Infineon MA12070 Class D

"One listens to a system and not individual components" a wise man once said while not being able to distinguish individual components 🙂

-input stage and 2 opamps removed, remaining 3 opamps replaced for OPA1642, output RCA removed including circuitry, almost all electrolytic caps replaced for Panasonic FC/FR, output coils replaced for 10A (5 mOhm) rated ones , coupling caps bridged with Wima 1 µF MKS. Bass boost switch removed and contacts bridged. That's it.

BTW I also used these coils in a Sabaj A10a and it made more difference in that one. Sabaj A10a has no bead or coil filtering except 4 x 1.5 nF caps used straight from each output to GND which one never sees elsewhere AFAIK. The Sabaj A10a is a newer revision without the ugly cutoff/sawn PCB and it now has bass whereas it previously was not as good in that area as the A8. This revision (no number) also has 4 identical 10 µF electrolytic caps for coupling the MA12070 to NJW1194 and it is already better than the previous 2020 version.

Just info for those that also like these amplifiers.
 
Last edited:
  • Like
Reactions: TNT and mamocel
Hi all! I need a broadband (1-60kHz) power amplifier (1-10W) to sweep the output signal over a serial RLC circuit (R = 8-16 ohms, C = 0.2uF, L = 0.5mH), not for music. I have already ordered the ma 12070 board from Audiophonics.fr. My problem is that I haven't found which elements define the frequency band. Most graphics from Infineon have a very sharp cutoff at 20kHz. I'm assuming it's the digital cutoff on the input. On some other graphs, I have observed a normal (slow) cutoff, even around 100kHz, but again without any indication under what conditions this was obtained. May I ask the DIY guru to answer my question.

 
Not a guru but I think you have some items wrong. It is not digital, there is no digital cutoff as it is not digital and neither does it cut off at the input AFAIK, it has output filtering due to the nature of class D.

You probably have success by using only ferrite beads at the outputs as mentioned in the datasheet to avoid very high frequency products.
 
Jean-Paul, many thanks for your reply. I'm afraid my English is far from to be perfect and this creates some miss-understandings. I mean that a step-like cutoff at 20kHz is not due to some low-pass filtering which typically is proportional to w/wo^(-n), with n=1,2,3...., or 20dbu, 40dbu...per octave depending on the filter order. I also found the following results obtained for the same chip.,
see https://www.audiosciencereview.com/forum/index.php?threads/loxjie-a30-amplifier-review.17547/
 
So whats your best bet on how to crack this nut open? I got it for 26USD shipped from Amazon.

Joints seem sturdy. There are 4 screws deep down under the feet which I undid without any apparent change. Brute force? Any ideas?

//
 

Attachments

  • DSC_0993.JPG
    DSC_0993.JPG
    123.2 KB · Views: 110
  • DSC_0994.JPG
    DSC_0994.JPG
    151.2 KB · Views: 101
  • DSC_0995.JPG
    DSC_0995.JPG
    167.5 KB · Views: 105
  • DSC_0996.JPG
    DSC_0996.JPG
    135.8 KB · Views: 97
  • DSC_0997.jpg
    DSC_0997.jpg
    144.3 KB · Views: 103
Last edited:
OK - so the bottom do come out once the screws are out and a bit of bending force is applied in one of the holes... looks like really well built, clean, structured with orderly marking of things.

I wonder:
  • If and how one could mobilise the BT - I dont want that HF around.
  • The volume pot - is that "digital" control of the Infineon chip or do the analog signal pass it?
  • In general make it into a power amp without any of the USB/card features...
  • Exchange input caps
  • Exchange output coils
//
 

Attachments

  • DSC_1002.JPG
    DSC_1002.JPG
    179.8 KB · Views: 116
  • DSC_1001.JPG
    DSC_1001.JPG
    242.8 KB · Views: 121
  • DSC_1000.JPG
    DSC_1000.JPG
    187.6 KB · Views: 115
  • DSC_0999.JPG
    DSC_0999.JPG
    180.3 KB · Views: 112
Last edited:
Can't be volume control of the MA12070 itself I think. Should be the other PCB. Nice amplifier board, I have seen worse. If you make detailed pictures of the sole amplifier board we can see if coils* will fit.

*On picture DSC_1001 we get a glimpse of SMD coils. Larger than on the picture in post 1327 (which also looks to be a completely different PCB).

Volume control seems to be logical when it really contains a MA12070P.
 
Last edited:
Yeah I guess. You may be OK with that but the loudspeakers and neighbors may not.

No wait I guess volume will be 0 as nothing connected so it will be open circuit. If you make a picture of both 3 pin connectors we can see that better. I think you'll need to connect those to eachother if you want to use the RCA connectors.
 
Last edited:
Careful pull the inner connector. Not the cable itself.

Hint: "open" power amplifier + volume control in DAC/source + small software error/glitch = catastrophe so sort of problema.

Also fiddly when pulling cables with the amplifier being switched on. Bzzz, BRRRRRR........ BTW you are busy with individual components if you were not aware of that 🙂
 
Not bad, they skimped on the capacitors as usual. Also they omitted the planned output network that you may wish to add being an audio rebel. I think removing the RCA connector block will make it possible to glue a heatsink to the chip and then you can use the cables to connect to better RCA connectors. If you'll use a new metal casing make sure to drill output connectors so that they'll fit exactly like the way it was in the plastic casing. Short signal path is always better, certainly with switching stuff. The back of the plastic casing is an excellent drilling template.

A small study to which output network and coils you wish may be necessary. The Infineon document will help you. Verdict from pictures: worth the money and probably fun to try out.
 
Last edited:
You may wish to use the "gentle total demolition ®" method. A 160 side cutter will do wonders. First cut off the heads of the snakes, so the white and red with shields sideways. Then the black part may be bent forward to do further connector dissection.
Power connector idem ditto. Gentle but remorseless demolition without any damage (= no mechanical force) to the PCB. One needs a fresh/sharp side cutter for connector sushi.

When the plastic is gone mostly removing the metal parts from the PCB is easy and done in seconds.
 
Last edited: