Mooshimeter Update
Hi Readers,
The Mooshimeters have been at the testing lab getting their regulatory certifications. We hit a bump but are hopefully past it.
The Good News:
The Mooshimeter passed emissions testing without a hitch!
The Bad News:
The Mooshimeter demonstrated susceptibility to radio interference that messed with the current reading. In the lab, when blasted with radio waves in the neighborhood of 200MHz, the meter would erroneously display a large current (up to 2A). That’s not acceptable.
The Problem:
The test lead is acting as a large antenna and blasting our poor current sense amplifier with RF noise. Instrumentation amplifiers have a well known but poorly characterized ability to rectify common mode RF noise. Analog Devices have a nice paper about it here.
The designer must protect amplifiers from RF noise through careful filtering. In this case there was a gap in the filters that subjected the reference input of the instrumentation amplifier to too much RF noise. This noise is rectified and appears as an erroneous voltage offset within the measurement chain.
The Fix:
We recreated the issue in our own lab using a HackRF Jawbreaker – ProtoDave, thanks for the loaner! This allowed us to quickly test the changes to the RFI filters. With the changes in place we are unaffected (error is below measurement threshold) by interference sources between 20MHz and 6GHz.
The production design has been updated to include new filter components, which were added by hand on the boards going through regulatory. They’ll be back in the compliance test lab this week to undergo radio immunity testing again.
Thanks
~James
Thanks very much for the update!
Really looking forward to the day it shows up and I can play with it. I want to plot the current draw of an RC car motor in real time!
Thanks for the encouraging update and for the link to HackRF, pretty cool !
Any chances to share more details on a final tests update like the open-source SmartScope team just did 3 weeks ago https://www.kickstarter.com/projects/751733865/smartscope-reinventing-the-oscilloscope/posts/828930 ?
Keep the great work going, you know where you are and where you want to be :-) 2 thumbs up!
Thanks Beat,
Whatever information we get back from the test lab we will share. We want you to know what you’re getting :-)
Congratulations on the fix! I know how much of a pain EMC testing can be.
I have a question on the accuracy specifications. You specify 0.5% DC voltage accuracy, but don’t give any offset error information. On a cheap Chinese product I would interpret that as 0.5% of the 600V full scale, for up to a 3V error even near 0V readings. On a Fluke I would expect to see an accuracy specification of something like 0.05%+1digit, meaning a gain error of 0.05% of the reading and an offset error of 1 count on the display. Could you list the voltage/current offset separately from the gain when reporting accuracy?
I know I just estimate those figures looking at the op-amp or in-amp datasheet offset voltage and factoring in the gain of the resistor network, but without seeing a schematic, I can’t do that myself.
I have been anticipating the arrival of the mooshimeter on my doorsteps since last year.
Hopefully this year, I can give it a test run at my lab.
Best,
Dr. D