off brand MMA45X board from aliexpress log error

[vikingmop]

picked up a couple of these more as a test than anything to see if they would work, I managed to get them working and they seem fine nudges are working etc.. but in my log file I get this red text not sure what it means doesnt seem to affect the function of the thing, I had to set the address to addr: 0x1C, instead of 0x1D to get it to work. just curious if there is something to fix the code thats coming up saying

2025/09/10 17:10:08 Error: MMA8451Q: WHO_AM_I (reg 0x0F)=0x2A (invalid; expected 0x1A)

[mjrgh]

The WHO_AM_I code means that the chip is actually an MMA8452Q, rather than MMA8451Q. They're closely related chips, and I've already been in touch with another person who tested one out and found that it worked correctly with the existing code, so hopefully it'll continue working for you, too. I've already updated my working version of the firmware to accept the 0x2A code and recognize the MMA8452Q specifically, so the error message should go away for you at the next release. In the meantime, it's harmless, and you can ignore it as long as the rest of it keeps working.

Note that the MMA8452Q is not quite as good a chip as the -1Q - it has slightly lower bit resolution (12 bits for -2Q vs 14 bits for the -1Q), so it can't resolve as fine gradations of nudge force as the -1Q. I suspect its noise level near rest might a little higher too, since most of the other chips with similar bit resolution are also a little noisier, but that's just speculation. It seems like the manufacturers are all prioritizing ultra-low power consumption over all else these days, at the expense of reading quality, so there's nothing on the market right now with the high resolution and low noise of the -1Q. These devices are marketed primarily for embedding in phones and similar devices, so they don't have to be very good - phone operating systems mostly only care about which direction the device is tilted, and beyond that don't need very detailed acceleration readings. Virtual pinball, on the other hand, benefits quite a lot from fine-grained readings so that it can distinguish between subtly different nudge forces. If you care about the better resolution, you might want to grab one of the Adafruit MMA8451Q boards while they're still available, since those are the real thing, and they'll only be available as long as Adafruit's inventory lasts, since the chip is no longer in production.

[vikingmop]

Makes sense, hence why they were a few bucks cheaper, good to know they seem to work pretty good so far, i'll give them a go I have the adafruit ones as well just in the testing stages and wanted to make a few up for cheap nudge solutions so trying out both options, good to know the next version will accept the 52 just seeing the scary red line in the log file made me think something was going wrong but then it worked as expected, any reason it would be a different address?

[mjrgh]

any reason it would be a different address?

The address on the MMA845x series is configurable with a pin labeled SA0. If you connect it to ground, you get one address and if you connect it to 3.3V you get the other address. It's just a matter of how that pin is configured on your breakout board. Most of the breakout boards expose SA0 as a header pin, so you can change the address by connecting that pin it to GND or 3.3V - it's probably connected internally on the board through a pull-down resistor to GND since you're getting the lower address. If you want it to be 0x1D, you should be able to connect the pin to 3.3V, although there's probably no reason to override it at the hardware level since you can just change it in the software as needed.

[vikingmop]

Installed the updated firmware today works like a charm no error in the log anymore and it seems to be just fine not noticing any difference from the adafruit 51 one as far as I can tell