Measuring very small resistances with ADALM1000 part 2
This blog installment continues the discussion on making a Milliohm meter using the ADALM1000 and AD8210 current shunt monitor IC. Using the solder-less breadboard to connect to the AD8210 can be a little flaky with the results shifting when the wires are wiggled. To try to minimize the variability, I hand soldered a small adapter board for the BOB mounted AD8210 from the ADALP2000 kit. The AD8210 BOB does not fit into a standard DIP IC socket so I had to improvise something using female pin headers. Not perfect but better than the solder-less breadboard.
Figure 1, Hand soldered adapter board.
Further testing of the configuration suggested in part 1 of this topic has shown that offset and linearity of the AD8210 when the output is near ground is not very good. Connecting pin 7 (VREF1) to 2.5 volts as shown in figure 2 will reference the "zero" current point at 2.5V/2 or 1.25 V. This takes away from the total range (by about 1/4) but gives much better accuracy. One or two hundred mV above ground is enough shift but this was the easiest way to shift the output.
Figure 2, AD8210 connections to center Vout at 1.25 V (2.5/2)
Going even further I designed a small 1" by 1" plug in break-out PC board to mount the SMD AD8210 and connect it to the ADALM1000 and provide a place to connect the 4 force and sense wires / test probes.
Figure 3 SMD break-out PC board
In the picture the Milliohm meter board is shown connected to a 4 pin Vishay (VPR221S) 2 ohm 0.05% calibration resistor.
Using the full blown ALICE desktop scope display for this milliohm meter is overkill. A first pass at a standalone tool much like the other DC tools offered in the ALICE software package has been written. It should be included with the most recent release of the ALICE tools. A screen shot of the standalone tool is shown in figure 4. It includes an example schematic at the bottom as a reminder for how to connect the AD8210. Controls to manually and auto zero the channel B offset voltage and channel A offset current are included.
Figure 4, Milliohm meter software tool
The AD8210 will likely have some small output offset. As the software runs, with the auto zero boxes checked, first the channel A current is set to 0, i.e. not sourcing any current, the average measured channel A current and channel B voltage will be the automatically entered in the offset entry locations. Then channel A is set to the test current and the unknown resistance is measured and displayed on the top line.
When the Auto Zero boxes are not checked you can manually enter the values. The second line reports the measured channel A current and channel B voltage. If the CA Test I is set to 0 these will be the offsets.
If the Test current is set too high for the resistance being measured such that the channel B voltage goes above 4.8 V the line displaying the voltage (and current) turns red.
The gain accuracy of the AD8210 in the datasheet is specified to be +/- 0.5% Max and the calibration accuracy of the ADALM1000 is likely in the same range. The software has entry places to adjust the gain as well. In the screen shot of figure 4 I have adjusted the current and voltage gains (really only need to change one) such that the reading is exactly 2.000 ohms for one of the Vishay calibration resistors. The total adjustment was 0.8% which is within the range that we might expect. I checked the second 2.000 ohm calibration resistor I have and it gave identical results.
It should also be noted that the board I used in this case had pin 7 connected to +2.5 V so the channel B offset (zero current value) was around 1.3 V.
Alternate way to auto zero
Also since the current source in the ADALM1000 is bipolar it should be possible in a test version of the software to alternate between both positive and negative test current and null out any offset that way. In figure 5 we show pin 7 connected to the +5V supply. Now the "zero" current output of the AD8210 will be at +5V/2 or +2.5 V. Because we have half the voltage range at the output of the AD8210 we have to also reduce the magnitude of the test current by half.
Figure 5 ADALM1000 Connections for bipolar auto-zero
A screen shot of a test version of the software that implements this bipolar test current technique is shown in figure 6. In this case there are no check boxes to enable auto zero or places to enter the offsets. We still need places to enter the test current and adjust the gain. The reminder schematic at the bottom is changed to show how pin 7 should be connected for this version of the software.
A second copy of the PCB was configured for this technique. The screen shot shows the results for the same 2 ohm calibration resistor. For this board the total gain adjustment needed was slightly different (1.3%).
Figure 6, Bipolar test current software version
Using the Milliohm Board as high current Ammeter
A side benefit of this AD8210 break-out board is that with a known external shunt resistor is can be used as a high current Ammeter. In the next photo I show a hand wired shunt using a 0.12 ohm power resistor and two screw terminals. The exact value of the shunt can of course be measured using the Milliohm meter software, after calibrating it against the Vishay resistor.
External shunt resistor connection example as Ammeter
One Final Thing
Here is another example of test leads for the Milliohm meter. Here I show using two mini-grabber cables I made up a couple of years ago for another project.
Another set of test leads
Schematic and board files for the AD8210 based Milliohm meter are attached below.
As always I welcome comments and suggestions from the user community out there on other fun ways to use the ADALM1000 and the ALICE software tools.
Additional resource links: