question

buddhafragt avatar image
buddhafragt asked

Smart Shunt inacurate

I use a Smartshunt in series with a shunt in my DIY-BMS which is very accurate. (Booth in minus string).

Whan I switch on a 10W load, my BMS shows 10W, the smartshunt shows 25W.... what is wrong here? Defect? Or the Smartshunt is in general so inaccurate?

(the problem is not the 0 point calibration, the values above are deducted from other loads)

Another point is:

I activate in Venus DC System.... Sometimes Venus calculate PV Power - consumers = DC System value, sometimes Venus use the value from the Smartshunt!

Everything not so dramatic, 10W or 30W (a Shelly for 15 EUR is much more accurate), but Venus calculate the charging power for the battery related to the consumers. This is dramatic!!

SmartShunt
2 |3000

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

6 Answers
Trevor Bird avatar image
Trevor Bird answered ·

hello @buddhafragt , the smartshunt is in essence a very simple device that would at the most basic component would be very accurate. I think it has a voltage drop from memory of 50mV for 500 amps so it is just an accurate resisitor. It may be helpfull to see what the voltage and current readings are for both shunts and make some observations relating to the likely source of the error.

For instance, obviously both shunts should be reading the same current. If that is not the case one of the shunts is reading incorrectly or and extra load is applied to one shunt and not the other. What happens if you reverse the posotion of the shunts in the series circuit? Does the anomoly swap to the other way around.

If both currents are the same now check the voltage that both shunts are reading. if the voltages are different it may point you in the direction of the anomoly.

I would exhaust all possible scenarios before concluding the smartshunt is that inaccurate because the probability of that being the case is very small.

Trevor


2 |3000

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

buddhafragt avatar image
buddhafragt answered ·

Hello Trevor, thanks for your response!

It is not a difference of 10%, we speak about 100%. Booth shunts are in series, so the current must be exactly the same, right?

I also notice, all Victron devices have the problem of extremely high inaccuracy.

But I try to measure the voltage over the shunt....

Another problem with the Smartshunt I have described here, which is also not yet solved:

https://community.victronenergy.com/questions/182087/smart-shunt-and-phoenix-strange-behavior.html

Perhaps Victron can comment on what is the problem with this device .......

Michael

3 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.

nickdb avatar image nickdb ♦♦ commented ·
The accuracy is published in the specs: 0.4% for current and 0.3% for Voltage.


0 Likes 0 ·
dancrompton avatar image dancrompton commented ·
I'm currently struggling with a Lynx Shunt that can be up to 250mv off. Given their sole purpose you would have thought they would be more accurate.
0 Likes 0 ·
Trevor Bird avatar image Trevor Bird commented ·
Well i agree the current should be the same but are they reading the same? What does the smartshunt current read compared to the other shunt? Now what does the voltage read compared to the other shunt? Is it the voltage that is reading differently or the current? Now lets see what the actual voltage is with a meter and verify which reading is wrong. You may have a faulty shunt but I would rate that as an unusual circumstance. Let's see what the actual voltages and currents are and see what is different between the two shunts. Something is different and one of those parameters is wrong. It is interesting to see where this anomoly lies.


0 Likes 0 ·
Nev avatar image
Nev answered ·

Perhaps your SmartShunt is faulty?

Our 600Ah LiFePO4 RV battery has a 500A Victron SmartShunt and a much cheaper 400A Junctek KG140F battery monitor shunt in series.

Both monitors always agree to within 1 or 2% SOC across the entire typical 40 to 100% SOC range providing that 100% synchronisation occurs at least every few days.

Either both monitors perform accurate calculations or much less likely they are inaccurate by a similar amount.



2 |3000

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

JohnC avatar image
JohnC answered ·

@buddhafragt

Have you performed a Zero Current Calibration?

Regardless, trying to measure down to like 10W and expecting laboratory accuracy from a 500A shunt is a bit much. The shunt doesn't even report Amps to better than 1 decimal point, and on a 48V batt, that's 5W steps at a minimum.

2 |3000

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

buddhafragt avatar image
buddhafragt answered ·

see my post: (the problem is not the 0 point calibration, the values above are deducted from other loads)

I have also tried it with a higher load, with the same result.

Also funny the difference between the result in VictronConnect and the Remote Console! ;-)bildschirmfoto-2023-03-20-um-104955.png


2 |3000

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

Trevor Bird avatar image
Trevor Bird answered ·

@buddhafragt I note on your remote console you have various boxes with PV on output, PV on input, loads on input,etc. Is the CerboGX set up for ESS? Do you need it to be ESS or is it just a normal setup not needing to be ESS?

Is it an ESS system? The system is very configurable so it is important to know exactly how the CerboGx is setup to know why the apparent anomoly occurs. Just to ponder for a moment the addition of all power readings equates tot he output of the Inverter. The 42 watts shown on "Connect" is about the thermal loss of the idle current of the inverter. Without information on how the system is exactly setup it is hard to comment more.

2 |3000

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