Shop OBEX P1 Docs P2 Docs Learn Events
MCP3202: SPI D[out] through 300 ohm resistor? And decoupling question(s) — Parallax Forums

MCP3202: SPI D[out] through 300 ohm resistor? And decoupling question(s)

mmgoodmmgood Posts: 19
edited 2015-05-16 16:45 in Propeller 1
Delurk post here. Hi!

1) I've been glancing around at various "cookbook" articles re MCP3202s. This one: http://www.gadgetgangster.com/tutorials/397 shows a 300-ohm resistor running from D[out] to the "commoned" D[in] + Prop input connection. I've seen other folks just show connecting all three points with no resistor.

Can anyone tell me what's up here? Is the resistor "a thing"?

2) The datasheet says the chip only draws 375uA in operation. I'll be running it from a Prop Mini with +5V external power fed directly to the 3.3 regulator's input. I'm thinking of powering the ADC with a separate TO-92 LM317 or perhaps some 3-pin LDO regulator (fed from +5V, too), and getting V[ref] either from that (with some cap bypass) or with its own regulator.

Are there gotchas? Am I likely to get 10 real bits out of this arrangement? [Yes, I knw that noise on the input could mess me up, too, but the source is a 5k pot so I have some hope there]

Informed opinions are very welcome. Cheers.

Comments

  • JonnyMacJonnyMac Posts: 9,108
    edited 2015-05-12 15:12
    You really should use 3.3K or higher, not 300.

    Here's what's up: Many of use connect the DO and DI pins through a resistor to minimize connections to the Propeller. By putting a 3.3K resistor in the line (and connecting to the DI side of the resistor) you prevent a possible short circuit should the Propeller DIO pin be in output mode when DO is active. Even if you're using separate DO and DI pins, you should put a 3.3K resistor in the DO line when using the MCP3202 at 5V -- I do this all the time (see attached)
    1000 x 542 - 258K
    1000 x 1000 - 347K
  • mmgoodmmgood Posts: 19
    edited 2015-05-12 18:50
    JonnyMac: Thanks! Makes sense. I appreciate the prompt and thorough response.

    I certainly understand the cheap mixed-voltage current limiting trick, but I hadn't thought of the "both chips driving" problem. D'oh!

    MM
  • frank freedmanfrank freedman Posts: 1,983
    edited 2015-05-15 21:00
    Might be worth noting that this family of chips can run happily on 3V or 5V. Depends on what the sample rate needs to be. If you are under 50ksps, you can run at 3V and maybe avoid the 5v and the regulator to supply it.

    FF
  • edited 2015-05-16 16:45
    I'm using 4.7k resistors in the data, clock and chip select lines of the DS3201 version powered at 5 volts. Using PASM, I toggle the clock line every 100 ticks which works out to 400 khz.

    Actually I'm running two chips at the same time from a single set of clock and chip select pins and two data pins. See attached code.

    Sandy
Sign In or Register to comment.