I installed a new 100Ah AGM battery, reset the smart shunt stats and fully charged so all sync’d at 100%. Then disconnected the EHU and turned on the fridge to do a duration test. Below are the graphs for the last 2 days as well as the settings from the shunt. I then read up on Peukert and adjusted that to be 1.0 which updated the battery to 73%. (I still need to figure out the right value here, my power usage should be small, I think it’s unlikely to go over 150w) But my battery is 100Ah, I set the discharge floor to 50% so I’d expect the reported battery percentage to be 51% based on a 100Ah battery with 50Ah of usable capacity and a used figure of 24.5Ah So it seems like the discharge floor is not being factored into the capacity calculation? Is there some setting I’ve missed? [soc] [consumed] [history] [IMG_0480]

So doing a bit more fiddling with the settings. At 50% discharge floor the display is showing 1d4h to go. Seems about right, actually impressively correct as the current load is 9w, that constant, the fridge kicks in about 3 times per hour at 40w (ish). It’s been running about a day, I’d expect it to run about another day. So that prediction seems to take into account the periodic increase due to the fridge. At 0% discharge floor that time increases to 4d1h. Seems reasonable. But in both those cases the battery percentage remains at 71%. I’d expect to see the percentage to reflect actual usable amounts. Not to need additional knowledge about the battery chemistry to understand that for me with an AGM 50% is bad.

I hadn’t but it doesn’t seem to address the fundamental problem. The prominent display of battery percentage does not take into account the discharge floor. So IMHO is kinda useless for lead acid batteries. I should also mention I’m running the beta firmware v3.50~30 in case the behavior I see is different.

The shunt does not take account of the discharge floor. SOC is how much you have taken out from the total capacity. It has always been like this, it is a industry definition of SOC. The 50% discharge floor is also not a hard limit, you can go beyond this, you just have fewer cycles. This is nothing to do with beta software. Your Peukert of 1.0 is too low for AGM, high quality AGM will end up somewhere in the 1.10 to 1.15 range, lower than the default for flooded leas acid.

Agreed I need to set the Peukert correctly, that was just part of my experimenting. While I appreciate what you are saying about SoC I don’t see why there can’t be a setting so we can display the available charge from 0-100% like every other battery display. How much I keep back (to prevent damage, or just to have a bit of spare) should be configurable. Discharge floor seems a handy number, but no reason it couldn’t be a different setting. The new gui is so lovely, and beta would be a fantastic time to introduce such a thing. [image] 64% means, put the lights on, charge that ipad, plenty of power, put the tv on. 14% means, I need to be a bit more frugal or my fridge is going to turn off.

Battery SoC is wrong

DIY

nesswill (Dave) 11 October 2024 12:19 3

Hi
Have you read This might help

Topic		Replies	Views
Wrong state of charge smartshunt DIY diy , ess , smartshunt	13	3795	13 September 2024
Battery not charging well on sunny days DIY beginner , multiplus , leisure_batteries , battery_charging	6	209	19 June 2026
BMV-712 Smart SOC not accurate with low current consumption Q&A and troubleshooting diy , lithium_battery	6	144	23 January 2026
SmartShunt not syncing to 100% DIY diy , smartshunt	24	504	31 March 2026
Smart shunt battery percentage adjustments Q&A and troubleshooting diy , offgrid , lithium_battery	6	518	17 January 2025

Battery SoC is wrong

Related topics