Antigravity LFP Start Batteries... "Long-Term" Test (TL;DR: Cannot Recommend)

Just for a little background, I'm the head tech for a distributor of high-current AC and DC power generation/modification/distribution components, including LFP (lithium iron phosphate or LiFePO4) batteries and the like, and although neither of my degrees are in electrical engineering I do have an extremely extensive engineering background both from my decade in the USAF as well as later private and professional experience -not to mention the last 5 years of testing/diagnosing/repairing batteries, inverters, chargers, and all sorts of things for this job every day- so I'm not just talking out my ass here.

About a year ago I purchased (not at retail, good lord, but my commercial price was still about $600) an Antigravity LFP start battery for my '15 VT; I purchased the correct model for this vehicle in the highest-capacity version available, the AG-H5-40-RS (Group 47/H5 form factor, 40Ah capacity). I purchased it largely for the increased usable capacity; the OEM AGM battery is 60Ah, of which as with most Pb batteries approximately 50% is usable before voltage drop causes issues for sensitive electronics and further discharge creates the potential for permanent damage to the battery, while LFP is capable of 100% discharge with a very flat discharge curve and no detrimental effect to the cells. Secondarily, I purchased it for the weight savings - the OEM AGM battery weighed 32.9lbs while the Antigravity weighed 11.9lbs... big and noticeable difference.

After install of the AG-H5-40-RS, I will say the drop in weight was noticeable in the behavior of the suspension and general responsiveness of the car, and for the first week or so the battery seemed to be performing as expected - engine starting with the AG's 1500CA was so instantaneous that it felt like it started before I'd even fully pressed the button, and my charging voltages and running behaviors were staying as expected - I watch ECU voltage on my UltraGauge, system voltage through a monitor on one of my DC power ports, and actual battery voltage from a bluetooth voltage/temperature sensor mounted to the battery itself. After about two weeks, however, I noticed that all voltages were dropping lower than they should while driving - I'm not talking the normal 12.5-12.8 range that the alternator will back off to, I mean 12.1-12.4V. Very unusual, particularly given that the normal resting voltage of an LFP battery is around 13.2-13.6 depending on the BMS and in the case of the AG, the resting voltage should be around 13.2.

I pulled the battery out of the car and got it on my test bench and after running multiple charge and discharge tests on the unit, I established that not only was the BMS glitching out and very rarely allowing the battery to actually charge at all, but the battery also had the entirely wrong cells in it, with only a total capacity of 14Ah, not 40Ah as it was labeled, boxed, and sold. I was, obviously, peeved.



I was of course immediately in touch with AG, going from my sales rep to their head tech to eventually the owner Scott, and we went back and forth for about a week with me sending them test results, photos of the install, specs of the alternator, photos of the OEM wiring and ground points and lots of other supporting evidence and they eventually agreed to warranty the unit and concluded that this particular battery must have been mis-labeled at the factory in China. The complication was that that was the very last Group 47/H5 they had available for the foreseeable future (this was April of 2022) so they suggested I find a different model that might work. I measured everything and compared to the dimensions listed on their various models on their site and eventually settled on a Group 75/78 form factor, as although it had slightly different dimensions I established that I could make it work.

So, around the end of April 2022 I received my replacement battery, an AG-75-40-RS - same Ah capacity as my first unit, and still a proper drop-in starting battery specifically designed for this purpose, just with slightly different case dimensions. This time I immediately ran charge/discharge tests on the unit and found that it did have the advertised capacity (actually slightly better, 42Ah at the 20-hour rate) and all behavior was within spec, so I happily installed it and, up until 3 days ago, had absolutely zero issues with it. Worked great, did what I needed, ran my high-demand components for as long as needed, etc. etc. Super happy.

Howthefuckever.

3 days ago on my drive in to work, for some unknown reason the battery's BMS decided to shut off the battery while my alternator was under load (HIDs) which caused the alternator output voltage to spike and my ECU momentarily saw 16.2V. It was brief; my UltraGauge is maxed out at 10Hz polling rate, so the spike lasted about 2/10 of a second before the alternator throttled back, but still not good although no damage was done. I pulled the battery as soon as I got to work, tested it thoroughly on charge and discharge, everything performed as it should. No indication of any fault in it whatsoever, so honestly I thought maybe the UltraGauge just glitched and read the wrong values for split second... Okay. Shit happens. Reinstalled the battery. But that night on the way home it happened again twice. I was watching the UG like a hawk this time, and the battery's BMS definitely shut off the battery, alternator output voltage spiked, ECU throttled back the alternator voltage, BMS turned the battery back on, then turned it off again and voltage spiked again. Only made it to 15.9V one time and 16.1V the second time, not quite the 16.2 I'd seen before, but this time the spikes burned out my battery-mounted bluetooth voltage/temp sensor which, though I know the ECU capacitors are likely a little more forgiving, still made me seriously worry about my ECU. Before the sensor burned out I did confirm that there was no valid reason for the BMS to have shut down the battery - battery temperature was 93F, well under its high-temperature cutout, and voltage at the battery was a respectable 14.1V, well under its high-voltage cutout. No, it just... freaked out... for no external reason.

Needless to say, I pulled the Antigravity out of the car and now have an AGM back in there; better to be heavy and be able to trust my battery than be light and wondering. And quite frankly, though the car definitely felt perkier without the weight, I'll say that having that weight back over the driver side wheel actually makes the front end behavior feel a little more solid and balanced, so eh... not a total loss!

At the end of the day, though, that's 0 for 2 in less than 1 year with Antigravity batteries... and worth noting, too, that these batteries were being used for the purpose for which they were designed. I know other people in many different car communities have had issues with AG batteries, but in many cases they were using the wrong battery type (ie, motorcycle battery rather than automotive starting battery)/connected incorrectly/things like that which are actually quite critically important, so there was a bit of a "yeah, but..." discussion to be had in many cases. In this case, nope, these were the proper batteries for the application, just an unfortunately poor quality product. Great idea. Really really fantastic idea. Poor execution.

The warranty, though, you might ask, what about the warranty? Hmmm yes, except that when you read the fine print, the warranty is good for one exchange when a battery is deemed defective. The replacement battery does not have a warranty. Since my first battery was exchanged under warranty, this latest one is not eligible for an exchange... and to be honest I wouldn't put one back in even if they were going to warranty it, because I don't trust the damn things anymore and I'm not going to risk blowing a capacitor on my ECU just to save weight. The upside, however, is that now I get to take this one apart and see how it's built and possibly establish what exactly failed, so I will add photos to this thread in the future. AG claims that they use prismatic pouch cells in these batteries and a BMS from a "reputable" supplier, but after my experience I definitely have my doubts... we'll soon see!