When Balmar introduced the UK developed Smart Gauge to the US I was pretty excited. I had tried to buy one two years ago but my emails went unanswered so I filed it under the scam/snake oil folder in my mental filing cabinet.


I had literally forgotten about the Smart Gauge until Rick J. of Balmar approached me at the Annapolis show to tell me they had added it to their product line and were now the US distributors. Even though I implicitly trust Rick and the guys at Balmar, the claims, still seemed too good to be true. Why had no one been able to do this before??


To make a long story short I tracked down an inside contact at EnerSys (name withheld) to see if I could get my hands on the white paper so often referenced by Smart Gauge.


EnerSys are the makers/inventors of TPPL AGM technology sold under the Odyssey & Die Hard brand for marine use and they are the inventors of the Optima spiral wound batteries (which has been sold off). EnerSys is much larger than their presence in the marine market and much of their business is in large UPS systems and military use. Because EnerSys has no financial ties to Smart Gauge I found their white paper to be a breath of fresh air in a credible independent test data manner. I don't believe EnerSys allows Smart Gauge to use that white paper, and I was asked not to reproduce it, so have only taken excerpts from it...


But RC how does it work?


You've got me..? I have no idea how it works, at least at the programing/algorithm level (proprietary stuff), but it is reportedly designed to track voltage. Many internet posters have assumed, posited and suggested, that it checks internal resistance and pulses across the battery etc.. It may, but I have not seen evidence of this on the power / volt sensing wires, even with an Oscilloscope. If I had a tracking o-scope I may have seen something but I don't.


As near as I can tell it simply tracks voltage, up to 1500 times per second, to detect trends and compare it to internally programed models. I do know it uses computer modeling of actual batteries in real world situations and then feeds this data into an algorithm that can "learn" the bank as time goes on. This modeling must have been time intensive to get to this level of accuracy but this sort of programing level is way to complex for a guy like me who can't even figure out gmail..


It may pulse the battery every so often and I just missed it..? All I can say is that over time it seems to adapt to learn your bank and give significantly more accurate SOC readings than an Ah/Coulomb counter can, as programmed and used by the average installer / boater..


How it actually does this is as closely a guarded a secret as the Frosted Flakes recipe that Tony the Tiger protects.


I think the best overview I can give, on its level of accuracy, is the executive summary of findings by EnerSys.


QUOTE = EnerSys White Paper


"EXECUTIVE SUMMARY

SmartGauge® is a Battery Monitoring Unit (BMU) that is intended to be fitted within military vehicles and to provide crucial information to vehicle commanders, such as State of Health (SoH), State of Charge (SoC) and the time remaining they have available to continue operation until battery power runs out.


The working partnership between EnerSys and SmartGauge® has resulted in EnerSys testing the SmartGauge® BMU whilst connected to a Thin Plate Pure Lead (TPPL) battery type within its electrical laboratory at Newport South Wales, to evaluate the performance and accuracy of its data.


The SmartGauge® BMU was tested using a 12V 100Ah TPPL battery which was subjected to a 100% depth of discharge test, followed by a full 12 hour recharge. This cycle was repeated continuously until the battery reached 80% of its rated capacity, the SmartGauge® BMU and EnerSys laboratory data logging equipment (Digatron) continuously monitored the voltage, current, time and from which the State of Charge and State of Health was calculated.


The correlation of State of Charge (SoC) and State of Health (SoH) between the BMU data and Digatron Data was excellent with insignificant variance between the two readings based upon resolution increments of 1%."