ADS1118 ADC - is minimum Vcm really exactly GND?

Dec 08, 2016 5 Replies

formatting link



I have noticed that with some delta-sigma ADCs, e.g. MCP3550, they don't stop working and their specs don't degrade if you go a few mV below the GND rail - and why should they? The most that should happen is that input protection diodes might just about start getting turned on.



The reason I ask is because I want to measure voltages of a few hundred mV but they can also go below zero by up to 10mV.



It's a long story but I am combining the measurement of various sensors, with the measurement of thermocouples, and thermocouples can produce small negative voltages. Very small; even -270C produces only



-10mV.



Obviously there are other ways to get around this (e.g. lifting the whole sensor interface up by 10mV) but I am trying to keep things simple because I am trying to measure various other voltages :)


Some d-s ADCs, like AD7793, have a buffer amp that can be enabled or disabled. When enabled, the input cm range no longer includes ground. Otherwise, a bit of negative input is fine.

Without the buffer, there will be some charge injection spikes sneakout out of the input pins, which are no problem for a passive signal source with some bypass caps. The spikes can mess up opamps.

ps - AD7793 sucks.

John Larkin Highland Technology, Inc lunatic fringe electronics

It's important to keep zero-signal on scale, so a small positive offset is a big win.

Cheers

Phil Hobbs

Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net

John Larkin wrote

OK - many thanks.

Just in case, I have put in a little software-switchable "lift" which lifts the base of the TC to +20mV and that will take care of even the lowest temperatures.

There is a specific type of TC which is apparently used in cryogenics, down to -270C (3K). Funny that... years ago I read an old book on cryogenic physics and they used carbon composition resistors to measure low temps.

From what you say I should put some caps on the op-amp buffer outputs, with a series R so the op-amp dooesn't go unstable. I am using the tlv2333 chopper stabilised op-amps as buffers on all four inputs. However the Zin is claimed to appear largely resistive (I know what you mean - obviously it isn't) at 710k - 22M depending on the range.

Sorry - two misconfigured usenet clients... John Smith is me :)

You can buy calibrated silicon diodes as cryo temp sensors. Below 20K, the voltage drop gets huge as the carriers freeze out, so you get lots (volts) of signal. Lakeshore makes some.

There are also exotic resistors, which work better in magnetic fields.

The ADC and the chopper amps will both kick out charge spikes, which can cause microvolt offsets.

John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required