12bit ADC MCP3208 Vdd 3.3V and Vref 5V possible?
StefanL38
Posts: 2,292
Hi,
I would like to know if it is possible to use the MCP3208-chip with powersupply-voltage 3.3V and Vref 5V.
The datasheet says All inputs and outputs w.r.t. VSS ............... -0.6V to VDD +0.6V. I interpret this as
If Vdd is 3.3V the maximum voltage to VRef and any ADC-input-pins is 3.3V + 0.6V = 3.9V.
Or does it mean maximum-input-voltage is Vdd-max + 0.6V which would mean 7+0.6 = 7.6V?
Of course I can set Vdd to 5V and then add current-limiting resistors of 4.7k for connecting to the propeller-chip.
But I would like to avoid this.
best regards
Stefan
I would like to know if it is possible to use the MCP3208-chip with powersupply-voltage 3.3V and Vref 5V.
The datasheet says All inputs and outputs w.r.t. VSS ............... -0.6V to VDD +0.6V. I interpret this as
If Vdd is 3.3V the maximum voltage to VRef and any ADC-input-pins is 3.3V + 0.6V = 3.9V.
Or does it mean maximum-input-voltage is Vdd-max + 0.6V which would mean 7+0.6 = 7.6V?
Of course I can set Vdd to 5V and then add current-limiting resistors of 4.7k for connecting to the propeller-chip.
But I would like to avoid this.
best regards
Stefan
Comments
You should only need one current limiting resistor when powering the chip with 5V. The Dout pin is the only pin connected to the Prop with 5V on it.
Technically 1K isn't enough to limit the current to acceptable levels (according to datasheet). I think a lot of people use 1K resistors but the official recommendation is to use a higher value.
Really? Only 500ua?
That's what I've heard (and read).
Definitely not a good idea to have Vref any higher than Vdd. From the absolute maximum ratings "All inputs and outputs w.r.t. VSS ..... -0.6V to VDD +0.6V". I got curious and tested a chip up to VDD + 0.7V. The readings became noticeably non linear after Vdd + 0.4 and the input was damaged at Vdd + 0.7 (constant 4093).
You can't even do that, its guaranteed input high value is 3.5V (0.7Vdd) so the Prop cannot be guaranteed to talk to a 5V MCP3208 without full level-shifting on the SPI bus.
I guess it's not guaranteed, but it's done that way a lot.