Hi all,
Long time lurker, first time poster
I've been working on a data logger for my local Scout county. I'm trying to interface a ATMega32U4 with a cheap Chinese RC522 RFID reader/writer (this contains the NXP MFRC522 chip).
Although 3.3v versions of the 32U4 circuit exist, they are significantly more expensive (for no particular reason). So, trying to keep costs down, for what is a very short run, I'm hoping to use the 5v version. However, this is giving me logic level conversion issues.
Testing things out on my Arduino Uno, I've been able to use SPI at 5v. I'm not sure what long-term damage I'm doing to the MFRC522 with this, so would prefer not to in the production version. The datasheet says that I can go Vdd+0.5v on these pins, and Vdd can be at maximum 4v.
Anyway, to the crux of my question. I could, obviously, and probably should, use a logic level convertor and a linear regulator. Bearing in mind the current draw for the MFRC522 is usually 10mA, peaking at 40mA when doing read/write operations. However I wondered if I could cheat a little. I've never seen any talk of this around which is either because I've not been looking or because it's a very bad idea. However, could I simply use a voltage divider to supply GND of the board with 1.7v. Then my power and data at 5v would then be (relative to GND on the board) 3.3v. Of course, signals will be -1.7v when pulled low, but I'm not sure what kind of issue a negative voltage might have.
Need to try and keep the overall cost down below £20 per unit, so if I could do this then it would be great. Otherwise, I may have to resort to using a 3.3v ATMega328P board and interface using a USB-Serial adapter. I know this is the more sensible solution, but it's likely someone will plug a pinhead in the wrong way round, whereas a USB connector is pretty fool-proof.
Many thanks for your help in advance, Chris.