1.875Mbps serial

I have a few sensors with serial output at 1.875Mbps, 8 data bits, 1 start bit, 1 stop bit, and doesn't seem to have a parity bit. It sends 4 bytes of data, waits about 4 bytes worth of time, then repeats. I need to send some of these 4 byte chunks of data to the computer connected to the USB port, say at least 1 out of 1000, but more often would be better. It would be nice if I can read more than one sensor per board if CPU speed allows.

Which board would I need to do this?

What do I need to do to get 1.875Mbps baud rate?

Which board would I need to do this?

One of the SAMD21 based boards should be able to do the job. Forget all AVR based boards.

What do I need to do to get 1.875Mbps baud rate?

Use a SERCOM based hardware serial interface.

Thanks, I bought the MKR Zero board and it works well now.... For some reason I had to use a lower baudrate in the 1.715 to 1.845Mbps range to get it to work. I am 99% sure the sending devices are 1.875Mbps so I guess the MKR Zero oscillator is off a little or something.

I am 99% sure the sending devices are 1.875Mbps so I guess the MKR Zero oscillator is off a little or something.

Did you check that with a scope?

Baud rates are approximations in most cases as they are constructed by dividing the main clock. As 48/1.875 is 25.6 there's no exact match. At 1.845 you get almost exactly a factor of 26. If you use the exact value (1.875) the divider will be 25 resulting in a frequency of 1.92Mbaud.

BillC48:
Thanks, I bought the MKR Zero board and it works well now.... For some reason I had to use a lower baudrate in the 1.715 to 1.845Mbps range to get it to work. I am 99% sure the sending devices are 1.875Mbps so I guess the MKR Zero oscillator is off a little or something.

Perhaps not.

Consider how your serial send and receive operate. The sending of a bit occurs at the the beginning of the Baud rate clock signal. The receiving of the bit MUST occur sometime after the beginning of the Baud rate clock signal. If perfect, the receive clock will be in the middle of the bit time.

Since the two Baud rate clocks are not synchronized, the receive clock will have to be different from the TX clock by some amount of time and at that speed, the window for determining the bit value is very tiny. So, if it works, it works.

Paul

I did check with a scope, but it says my 1KHz ref is 1.1KHz so I can't trust the scope at all, time for new scope and reference.
The factor /26 works, the /27 works too, but /25 doesn't.
Now on to my next 2 problems... (in another post)