question

linwood-f avatar image
linwood-f asked

Smartshunt 500 losing state

I have one of these on a 200ah LiFePO4 battery that I use for astronomy. It is permanently wired and powered all the time to be a SOC indicator. I bought Jan 2022 and for about a year it worked perfectly.

Of late however it loses its information. In the most recent example, I started at 100% and a full charge, took it out and used it for a night, came back went into my closet, and it showed about 75% charge remaining which was reasonable.

Two days later (today) I checked it, and it shows 100% charge. It is not on charge, just sitting on the floor. It shows no current in use, or used. It basically forgot its state.

I use the setup infrequently, charging only before each use. This is the second time in about 2 months that when I checked it, it had reset.

This is a new "feature" after a year of stability. Did it come with recent firmware (I generally update if it prompts). Or a sign it is failing? Some kind of a reboot?

This thing should work in an always-on mode, right? That's rather the point of a SOC coulomb counter, right?

Linwood

SmartShunt
1 comment
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

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

klim8skeptic avatar image klim8skeptic ♦ commented ·

@Linwood-F it is possible that the "charged voltage" is set a little high, and that the battery voltage has bounced back to above that level.

Pls check, and post a screenshot of the BMV battery settings.

0 Likes 0 ·
7 Answers
linwood-f avatar image
linwood-f answered ·

It's not supposed to go by voltage but by current used. And it did -- it correctly counted down to 75% (or so, I forget the exact amount), then the next day it reset to zero used, 100% charged. It is acting as though it reset. The 75% remaining was about right given the length of the night and equipment in use.

Here are what I think are the key screens. The history shows 4 cycles, that's too low, probably more like 12-15.


screenshot-firstscreen.png

screenshot-settings.png

screenshot-history.png


screenshot-firstscreen.png (99.2 KiB)
screenshot-settings.png (143.8 KiB)
screenshot-history.png (164.0 KiB)
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

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

linwood-f avatar image
linwood-f answered ·

By the way, I have no idea where that "Charged voltage" came from. This is the table from the battery manual I would have used to set it up. If that's a user entered field either (a) for a year it was wrong and worked, or (b) it reset itself. But SOC should be from current flow not voltage, right?!


charged.jpg


charged.jpg (33.4 KiB)
1 comment
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

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

klim8skeptic avatar image klim8skeptic ♦ commented ·

@Linwood-F Any time that the battery voltage bounces back above 13.2v, the SOC will reset to 100%.

Set your "charged voltage" to 0.2v below the absorption voltage.

The rest of your settings look OK.

0 Likes 0 ·
linwood-f avatar image
linwood-f answered ·

@klim8skeptic I'm unclear what absorption voltage is, does it correspond to float in the below?

I tried setting it to 13.6 and no change.

But please help me understand -- SOC is supposed to be based on current not voltage. Why does the charged voltage matter in this case?

Why did it read correctly on one day, then reset the next with zero activity? (And is that why it was 13.2, which pushes coincidence was the actual battery voltage?).

This is supposed to integrate current for SOC, not rely on voltage. It did that, worked perfectly for a full year. Now it is not. What happened?

But anyway -- at 13.6 there's no change in the SOC, still says 100%. I assume at this point it has lost all record of actual SOC and reset to 100%. It seems to be resetting itself. It seems broken.


parameters.jpg


parameters.jpg (95.6 KiB)
1 comment
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

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

klim8skeptic avatar image klim8skeptic ♦ commented ·

@Linwood-F As with all things, it is best to read the manual.

0 Likes 0 ·
linwood-f avatar image
linwood-f answered ·

@klim8skeptic thanks for the pointer. It confirms it should not reset without being charged (one of three conditions all of which must be met). I did not charge it, and it reset.

Unless I am just misunderstanding what appears to be plain english?

Or misunderstanding you?

Again -- it was at 100% after a full charge.

I used it for about 12 hours and it went to about 75% of charge, and indicated that correctly.

A day later, not connected to anything, not moved, just sitting, it indicated 100%.


1 comment
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

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

klim8skeptic avatar image klim8skeptic ♦ commented ·

@Linwood-F I used it for about 12 hours and it went to about 75% of charge, and indicated that correctly.

Under load the voltage would have dropped below 13.2v. Without load, the voltage bounced back above 13.2v. (screenshot says 13.21v)

Conditions have been met (13.2v / 3 mins / 0 amps) for the SOC to be reset to 100%.

0 Likes 0 ·
linwood-f avatar image
linwood-f answered ·

I guess I assumed the tail current had to exist, i.e. be above zero, for three minutes. There was no charge current, none, zero. It is supposed to detect fully charged by seeing charging activity, right? Not just voltage. While not stated, I assumed that mean < 8A but > 0A.

Are you saying that it does not have to see a non-zero charging current? That doesn't seem to make sense.

I also noticed that it says "time since last full charge" as zero. If it synchronized because it thought it fully charged, would that not be non-zero?

It also shows 10 sync's but 4 charge cycles. Other than when I first set it up, it seems those should be equal. The only other reason for a sync is a reset (it is set to sync to 100% on reset). And the only reason for a reset in this case is a spontaneous reset. It was just sitting there. And the last discharge should be about 50AH not zero.

I apologize if I am seeming argumentative, it is not intended, but this just is not making sense to me. Maybe in the morning with fresh eyes.



1 comment
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

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

pwfarnell avatar image pwfarnell commented ·
I will try and explain. The coulomb counter as you call it is not 100% perfectly accurate, slight discrepancy in peukert factor or charge efficiency or small loads below threshold value. Often therefore for most users the SOC may get back to 99% during charging when the battery is full. The synchronisation settings are then used to indicate to the SmartShunt that the battery is full, i.e. the battery is under charge because the voltage is above the charged voltage setting AND the charging current is lower than the tail current for 3 minutes. This is a necessary feature of any shunt based counting system to avoid errors building up. The trouble you have is your charged voltage setting is within your normal battery range so the SmartShunt thinks it is being charged. Just increase it to something like 13.6V or 13.7V, just below the voltage you charge at and it will work. It will not go back down to 75% SOC but will work on the next cycle. Suggest you recharge to full and start again.


You also state there should be more cycles, the manual explains that a cycle only occurs if the SOC goes below something like 65% and recharges above something like 90%. I have had a set of batteries on a boat for nearly two years and because they only get down to 80% minimum have zero cycles.

1 Like 1 ·
kevgermany avatar image
kevgermany answered · 
There's also the possibility that you have a wiring/connection problem. If the shunt disconnects, it will reset to 100% charge. See the battery starts synchronised setting.
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

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

linwood-f avatar image
linwood-f answered ·

@Kevgermany It's not impossible but unlikely. This last reset for example, occurred after I transported it home and had it in the closet. After being there it showed 75%, a day later with quite literally nothing touching it and no movement it went to 100%. The little spring loaded positive feed to the shunt's electronics is not exactly the most reliable type of connection, but it didn't reset in transport from a disconnect so I doubt it reset sitting still.


@pwfarnell OK. I'm still unsure why it would reset without any charge amp flow, ever, but I can give it a try. I looked on the charger and its absorption voltage is 14.2 (didn't know to look there when it came up above, this is a Victron BSC IP67 12/17 charger) so using 14v as the "charged voltage" on the shunt and will see what happens.

But when you say "thinks it is being charged" -- it has an absolute indication of charge across the shunt, which direction current is flowing. I just do not see why it should look at voltage to decide if it is being charged or not?

But... I just want it to work, so let's see if this fixes it.

2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

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