Hello!
My JS110 seems to be broken, seeing -3na current which messing up with our battery estimation.
When nothing is connected to it settles down at ~-3na and +15v.
Can you help me calibrate the joulescope, please?
Hi @tmbraja - Based on your message, I suspect that your JS110 is still working as designed.
For the current, the spec from the Joulescope JS110 User’s Guide is:
So -3 nA is well within the ±30 nA offset specification. For what it’s worth, the JS220 has an in-field offset calibration that usually brings the offset to within 1 nA. The JS220+ spec is ±1.0 nA for ±5 °C over 24 hours with trim offset calibration.
For the voltage, the JS110 and JS220 both omit the pull resistors that keep the inputs from drifting due to opamp input bias currents. Most multimeters and oscilloscopes have 1 MΩ to 10 MΩ resistors, but these cause too much error for many Joulescope applications. Instead, Joulescopes intentionally let these inputs float. The downside is that they will drift due to the opamp input bias currents when left disconnected. As soon as you connect the voltage inputs to the circuit, it should work. If you short IN+ to IN-, you should see:
Do you agree that your JS110 is operating within specifications?
What is the current that you are trying to measure, and what accuracy do you need?
Hi @mliberty
Thanks for the quick response. We are using the JS110 to measure/accumulate current to calculate the battery life. The average idle current of our device is around 60uAmps. The JS110 used to work well before, but for some reason, it is now accumulating around -0.2nA with the same setup as before. Do you have any recommendations for us to try?
Here is our setup,
Hi @tmbraja - I am not sure I am following what is wrong. The JS110 has always been specified as ±30 nA offset accuracy in the most sensitive 18 µA range. Both the -2.8 nA you originally shared and the -0.2 nA you just shared are well within this spec.
For 60 µA, -3 nA is -0.005% error, which is very good. Note that you will also observe gain error. In addition to being within spec, it seems that the JS110 is more than accurate enough for this measurement.
What am I not understanding? What has change from when you thought the JS110 was working well? Is it the negative offset error that concerns you?
You are not showing how the power supply is connected. You have to be very careful with unintentionally providing other ground return paths. Your JS110 is isolated, but the Geneva RS232 Debug Cable is likely not. If you unplug the debug cable, do you see any difference in the current measurement?
Hi @mliberty
We have been using JS110 successfully for a while; it is just that the one I have seems to be having some issues recently, it worked fine before. I am not able to figure out why/how.
The working JS110 with the same setup
The failed JS110 with the same settings and setup, the detect current amplitudes are much lower than expected.
Do you have any recommendations on how to calibrate the JS110?
And don’t see any change whether I have RS232 cable or not.
Hi @tmbraja - Let me see if I understand what you are saying:
You have at least two Joulescope JS110s. Let’s call them working and misbehaving. When you connect the working JS110 to your system, you measure the first waveform image. If you change nothing except swapping the misbehaving JS110 for the working JS110, you measure the second waveform where the amplitudes are 1 / 1000 of the working JS110.
Is this correct?
Note that Joulescopes are designed to never need recalibration, and there certainly is no calibration process that will fix a factor of 1000x error. Although it does not happen much, the most common JS110 failure mode are the capacitors for the Ćuk converter. See Section 17.4 in the Joulescope JS110 User’s Guide.
Yes, correct, the measured current amplitude from misbehaving JS110 is significantly lower, like 1/3000. I don’t have both JS110s, I am looking for some kind of self-test/calibration test to ensure it’s an issue with JS110.
Is there a way to test the Ćuk converter without opening the device?
Hi @tmbraja - Ok, so you suspect the JS110, but it sounds like you have not yet performed any tests to confirm.
The JS110 does not include any self-test or self-calibration. You can roughly follow the Verification Procedure in Section 16.1 of the Joulescope JS110 User’s Guide. For your purposes, even a power supply and 5% resistors should be enough to give you confidence.
You need to open the instrument to test the Ćuk converter.
Hi @tmbraja - I am following up. Where you able to perform any simple measurements to confirm if the JS110 is actually working correctly?