Balmar smart gauge (SOC) problem

[HerbertFriedman]

I installed a Balmar Smart Gauge a few years ago at the recommendation of the Maine Sail review. That gauge uses a proprietary set of algorithms supposedly developed by the UK Air Force which measures the Battery State of Charge by just monitoring the voltage profile during charge and discharge cycles. The more conventional approach (Victron) measures the AmpHours discharged by the battery by monitoring the current in a ground shunt. Apparently, while this method accurately measures the charge delivered by the battery, the SOC as a percent of the batteries capacity is in doubt in older batteries because the SOC when the battery is fully charged, can degrade with age, hence the SOC as a percent is in doubt. According to Maine Sail and Balmar their algorithms avoid this problem.

My problem is that with a normal set of instruments (depth, speed, chart plotter and VHF, the fridge and the auto pilot being turned off), the Balmar SOC drops from 98-100% (after days worth of shore power charging) to about 61% in the course of a 3 hour day sail. So going a short distance to an anchorage for the night means I have to run the engine to charge the batteries back up.

The house battery bank is two G27 Trojan wet cells in parallel, each one listed at 105 Amp hours for a total of 210 Amp hours, not insignificant for my E34. Both batteries are only one year old and the water level looks OK. So a difference of 98% - 61% is a charge loss of 29% which for the 210 Amp hour bank represents a charge loss of 61 Amp hours in 3 hours or an average current draw of about 20 amps. That current draw sounds quite high for the instruments and indeed, a clamp on ammeter around the current lead of the bank reads only about 5 amps. Also the battery voltage at the end measured 12.22 V, sounds about right for a fully charged battery.

Any ideas out there? I will call Balmar but when I called them a few years ago about an error message, E10, the tech said he never heard of such an error message, so I am skeptical.

 

[Loren Beach]

Ours still works fine, but I am not smart enough to know much about it.
Here is our install story- https://ericsonyachts.org/ie/ubs/2016-new-batteries-new-smartgauge.485/
You might try calling or emailing again. They may be short of staff with the pandemic and all.

 

[Jerry VB]

The house battery bank is two G27 Trojan wet cells in parallel, each one listed at 105 Amp hours for a total of 210 Amp hours, not insignificant for my E34. Both batteries are only one year old and the water level looks OK. So a difference of 98% - 61% is a charge loss of 29% which for the 210 Amp hour bank represents a charge loss of 61 Amp hours in 3 hours or an average current draw of about 20 amps. That current draw sounds quite high for the instruments and indeed, a clamp on ammeter around the current lead of the bank reads only about 5 amps. Also the battery voltage at the end measured 12.22 V, sounds about right for a fully charged battery.

The Balmar Smart Gauge clearly is not displaying a correct "state of charge." The question is why.

The first suspect in any "smart" anything is configuration and the second suspect is calibration. I presume you have the gauge configured for the proper battery type. The manual (pdf) recommends "Leave Smartgauge™ working for 48 hours. Use the battery system as usual, Smartgauge will automatically catch up over the next 2-3 charge and discharge cycles of the battery bank. Unlike all other battery state of charge meters currently available, Smartgauge™ becomes more accurate the longer it is used. All other battery state of charge meters become less accurate the longer they are used and require multiple recalibrations."

If you are never putting any significant (20% or more) discharge/recharge cycles on the battery, the "smart" part of the gauge may get confused because it doesn't have any data to use to recalibrate it's estimate of the battery's state of charge. Per your measurements, you are pulling ~5A over 3 hours = 15AH out of a 210 AH battery bank so your cycling is only about 7% of the battery's nominal capacity. The specification on the gauge is 10% of the state of charge (charge) and 5% of the state of charge (discharge), so the level of discharge that you are cycling the battery at is at about the same level as the accuracy of the gauge itself. I would try a couple of deeper discharges and see if the gauge "learns" better and gives you more realistic SoC values.

The next suspect in any battery problem is corrosion built up on the battery posts and/or the terminals. Corrosion causes extra voltage drops that the gauge will interpret as heavy (high amperage) discharge even though the actual current drawn is low (voltage drop = current x resistance and corrosion increases resistance significantly).

If you get to this point, I would do a "factory reset" (manual p.19) to start over with a "clean slate."

If the terminals are clean and tight and a factory reset doesn't fix it, the gauge might be defective.

Marine Sail has a very positive review of the Balmar Smart Gauge. I was skeptical of how accurate it could be when measuring only the battery voltage, but Rod did a very thorough test and was happy with how well it worked.

 

[HerbertFriedman]

The Balmar Smart Gauge clearly is not displaying a correct "state of charge." The question is why.

