Re: Battery Beak
 

If your team has a "battery wisperer" (somone who knows and understands batteries) that tracks and manages your batteries, they would probably
give it at least a 9. It takes two measurements at different currents
(small and large, the small (1A) is like the current of the cRIO alone, the larger current is like one window motor (18A) in less than a second, and displays both voltages and currents for you to review. You can tell how the battery will support a load even though it looks fine with small or no load.

If your team doesn't have a "battery wisperer", your electrical team would probably still give it at least a 9. The quickness of the test, and the simplified classification of Charge, Bad, Fair, or Good will help your team screen out potentially bad batteries. The multi-chemistry feature lets you test the NiMH batteries from those smaller 'bots, although you have to go into a setup menu to switch back and forth between battery types.

Does you team absolutely need it to participate? Not if you have a good "battery wisperer" on the team.

Will it help your team compete with a mix of older and newer batteries? Yes, I would buy one as soon as I could purchase it. It quickly provides real information about the battery's health. The simple classification and state of charge display may flag a marginal battery as bad, but the measurements are displayed so you can make your own judgement.

Very cool product, a tad expensive, but very quick and convienient, based on fundamental voltage and current measurements.

So far, very happy with it.

Re: Battery Beak
 

Ummm "battery whisperer", an interesting title.

I have both the Battery Beak and the CBAII load tester. I've done tests on the VEX battery packs using both devices. The Beak is good at pulling out a quick snapshot of the existing battery status. The CBAII is good at looking at the long term load. I have found a battery that has a cell that will fail under load. The Beak does not show it (and I would not expect it to), where the CBAII picks it up about 30 seconds into the test.

I'll work with the FRC team to get them to charge all their battery packs and run a dual Beak/CBAII run to rate the battery sets into practice / competition battery sets.

It would be nice to have the CBAIII, since it can put a heavier load on the battery packs. The prior year swerve base with the 4 active CIM's can draw more than the CBAII can simulate.

Re: Battery Beak
 

Would you be willing to post a plot of the CBAII data for the above battery?

Does it look something like the link in this post?

Re: Battery Beak
 

Team 342 is already using this. It makes things much easier and since you can let it hang around your neck it also makes it faster. we love it

Re: Battery Beak
 

Pretty much like that, swap 8V for the peak, 30 seconds for the first drop point and you've got it. There is only one dead cell, so only I have only the first drop. Mine then shows a flat line of power (good) after that. The battery at the link shows two bad cells (the second voltage drop)

I'm going to say that the battery would also show as good by the Beak. This battery would cause the classic "The robot works for the autonomous then stops running, must be a (software, mechanical, field control) problem."

Re: Battery Beak
 

Very interesting. So one (or two) cells have only 3 amp-hour capacity,
and those weak links cause the battery voltage to drop when they are
discharged. The Beak does not significantly discharge the battery,
but if those cells have a high internal resistance, they would show up.
But if they have low internal resistance (like the rest of the cells), just
reduced capacity, they will not show up in a Beak test. It would be
interesting to see what the Beak reports for Rint on the known bad
and a known good battery.

If you have access to a 50A or 100A car battery load tester, I would
be interested to see if the subject battery tests "bad" compared to
a known good one. (Normally test is something like apply a 50A
or 100A load for ~10secs, read the voltage before removing the load. The load tester will get hot, don't leave it connected more than 15sec)
Terminal voltage under such a heavy load should be >10.5v, and >11v
is pretty good. If the voltage is not slowly dropping (unsteady, jumping, or
quick drops), then individual cells are not contributing properly. I bet
the bad battery charges real quick - you only get to put in 3-4 amp-hours
before the terminal voltage rises, and the battery appears full. Hence you
only get 3-4 amp-hours out, more load is less run time.

I also have a CBA-II, and capacity test batteries when we get new ones, both to see if the tested amp-hours matches spec, and the shape of the
discharge curve. It takes quite a while (>3hours at 6A load), and you really
want to recharge the battery imediately after the test - don't leave it in a discharged state, that limits our ability to test during working time.
The Beak is quick, but dosen't do a discharge test. Both have a place
in our shop.

