I'm working to repurpose the trackball assembly of a Logitech Trackman wired mouse for use in an experimental rig. I assumed that Logitech would use Avago's ADNS series of optical flow sensors, but it doesn't appear that way. I've managed to carefully disassemble down to the sensor, but still can't find a part number. One side of the chip reads "M8aSN" but this returned nothing from Google. The sensor assembly had 6 leads running to a PIC, while the sensor itself has four pins.

I'm hoping that some sleuths here might have some ideas. Thanks in advance.

Logitech mouse specs: https://support.logitech.com/en_us/product/trackman-marble/specs

Avago ADNS series sensors (seemingly not what I'm looking at due to the difference in pin-out) [PDF!]: http://media.digikey.com/PDF/Data%20Sheets/Avago%20PDFs/ToolKitSelectionGuide.pdf

Lastly, here is the most similar-looking component I've been able to find so far. It has four pins though it is surface-mount: https://www.vishay.com/ppg?84286

Photos of what I've got:

chip-side: chip-side

mysterious "M8aSN": mysterious "M8aSN"

full assembly of LEDs, chip, components-- this has 6 leads full assembly of LEDs, chip, components-- this has 6 leads

  • \$\begingroup\$ Do you have schematics of how it was used in the original application? \$\endgroup\$
    – Hearth
    Commented Feb 1, 2019 at 19:56
  • \$\begingroup\$ That photo doesn't appear to be at all similar to Avago ADNS chip. \$\endgroup\$
    – glen_geek
    Commented Feb 1, 2019 at 21:30
  • \$\begingroup\$ @Hearth I have no schematics whatsoever for this unfortunately, which is why I've turned to the EESE community for some help. \$\endgroup\$ Commented Feb 2, 2019 at 14:38
  • 1
    \$\begingroup\$ @SpencerWilson It doesn't look like that complicated a module, though. It seems someone else has already answered, but in the future it would help if you could trace out a schematic for any future questions like this. \$\endgroup\$
    – Hearth
    Commented Feb 2, 2019 at 14:40
  • 1
    \$\begingroup\$ @SpencerWilson The extra two no doubt are for driving the LEDs. \$\endgroup\$
    – Hearth
    Commented Feb 2, 2019 at 14:47

1 Answer 1


I thought those things had become obsolete!

This is more anecdotal than anything, but the Trackball Marble by Logitech was the first to use optical sensors in this way. I actually worked with some of the researchers that designed the first versions of those ICs, some of the historical pedigree is here. The one you found might be more recent versions of the design (and very likely proprietary to Logitech).

The original ICs were actually what is known as analog VLSI (aVLSI) optical flow sensors, these come from the field of Neuromorphic engineering. They do analog calculations based on the optical array that represents the movement of the optical field (loosely based off an insect's eye). This is the original conference paper (PDF download link) from the research team that created it: A CMOS Motion Detector System for Pointing Devices, Xavier Arreguit et.al.

You can see that Figure 1 of the paper very closely resembles the layout of the sensor you found.

Optical flow sensor concept

According to the 1996 paper, the output of the sensor IC is two 5-bit digital words one for delta-X the other for delta-Y. It only uses 4-pins, two for power one for interface clock, and one for data output.

You should be able to decode the pinout by tracking the signals on the PCB, but it is clear that the large mounting pad in one of the two center pins must be ground.

Although the fundamental idea is similar to what Avago's sensors use, the principle of operation is completely different. Avago's design is fully digital, consisting of a low-resolution camera and digital signal processing circuitry.

  • \$\begingroup\$ Wonderful answer, thank you kindly. Lyon's history is particularly fascinating for me as this experimental rig will be used for neuroscience research, in a vision lab! Full circle, it seems. I am extending my question here, but do think it's likely that I'll find some sort of data sheet for this chip, or enough information to decode the bytes from the sensor, through this unknown PCB, to a new microcontroller? Any tips/references for how to go about trying? \$\endgroup\$ Commented Feb 2, 2019 at 14:37
  • 1
    \$\begingroup\$ @SpencerWilson finding a datasheet is extremely unlikely as this is a proprietary sensor, but you should have much better luck with the researchers (unless NDAs are involved). Vision is a big topic in aVLSI research and there have been a lot of developments in this area. You might even be able to get access to research ICs. Some names that come to mind: Tobi Delbruck, Kwabena Boahen, and Reid Harrison (the founder of Intan technologies). All of them have done vision aVLSI ICs at some point in their careers. \$\endgroup\$ Commented Feb 2, 2019 at 14:54
  • \$\begingroup\$ Thanks Doc Brown! I will try my luck there. Much appreciated. \$\endgroup\$ Commented Feb 2, 2019 at 15:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.