The first suspect in any "smart" anything is configuration and the second suspect is calibration. I presume you have the gauge configured for the proper battery type. The manual (pdf) recommends "Leave Smartgauge™ working for 48 hours. Use the battery system as usual, Smartgauge will automatically catch up over the next 2-3 charge and discharge cycles of the battery bank. Unlike all other battery state of charge meters currently available, Smartgauge™ becomes more accurate the longer it is used. All other battery state of charge meters become less accurate the longer they are used and require multiple recalibrations."

If you are never putting any significant (20% or more) discharge/recharge cycles on the battery, the "smart" part of the gauge may get confused because it doesn't have any data to use to recalibrate it's estimate of the battery's state of charge. Per your measurements, you are pulling ~5A over 3 hours = 15AH out of a 210 AH battery bank so your cycling is only about 7% of the battery's nominal capacity. The specification on the gauge is 10% of the state of charge (charge) and 5% of the state of charge (discharge), so the level of discharge that you are cycling the battery at is at about the same level as the accuracy of the gauge itself. I would try a couple of deeper discharges and see if the gauge "learns" better and gives you more realistic SoC values.

The next suspect in any battery problem is corrosion built up on the battery posts and/or the terminals. Corrosion causes extra voltage drops that the gauge will interpret as heavy (high amperage) discharge even though the actual current drawn is low (voltage drop = current x resistance and corrosion increases resistance significantly).

If you get to this point, I would do a "factory reset" (manual p.19) to start over with a "clean slate."

If the terminals are clean and tight and a factory reset doesn't fix it, the gauge might be defective.

Marine Sail has a very positive review of the Balmar Smart Gauge. I was skeptical of how accurate it could be when measuring only the battery voltage, but Rod did a very thorough test and was happy with how well it worked.

The battery terminals are very clean so I do not think that is the problem but your suggestion to reset the gauge followed by a deep discharge to "teach" the gauge sounds reasonable. I wonder if others with this gauge has tried to correlate the gauge with a direct ammeter-time reading, subject of course, to the uncertainty of what the actual SOC for a "fully charged" battery is. That uncertainty is exactly the reason the Victron method is not always accurate, especially in older batteries.

A second point is that now that I have some data on amperage draw from a set of instruments normally used in sailing, if one were to go cruising and sail for say 24 hours at a time, that would be a draw down of 5 amps x 24 hours = 120 amp hours which is at least half of the capacity of normal house banks. That would require recharging every day, wonder what cruisers do?

 

[Kenneth K]

12.2 volts would indicate about a 40% discharge for deep cycle batteries so the Balmar was consistent showing 61% SOC.



If your bank discharged 40% after 3 hours, then (absent gauge error/malfunction) you were either draining more than 5A, you started with less than 210A-h available(a batt/bank problem) or some combination of both.

I'd probably check the individual current draw of each item with a VOM meter to further verify what you measured with the clamp-on gauge.

 

[Loren Beach]

Probably just me liking the "belt and suspenders" approach, but I maintained the expanded scale VM that Ericson installed on the breaker panel.
Even replaced the old meter with an identical new one a few years ago when it finally stopped indicating.
As much as I rely on the sophisticated new device, I also appreciate the reliable analog readout as a confirming information source.

 

[Jerry VB]

@Kenneth K's comment made me notice a math error in the percentage of discharge:

So a difference of 98% - 61% is a charge loss of 29% 37% which for the 210 Amp hour bank represents a charge loss of 61 82 Amp hours in 3 hours or an average current draw of about 20 27 amps.

That level of discharge (27 amps) seems unlikely, especially if your clamp-on meter is reporting ~5A. If true, something is getting pretty warm (325 watts x 3 hours). I'm suspicious of your Balmar Smart Gauge's readings. When you reported 12.22v, I presume that was that reported by your Smart Gauge? I would use a DMM to get a second opinion.

Update: one thing that could be drawing 20+ amps without you noticing it (smelling it) are your engine glow plugs. If your glow plug switch is sticking, that would explain the current draw.

 

[HerbertFriedman]

thanks for all the advice. I will repeat the test but this time I will measure the battery voltage (using both the Balmar gauge and a digital VOM) at the start (assuming the Balmar SOC reads well into the 90's per cent) and after. Ken, thanks for the voltage chart but I am somewhat skeptical since I was under the impression that voltage alone is not a good indicator of SOC, with the caveat that a reading under about 11 volts certainly indicates a low SOC. But recording the voltages and comparing the Balmar voltage reading with a digital VOM is certainly worth a try. i will let you know.