Anybody have problem with MCP3202 with Vdd=+5V ?

Rayman · 2012-01-19 15:54

Asking because I'm getting ready to make another version of our Merlin board with an ADC instead of motor driver and have this question about the MCP3202...

Anyway, I'd like to power the MCP3202 with +5V, but the datasheet shows the digital input threshold as 0.7*Vdd minimum, which is 3.5 V.

I think I remember other chips rated for 0.7*Vdd not working reliably...

But, I think I see a lot of people here using it without problem...

Has anybody had problems at 5V and had to switch to 3.3V to make it work?

JonnyMac · 2012-01-19 15:59

I do it all the time in my own and in client projects; have never had a problem. The only thing I add to the circuit is a 3.3K resistor between the DO (MISO) pin and the Propeller.

Rayman · 2012-01-19 16:07

Thanks Jonny. I was just looking at your MCP3202 driver in OBEX and saw how you hooked it up.

JonnyMac · 2012-01-19 16:10

I've actually been going a bit more traditional lately, keeping DO and DI separated. I seem to be using more chips with SPI interfaces so I'm using SCK, MOSI, and MISO lines as one usually would, with a CS\ for each chip that shares the buss. Recently did a customer project with an MPC3208, MCP2515 (CAN interface), and four thermocouple interfaces on a shared SPI buss. We actually used and I2C expander to handle all the chip select lines.

Rayman · 2012-01-20 05:45

I'm still worried about using MCP3202 at 5V, but I think I'll do it anyway since Jonny does it all the time and nobody seems to have problems with it.
Still, I really don't like using things outside of their specified range...
But, I don't want to sacrifice the increased functionality just to be picky about specifications...
Maybe it's only 1 in 1000 chips that would have a problem.

BTW: I've noticed that a lot of chips I look at have the 0.7*Vdd specification. Must be something fundamental in their I/O...
Makes me wonder if we'd be better off running the Prop at 3.5 V...

Duane Degn · 2012-01-20 08:53

I've used the MCP3208 at 5V a lot (with the Prop). I haven't had any problem. I've seen other chips with the 0.7*Vdd logic claim (some shift registers) that work fine with 3.3V logic.

Would using a MCP3208 instead of a MCP3202 cost much more? I find I can always use more ADC channels on my robots. Two just don't seem like they would be enough.

Rayman · 2012-01-20 09:17

Thanks Duane. This is for a robot, so I'm iterested... What do you need more analog channels for?

I was thinking maybe 1 for an IR proxy sensor was enough...

Duane Degn · 2012-01-20 09:39

I really like to be able to monitor the battery packs voltage. I use a voltage divider if needed so the max voltage isn't about the ADC's Vdd. I also like to add light sensitive resistors (or transistors) to monitor light level or to look for the brightest/darkest areas.

There are lots of sensors that give an analog output. The accelerometer in a Wii mote has three analog outputs (I keep intending on reusing some of these).

If you make your own force sensor from anti-static foam, an ADC could read the analog value from it. Those little thumb sized joysticks use two pots which give analog outputs. And you can never have too many knobs (pots read by ADC).

It seems like a lot of homemade sensors are analog.

I don't know if there is a big difference in price when you buy the chips in bulk, but I noticed the other day, the price difference for a 2-channel ADC chip verses an 8-channel ADC chip from Parallax's store, wasn't very much.

jazzed · 2012-01-20 11:48

I use this for battery monitoring to save pins and real-estate.

Rayman · 2012-01-20 13:02

I think I've decided to switch to the MCP3204. This actually fits the stencil I have already pretty good and gives 4 channels instead of 2.
It's still rated at 0.7*Vdd, but hopefully it's just as forgiving as MCP3202...

JonnyMac · 2012-01-20 13:19

I've used all three of that family of chips in my own and commercial circuits without a lick of difficulty. Like most vendors, I think Microchip is being very conservative with its specs; not that I think one should abuse them, mind you.

frank freedman · 2012-01-20 23:02

Rayman wrote: »

Asking because I'm getting ready to make another version of our Merlin board with an ADC instead of motor driver and have this question about the MCP3202...

Anyway, I'd like to power the MCP3202 with +5V, but the datasheet shows the digital input threshold as 0.7*Vdd minimum, which is 3.5 V.

I think I remember other chips rated for 0.7*Vdd not working reliably...

But, I think I see a lot of people here using it without problem...

Has anybody had problems at 5V and had to switch to 3.3V to make it work?

No issue using the 3201 version. Used 5V because allows max sample rate. Only need to remember series 4.7k series resistor on data out line into prop.

FF

Anybody have problem with MCP3202 with Vdd=+5V ?

Comments