How does a DMM measure positive and negative voltages?

[JuanGg]

Just out of curiosity, what are the common ways used to measure positive and negative voltage on multimeters?
I have searched online and looked at some schematics, but I couldn't find much.
What I came up with is that, after the voltage goes through the input dividers, it could have some offset added to it so it falls in the midpoint of the ADC's range (but then you loose half the resolution), or it could go through a precision rectifier and then to the ADC, having to detect beforehand if the voltage being measured is positive or negative.
Same thing with positive and negative current.
Any ideas?

    Juan

[Kleinstein]

The usual ADCs used in DMMs have a bipolar input range. So zero is about in the center of the range.  A few add an offset and use a single polarity ADC.

The way with rectifier is not common with DMMs - I have seen it with an analog meter once.

[magic]

Find datasheet of ICL7106, the "brain" of most cheap 3.5 digit DMMs.

Ground of the DMM is generated somewhere in the middle between the two battery terminals, which become a positive and negative supply. Input dividers are referenced to ground rather than to the negative battery terminal and the ADC also measures with respect to ground. No precision level shifting or polarity inversion is required.

Nowadays, much better delta-sigma ADCs are available with differential inputs and 0 to ±VCC input range. For those, I suppose a circuit could be made which shifts the virtual ground between 0 and VCC so that the input signal always stays within range.

[radiolistener]

When there is 0 V between two probes of multimeter, it means 0 volt.
When there is -311 V between two probes, it means minus 311 volt.
When there is +311 V between two probes, it means plus 311 volt.

Just try to measure battery in different polarity, multimeter will detect where is plus and where is minus with no problem.
The same story for AC. AC voltage is the same as fast polarity switch of the battery.

Sometimes negative pulse on AC wave may be different from positive one.
It's the same as you put one battery with right polarity and then put the second battery with inverse polarity.
The multimeter can distinguish them by sign.

[soldar]

radiolistener, we all know DMMs can read positive and negative voltages. The OP is asking *how* do they do it. Several posts before yours have offered explanations.

[JuanGg]

 Thank you everyone. I'll take a look at a few datasheets / service manuals.
    Juan

[Doctorandus_P]

DMM's ofthen have a bunch of analog switches.
They first charge a capacitor to a voltage proportional to the input voltage, and then they measure the voltage (or charge) of the capacitor.

Datasheets of DMM chips are of course a good reference to learn how they work.

You can also have a look at chips like the ADS1118.
This is a 16-bit delta sigma ADC with a PGA as frontend and a bunch of integrated analog switches to adjust the polarity.

Another source of high quality info are the service manuals of High-end bench top DMM's.
Service manuals for 6, 7 and 8- digit benchtop DMM's from HP / Keithley and other brands are floating around the web.
Some of them are 100+ pages long with complete schematics and very detailed descriptions of all the circuits inside.

[JuanGg]

The usual ADCs used in DMMs have a bipolar input range. So zero is about in the center of the range.  A few add an offset and use a single polarity ADC.

The way with rectifier is not common with DMMs - I have seen it with an analog meter once.

The Keithley 168 does in fact use this approach. See screenshot from the service manual attached.

I've been looking at the service manual of the HM8012 (http://www.sm5cbw.se/hameg/hm80/hm8012-serv.pdf), but I haven't been able to figure it out. It uses the LTC2400 ADC, which is single polarity only. I don't see any offset or rectification, although I'm probably missing something obvious.
    Juan

[Kleinstein]

The HM8012 is in deed using an extra switchable inverter stage. An odd solution in my opinion, as it add another offset and a different gain factor for the negative side.

The part from the Keithly 168 looks more like the AC part - for DC one would not want to have the large capacitor to slow things down.

[JuanGg]

The HM8012 is in deed using an extra switchable inverter stage. An odd solution in my opinion, as it add another offset and a different gain factor for the negative side.

The part from the Keithly 168 looks more like the AC part - for DC one would not want to have the large capacitor to slow things down.

The Keithley 168 seems to use that circuitry for both AC and DC, at least as per the footnote. There is also no need to measure polarity if only used in AC.

Just saw the inverting op-amp on the HM8012  . Thank you
    Juan

[Dave]

I think you missed these.