1
\$\begingroup\$

I am using a few LTC6102 ICs in a current measuring system. My main goal is to use these ICs to measure current in different areas of a circuit and then display them on an LCD connected to an Arduino. I am feeding the Vout of the IC to an analog-in port on an Arduino Mega to read it. Schematic

I have confirmed that my AnalogRead() code is working correctly, and it is very accurate when measuring a steady DC input. I tested this with a simple voltage divider before hooking my LTC6102 Vout to it.

Now, when I connect my Vout from my LTC6102 to my Analog Input, my Serial Monitor shows this mess:

AnalogRead on Serial Port

No consistency. When I connect the probe of my oscilloscope to Vout, I get this:

Scope shot of Vout

I have come to a conclusion that my Serial Monitor mess is due to the fact that my IC is not providing a steady output (like shown in the o-scope shot) and that is causing my AnalogRead() inconsistencies since it only reads every so often and the value is always changing. I attached a 100nF capacitor to my Vout, and that keeps the value much much more consistent. I would like to better understand why this looks this way and make sure I have things correct.

My main question is: Does the oscilloscope shot of my Vout from my 6102 IC make sense?

\$\endgroup\$
6
  • 8
    \$\begingroup\$ You need to show a schematic or it's hard to guess what the problem is. \$\endgroup\$
    – user1850479
    Feb 17 at 1:45
  • 2
    \$\begingroup\$ Please clarify your specific problem or provide additional details to highlight exactly what you need. As it's currently written, it's hard to tell exactly what you're asking. \$\endgroup\$
    – Community Bot
    Feb 17 at 1:59
  • 1
    \$\begingroup\$ You're measuring a 10kHz spiky signal. The multimeter will average it, the micro's ADC will sample it when the ADC is triggered, that's probably a random point on the waveform, so the values you get are normal. \$\endgroup\$
    – bobflux
    Feb 17 at 9:01
  • 2
    \$\begingroup\$ Please show us a schematic, and a layout. It's possible you made a mistake there (no bypass capacitor? too much output capacitance? noise from another noisy circuit nearby?) \$\endgroup\$
    – Hearth
    Feb 17 at 18:34
  • 1
    \$\begingroup\$ Firstly, isn't R1 too low for the current that you are trying to measure? Voltage across R1 is just 2.1uV (lower than even the offset voltage of LTC6102 hence may not be possible to reliably measure) and the expected output is just 116uV isn't it? May be you should change R1 to say 1ohm or 5ohm and check if there is any improvement. \$\endgroup\$
    – sai
    Feb 26 at 18:13

2 Answers 2

Reset to default
1
\$\begingroup\$

Thank you all for your input! I found the source of my problem. There is a mistake in the Mouser description for the shunt resistor I ordered. I was looking for a 1mOhm resistor so I keep the losses to a minimum (20A system so losses multiply quickly). I ordered these without looking at the datasheet because I thought there wouldn't be much to look at due to the simplicity of a resistor. Whoops! Turns out they are 1Mohm.... Mouser lists 1M in the description but 1m in the specifications. I will definitely learn from this mistake, especially considering the headache it has caused.

Mouser Info
(Image source: Mouser - YAGEO AC2512FK-7W1ML)

\$\endgroup\$
0
\$\begingroup\$
  1. Scope measurement: The scope may be showing you measurements of the average value or RMS value of the signal waveform. Play a bit with the measurement settings such as max, min, average, dc volts and ac volts on the scope and compare that with your DMM reading. I would trust the scope measurement more than the DMM measurement. Note that what you see on the display of the scope is the actual real life voltage waveform.
  2. DMM measurement: If your DMM is set to measure DC voltage it will essentially be measuring something close to the average value of that spiky signal. If set to AC voltage measurement it will probably be measuring the RMS value of that spiky signal - this is a type of average AC signal value. Standard digital multimeters are not very good at characterizing any signals other than pure DC or pure sinusodial AC signals. Don't make too much of this measurement.
  3. Microcontroller ADC: Remember that you microcontroller can only sample every now and then - say every 10us (really depends on the microcontroller and how it is configured). The values on the serial terminal may actually be correct and they represent the instantaneous value of your spiky waveform at the time it was sampled. It may be a good idea to plot the values measured by your ADC in a graph or spreadsheet to get an idea of the waveform that your ADC is seeing. If your ADC has too much capacitance on its input it will also not be able to see fast signal changes.
  4. Current measurement: Try placing your oscilloscope probes across your current sense resistor and verify that the current waveform looks like your output voltage waveform. Check that your sense resistor is correctly sized according to the datasheet. Make sure that you don't place your scope ground on a point in the circuit that is not grounded as this will interfere with the circuit and might damage things.

Possible solutions: If you are interested in the average current being measured you can place a capacitor on the output of the current sense amplifier to filter the voltage to a DC value. You can also try increasing the ADC sampling speed and averaging the reading over e.g. 10 samples.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.