Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Batrium Shunt inaccurate current measurment
Chances of two faulty units seems pretty slim?
Have you looked for electrical noise when charging? Don't suppose you have an oscilloscope?
Running off solar, DIY & electronics fan :-)
I disagree, chances of two that have the same error are very likely if they have been programming each of them with the same incorrect calibration curve, which is what I think is going on. Maybe something changed in the physical shunt manufacturing process and they have been using an old cal curve?

I do have a scope, and have not moved it over to try. I would be surprised if that was the case with a shunt measurement scheme. I would suspect it for hall effect. Really if it was a noise issue I'd still ding them for not having proper noise filtering /decoupling in their hardware design. But depending on what they tell me I'll give it a try.
I did a detailed test to characterize the issue. I have a FLIR voltmeter accurate to 1mV, so I ran a test where I measured shunt voltage to 1mV accuracy, and current through the shunt reported by the clamp meter (which agrees with the sunny island exactly) and the Batrium. I sampled all three at 5A intervals. 

Attached are the results.  

You can see a region where everyone agrees from 10A to 70A. Then a region of growing error from 70A to 140A, and quick correction after that to 150A. But the shunt mV remains linear the whole time. 

This SCREAMS calibration issue. 

[Image: 0SzfYJNbImQLqvfuyRtNJ5TqRxy324jpYKM8rbth...14-h670-no]
Good data :-)
Well good in that it shows the fault!
I didn't think shunts would have point by point correction table type calibration so I'm surprised by the data.
Was thinking they'd just have a linear correction value & maybe an start point offset value.

There has to be something unusual going on (obviously).
You've got us all scratching our heads here!

Could the sampling frequency of the shunt be getting affected by switching noise from the charger?
The meters are going to be sampling differently so may not be affected by noise?
It only happens when this particular charger is running right?
At these sort of amps there'd be some noise right?

Can you post a pic of the setup?
Running off solar, DIY & electronics fan :-)
These are Sunny Islands I'm using to charge. If it's a noise issue I'd still fault Batrium EE design for insufficient filtering in the shunt. I am tempted to open one and look at their design to rule this out, but risk voiding my warranty and they don't take it back. Also them may have potted the PCB anyway. I don't have another 56V 120A charger laying around to prove that theory, and there's not much we can do in the way of filtering external to the shunt, without some major capacitors.

Well Maybe I can wait for a really sunny day and verify the 80-90A range as that is what my solar system max should be. But that's a month or so away probably.
I've isolated the issue and tested a fix. During the range of inaccurate reading, there's a 20Mhz ringing present across the shunt with a large magnitude of 120mV. I separated the shunt PCB from the shunt and added a low pass filter between the two. This fixed the error, and all readings are accurate now.

I've reported it to Batrium so we'll see where it goes from here.
cowpen likes this post
Nice to hear you've been able to isolate it & thanks for getting back to us so others can look out for similar issues!

Wiring inductance might be a factor here on the major DC cable paths?
Running off solar, DIY & electronics fan :-)

Forum Jump:

Users browsing this thread: 1 Guest(s)