Anyone else struggling to get accurate SOC readings with a cheap shunt on a lithium battery?

by Ben Webb · 1 week ago 63 views 4 replies
Ben Webb
Ben Webb
Member
6 posts
Joined Jan 2025
1 week ago
#8018

I picked up a 100A shunt monitor off Amazon for about £12 to track my 100Ah LiFePO4 leisure battery in the van. Set the capacity, wired it in correctly (negative side, all loads through the shunt), and it seemed fine for the first few days. But I'm noticing it drifts quite badly — after a week or so it's showing 40% when the BMS and my voltage readings suggest it's closer to 70%. I've been resetting it manually when I get a full charge but that feels like a bodge.

I'm wondering if the shunt itself is inaccurate, or whether it's a calibration issue with the way LiFePO4 voltage curves work. The flat discharge curve always seems to confuse these cheaper monitors. I've also got a slight suspicion there's a small parasitic draw somewhere that's not going through the shunt, though I haven't tracked it down yet.

Has anyone had similar drift issues with budget shunts on lithium? Would stepping up to something like a Victron BMV-712 actually solve this, or is it more likely to be a wiring/parasitic draw problem that a better monitor wouldn't fix either? Keen to hear what others have found before I spend £100+ on a BMV.

Chopper
Chopper
Active Member
10 posts
thumb_up 2 likes
Joined Nov 2024
1 week ago
#16046

@BenWebb yeah classic issue with cheap shunts — the calibration is usually miles off from the start, and LiFePO4 makes it worse because the voltage curve is so flat it can't "reset" itself properly like lead acid does.

Main culprits:

  • Incorrect shunt resistance value hardcoded in the firmware
  • No proper full-charge sync point to reset to 100%
  • Peukert exponent set wrong for lithium (should be ~1.05, not the lead acid default)

Ditched my Amazon unit after a week. Running a Victron BMV-712 now — pricey but the SOC accuracy is genuinely solid and syncs to 100% reliably. Fogstar cells deserve better than a £12 monitor tbh.

If budget's tight, the Daly BMS built-in monitor is at least a step up from the random Amazon shunts.

Crispy Welder
Crispy Welder
Active Member
10 posts
Joined Jul 2024
1 week ago
#16145

@BenWebb had exact same headache with my garden office setup before I upgraded. The flat voltage curve on LiFePO4 basically exposes every flaw in a cheap shunt — voltage-based estimation is useless so coulomb counting accuracy is everything.

Biggest culprit I found was the tail current at full charge — if your shunt doesn't see a proper low-current absorption phase to reset to 100%, it just drifts further out every cycle.

Switched to a Victron SmartShunt and it's night and day. Syncs via Bluetooth, lets you fine-tune the charge efficiency factor, tail current threshold, the lot. Worth every penny of the ~£60.

If budget's tight, the Renogy 500A shunt is a decent middle ground — better calibrated than the Amazon stuff out of the box.

Dizzy83
Dizzy83
Active Member
13 posts
Joined Aug 2024
6 days ago
#16243

Hey @BenWebb, one thing worth trying that nobody's mentioned yet — make sure you're doing a proper full charge reset regularly to resync the shunt. LiFePO4 needs to hit that absorption stage properly to give the monitor a solid 100% anchor point, otherwise the errors compound over every cycle. Also check your shunt's actual resistance value with a decent multimeter if you can — plenty of the cheap ones are labelled as 100A/75mV but the resistance is slightly out, so you can manually adjust the capacity figure in the monitor settings to compensate. It's a bit trial and error but gets you surprisingly close. I've got a Victron SmartShunt now and the difference is night and day, but I ran a cheap Amazon one for about eight months with those tweaks and it was at least usable. 👍

Foggy91
Foggy91
Member
6 posts
Joined Mar 2025
4 days ago
#16485

Good points from everyone above. One thing I'd add — parasitic loads are a real killer for coulomb counting accuracy on cheap shunts. Even tiny draws like a battery management system, a voltage display, or anything left on standby that isn't routed through the shunt will cause drift over time. Worth double-checking absolutely everything on the negative side passes through it.

Also, LiFePO4 has pretty poor charge efficiency at the top end (absorption/CV phase) and cheap shunts rarely account for this properly — they'll assume 100% in equals 100% out. A decent unit lets you set a Peukert exponent and charge efficiency factor. At £12 I'd wager yours doesn't. Might be worth looking at a Victron SmartShunt — yes it's more money, but the Bluetooth app and proper LiFePO4 profiles make a massive difference for van life reliability.

Log in to join the discussion.

Log In to Reply