that is part of this Buck circuit is far too small.
From the photos, it does not appear that you have replaced the Inductor with one of a higher value so it is most likely just going to burn out again !!
Just a thought
James.
PS. All my OPA1656 chips arrived from Mouser today so I will get them assembled onto DIL adapters ready to use in the various amps. Anyone else want any ?
Hi amigos,
I am finaly back with my DIY Amplifier based on the "non public" Zero Zone TPA3255 board with Nover caps. This is gonna be my second attempt....
As reminder : months ago, my first attempt was a disaster because the LM2575 IC burnt with my 28V AC Linear PSU at first power ON. We ignored the reason : deffective board, deffective caps ? etc
December 2019 : first attempt :
Before I start the second attempt amigos (I am so scared lol) :
April 2020 : second attempt
1) Can I try without any speakers connected ? I guess TPA32XX are self protected
2) Would you try before with a 22V Toroidal PSU ?
3) Any other precautions you would think of?
Cross fingers now
I saw that... what would you think to try with a 22V PSU... My first attempt was with a 28VAC... With the 22V it would give me about 31V DC maximum.
LM2575 IC inductor is given to this Wide Input Voltage Range: 4.75 V to 40 V
Do you thing, it's better to give a try without or with the speakers connected ? in any case I have to test....
Last edited:
Hi Daniboun.
There are lots of people on here that are much more knowledgeable than me !!
These Voltage regulator chips oscillate at high frequency and create a voltage across the Inductor to produce the regulated voltage. My concern is that if the inductor is too small as in this case that it will still cook the chip regardless of the incoming voltage. This regulator is creating a 12 volt supply from whatever voltage you apply to the whole board whether this is 30 volts or 48 volts.
If you check out the TI Data sheet for this chip you will see they use a 330uH Inductor. You currently have a 10uH Inductor in the circuit which only a small fraction of what appears to be required.
I think most of the TPA3255 boards that use this regulator chip normally specify a 220uH Inductor at this point in the circuit. All the little bucks I have for other projects seem to have a 220uH Inductor.
I have many examples of chinese boards where the wrong inductor has been used - Usually a factor of 10 wrong. IE. a 100 Inductor (10uH) installed instead of a 101 Inductor (100uH) - Easy mistake to make if you are just reading the code off the inductor !!!
I am surprised the chip is only rated at 40 volts as I thought you had already replaced it with the higher voltage version which I think goes up to 60 volts. The regulator chips in the little Aiyima TPA3255 amps is rated at 80volts but this is from a different chip family. Unfortunately this is where many of the Chinese products cut corners if there is not a good quality control department.
James.
PS. Just soldered up my first two OPA1656 Chips so will test them shortly. Also just received a pair of OPA1622 chips so plenty to listen to over the next few days. Do you want any more OPA1656 Chips ?
A pair of OPA1622 chips - Made in China ! - Soldering questionable !!
A pair of OPA1656 chips - Made in UK by me ! - Soldered using 5.5% Silver solder !! Op Amps sourced through Mouser - Shipped from Germany !
Ready to start doing some Op Amp Rolling with my Aiyima 07, TPA3255 Amplifier.
I only have some little KEF Uni-Q Q15 Speakers to play with at the moment. First tests with original NE5532 chips I thought the speakers were bad ! Took me ages to find some decent quality source music and actually they now sound pretty good but could also because the amp has now been burnt in for a week. This amplifier shows up a bad music source very dramatically - With a good source, it sings.
A pair of OPA1656 chips - Made in UK by me ! - Soldered using 5.5% Silver solder !! Op Amps sourced through Mouser - Shipped from Germany !
Ready to start doing some Op Amp Rolling with my Aiyima 07, TPA3255 Amplifier.
I only have some little KEF Uni-Q Q15 Speakers to play with at the moment. First tests with original NE5532 chips I thought the speakers were bad ! Took me ages to find some decent quality source music and actually they now sound pretty good but could also because the amp has now been burnt in for a week. This amplifier shows up a bad music source very dramatically - With a good source, it sings.
