question

imoore avatar image
imoore asked

BVM712 voltage inaccurate

Hello,

I have read through some threads discussing this, but am still not sure the best course of action. I have the BVM712 hooked up. Everything is hooked up correctly and seems to be working well, but the voltage is consistenly lower than the actual measured voltage at the battery. The only connection that could be faulty is the supplied cable with fuse that plugs into the shunt. I have checked it and it is fine. I measure at the point where the cable connects to the battery and get the proper measurement, but the BVM shows less voltage.

I read that one user subbed in a 1A fuse for the supplied 100A fuse and that corrected the problem, but this seems strange that Victron would supply something that gave a faulty reading when the whole point was it's accuracy.


Is this really the best route to try to solve this problem? What are other users doing to correct this?

monitoring
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

4 Answers
Mark avatar image
Mark answered ·

Yes, if you have a 100mA fuse fitted change it for a 1A fuse.

The 100mA fuse has a high resistance that results in a voltage drop across it, which effects the accuracy of the BMV voltage reading.

8 comments
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

imoore avatar image imoore commented ·

Thank you Mark. I have replaced the fuse with the 1A and it has definitely improved the accuracy. The BMV is still showing slightly lower( at this point 12.60 instead of 12.64). Is this considered in "spec"? It just seems strange to me that monitor who's main function would be to give accurate battery readings would be inaccurate and not have some type of function to compensate for this, if so.

0 Likes 0 ·
Mark avatar image Mark ♦♦ imoore commented ·

What are you using as you 'master reference' equipment to measure the voltage and was this calibrated recently?

If your just using a cheap multimeter there is a good chance that it could be a little out. You can also try reversing the multimeter probes & seeing if the reading changes apart from just being a negative voltage.

Anyway the spec for BMV voltage accuracy according to the manual is +/-0.3%, which is about +/-0.04v when measuring 12v.

No this can not be user calibrated.

Extract from manual;

Out of interest, can you confirm that you have connected all negative cables to the large M10 shunt bolts and NOT to the PCB screws?

Also check all other connections including on the negative side. A poor connection / high resistance anywhere in the circuit will reduce the voltage reading.

0 Likes 0 ·
imoore avatar image imoore Mark ♦♦ commented ·

Mark. thanks. I tried it with 2 separate high end DMMs; a Fluke and a B&K. The voltages are right. The disparity is in spec with the manual. I apologize, but had not yet looked at the tolerances as when i started they were so far out. I will call this good for the time being. I have the negative cables connected to the shunt bolts. I'm not sure which PCB screws you are referring to.

My ground reference is completely solid, and i can confirm that by checking different reference points to it. That's why this was initially so confounding. The only possible culprit was in the supplied power cable, and indeed, it turned out to be a fuse that was creating too much resistance and throwing the voltage reference out of wack.


Thanks again for your insight.

Ian

0 Likes 0 ·
Mark avatar image Mark ♦♦ imoore commented ·

No worries.

Out of interest did you try probing with the DMM directly between the brass shunt end block (battery side) and B1 terminal on the shunt PCB?

Is the voltage measured between those points closer to what the BMV is reporting?

Also, have you tried measuring/comparing the voltage with 0/no load? - as you will also get a voltage drop over the short negative cable between the battery & shunt when under load.

0 Likes 0 ·
imoore avatar image imoore Mark ♦♦ commented ·

I have the 712 mounted in a pretty tight location. I will try to get my DMM into the B1 location to measure. Is there a place on the PCB to do the same. It may be difficult to get a measurement from it.

The voltage seems to be more accurate when there is more input voltage than output.

0 Likes 0 ·
Mark avatar image Mark ♦♦ imoore commented ·

I am a little confused by your statement - you don't need to access the BMV-712 display unit (if that's what your referring to).

My suggestion to measure directly only requires access to the shunt & the PCB mounted to it, B1 is the positive input terminal on the PCB.

My suggestion was to measure voltage directly between the 2 points circled in red (brass shunt end block on battery side & B1 terminal).

Due to resistance & voltage drop in the short negative cable between the battery & shunt, during discharge the voltage measured directly at the battery terminals will be slightly higher than the voltage measured by the BMV. During charging the opposite will occur.

If you have a negative cable with that has a small cross section &/or long length &/or a poor connection - this unavoidable small difference will be worse.


0 Likes 0 ·
1561949537516.png (77.9 KiB)
powerstroke2000 avatar image powerstroke2000 commented ·

I'm trying to figure out why Victron wouldn't send out two different fuses, if this is truly the case?

0 Likes 0 ·
cpt-pat avatar image cpt-pat commented ·

I had the same issue -- and the same solution, but with a twist:

First, my test instrument: I'm using a Fluke model 117 multimeter that just recently received a factory NIST-traceable calibration. I was reading 3=13.17 volts at the battery terminal, and 13.07 volts on the BVM monitor. A full 0.1 volt error (which matters a lot with LiFePo4 batteries).

I unscrewed the fuse holder expecting to replace it with a 1 amp 3AG type fuse and.. "oh no. They used a mini glass fuse. I don't want to have to stock those on board." So, I temporarily reinstalled the fuse and.. the reading on the BVM hopped up to 13.13 volts - with the same 0.1 amp fuse. "Corrosion! Drat!" Those cheapie fuses Victron used only have to smell saltwater to corrode.

I replaced the fuse with a proper water-tight ATI fuse holder and 1 amp fuse -- and now the voltage reading is correct.

Comeon Victrom: you make a great monitor, so splurge a little on a batter fuse holder!

0 Likes 0 ·
offgridsa avatar image
offgridsa answered ·

I have referenced my BVM 712 along side my Fluke 87v and HP Agilent 8.5 count bench meter.

I was pleasantly surprised the BVM 712 was bang on within its full resolution.

As mentioned above I to took readings as close to the where the BVM pcb is reading the voltage and not the battery terminals. Though I did take battery terminal readings for a less critical comparison and there was little difference.

I could have just got lucky with a nice one that fell bang in the middle of tolerance spec range by chance though.

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Chris avatar image
Chris answered ·

Mine measures within 0.03v on my new and certified Fluke 87v.

0.03v is pretty great IMO. I measured at the battery which could account for the small difference. I would like the ability to calibrate the BMV-712 to match exactly.

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

powerstroke2000 avatar image
powerstroke2000 answered ·

I'm having a similar issue, albeit, the difference between the Victron 712 and my multimeter (I have two and both so the same result) where my Lithium batteries are showing 13.91V consistently, yet the 712 is showing 13.45V. To me, that is an inconsistent variable, so I'm trying to get to the bottom of this issue. I've been keeping an eye on this daily, and the difference between my meter and the 712 is always there...why is this, and how do I correct it? I do have the Victron set up as the seller of my Battle Born Lithium batteries have suggested.

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Related Resources

Additional resources still need to be added for this topic