Without getting too technical, the smaller battery testers are also "real" and unlike the high current DC loading testers (the so called carbon-pile which the cheap DC testers are made with, it falls under the resistive loading category), are based on injecting an AC constant-current thru the battery and measure its impedance, which in theory it should reflect the internal resistance of the battery from which we could calculate the potential current capability, i.e at 12V a 1,000A battery would have 12milliOhms resistance, however a good test instrument would be also in the thousand$ range (some are over $10k), has a four wire measurement (two for current delivery and two for voltage measurements) and the cheap ones like in the picture above they are maybe in the 15-20% accuracy range, therefore a 1,000A battery could be off by as much as 150A-200A.
Also, in order to use the AC type tester, the battery SoC (State-of-Charge) should be at least in the 70-75% range, so when dealing with an unknown battery that would be another variable.
The high current DC loading testers also stress all the connections thermally to expose any potential hot spots.
The HMMWVs have that 1,000A-100mV shunt, which could be used to get a good idea about how much current the batteries are delivering at start, by connecting a DMM across it, in the 100mV range or next closest range above it (ideally a DMM with Peak/ Max function would work best as it would record and display the highest value, otherwise difficult to observe as the numbers are changing quickly during starting, plus the DMMs with Peak function have a faster acquisition time which can also capture the highest peak more accurately, where the lower integration time/ slower acquisition DMMs would show a lower Peak value as it takes longer to integrate the measurement that changes quickly).
Generally, the Peak functions caption sub-millisecond events in the hundreds of microseconds, while the Min/Max function alone runs at the sampling speed of the DMM, which is usually in the 3Hz range or about 3 times per second at 333mS intervals, which it could still be useful.
Another DMM with the Peak function (or use the same one and repeat the staring test, it would be close enough), but set for the Min measurement could be set to measure the battery voltage, which would record the lowest value it dropped to during starting.
If the 24V battery bank drops too much, then each battery should be measured separately to determine which one is the weakest link, since they are in series.
Just a few other ways to test batteries.
EDIT: Here are two DMMs that could do MIN/MAX, Fluke 77 IV and Klein MM420, which is in the $40-$50 range:
Find many great new & used options and get the best deals for Klein Tools MM420 Digital Multimeter at the best online prices at eBay! Free shipping for many products!
www.ebay.com