Re: Battery Beak
 

Foster and I have talked about the 2 units. I suspect that to really know a battery a fixed load like 50 amps for at least a minute will tell the true story. That is a hard test to make an instrument for. We have many batteries in varying condition and age. Since Foster has both units, we will go thru all our batteries and that should tell allot about the value of both devices. Good project for in a couple weeks when students are sick of drilling holes and machining.

Re: Battery Beak
 

Agree - best test is a real 2 minute match at 100A ::ouch:: !

When you run a CBA load test, please use a reasonable cut-off voltage to end the test - I usually test at 6A (which is 18AH/3, or C/3 in battery speak) to a end of test voltage of 10.3v (look at the discharge curves in the battery spec sheet, there should be at least C/5 (~4A), C/2 (9A), and C (18A) curves, and they stop at different voltages, just interpolate to get voltage for C/3). There is no value to running the battery down to 6 or 7 volts - you can't compare to manufacturers specs, and the deep discharge may
actualy damage the battery.

We do the same - look for a time when students are looking for a change of pace.

Re: Battery Beak
 

I know this JUST came out, but are there plans for an in-series version to measure a battery under a true load while connected to the robot?

Just a thought I had (all while COMPLETLY ignoring the fact that the HUD on the classmate can show the same thing (I believe?) & that the old IFI controllers did the same thing as well somehow). lol

Re: Battery Beak
 

Marvin et al.
West Mountain does make an higher current version of the CBA II. When we run our tests, I set the CBA for the highest current it can test at, about 7.5 amps as I remember. We run the test to match manufacturer specs which I believe is 8 volts terminal voltage to duplicate the manufacturer's amp hour test. In the case of two damaged cells, a 10 volt cutoff might not show the second cell in Hugh' test.
The Battery Beak cannot test the depleted cells shown in Hugh's linked graph because the battery is good until the drop in cell voltage. It would likely show a different internal resistance on the Beak. How different only time and additional testing will show. Mike may have already done this testing. I believe the change we are seeing is due to damaged plates within the cell. Either a partial beak within a plate or an entire plate breaking off the plate connector. Any battery I have found with these deficiencies has always had some evidence of a drop when looking closely at the outside of the battery case.

Re: Battery Beak
 

You are correct - the discharge cutoff voltage depends on the discharge rate. Discharge at 7.5A would be 7.5A/18AH = 0.417C. The best datasheet I ran across tonight was from YUASA http://www.farnell.com/datasheets/575631.pdf, the discharge curve for the NP18-12 shows a cutoff voltage of 10v for 0.4C, 9.5v for 0.6C, so I would cut off testing at about 9.9 volts. You would measure only 12.4AH for a good, "nominal" 18AH battery (see the General Specs on the right side of page 1, the discharge would be to 1.65 volts per cell) [18AH is at the C/20 rate] The difference in measured capacity is due to power lost in the internal resistance of the battery, and reduced conversion efficiency at higher discharge rates (google "Peukert exponent").

I agree - the Beak does a very short time load test, if that's not enough to discharge the weakest cell, it won't show up as a voltage drop. But the battery might still show a higher internal resistance, since the other cells are probably stressed by overcharge/discharge as the weak cell drops out and charges prematurely.

Dropping batteries should be avoided for a lot of reasons, we don't usually think about internal damage, but plates and internal connecting bars can break inside the case. I'll be sure to check the battery case on any "problem" batteries from now on!

Thanks!

Re: Battery Beak
 

Marvin,
As you may note on the discharge graphs, the curves imply that they can go down to 8 volts so that is the one I use to insure as close to printed data as possible. If the total battery runs a curve as shown for 0.4 C discharge or very close we will consider it good. By choosing 8 as a cutoff I can prevent total discharge yet still make a reasonable curve. If it doesn't match, we will let it rest, then recharge and then test again. Two failures are enough to add it to the practice pile. A spectacular will get it a free trip to the recycle pile by the entrance.