Batteries

[ChuckDickerson]

Our team has been using the standard Enersys NP18-12 batteries for the last several seasons. We typically buy a new set of competition batteries every year or 2 and relegate the old batteries for testing and practice. With the expansion of the list of legal batteries this past year we are looking for feedback from teams that have experience with batteries other than the Enersys NP18-12. We had a battery start leaking this weekend around the top seal and are looking to replace batteries this coming season and would like to know which batteries are “top quality” and which have a reputation for not holding up. Our experience with the Enersys batteries has been hit or miss. Some seem much better than others. For reference we have a nice battery testing setup including a West Mountain Radio CBA with dual 500W amplifiers. We are building new battery charging station(s) this fall around Autometer BusPro-660 chargers and are looking to replace our competition batteries with “top quality” batteries. Now that we have a competition season behind us does anyone have any advice on the “best” batteries to go with based on the 2014 legal battery list*?:

Enersys (P/N: NP18-12)
MK Battery (P/N: ES17-12)
Battery Mart (P/N: SLA-12V18)
Sigma (P/N: SP12-18)
Universal Battery (P/N: UB12180)
Power Patrol (P/N: SLA1116)
Werker Battery (P/N: WKA12-18NB)
Power Sonic (P/N: PS-12180 NB)
Yuasa (P/N: NP18-12B)
Panasonic (P/N: LC-RD-1217)
Interstate Batteries (P/N: BSL1116)
Enersys (P/N: NP18-12BFR)
Enersys (P/N: NP18-12B)

*Yes, we know what was legal last season may not be legal next season.

[AllenGregoryIV]

Hey Chuck, I have been very happy with the MK batteries. If you call them and tell them you are a FIRST team they give you a pretty good deal. We have paid about $40 each including shipping. We normally order 6 during each build season.

They're customer service has been great. We had an event we needed the batteries for and they were able to work it out with the UPS driver to deliver the batteries to me in a parking lot after they got picked up from their distribution center in Houston. It was pretty crazy.

[pwnageNick]

They're customer service has been great. We had an event we needed the batteries for and they were able to work it out with the UPS driver to deliver the batteries to me in a parking lot after they got picked up from their distribution center in Houston. It was pretty crazy.

Did you use the batteries that weekend? Did you cycle the new batteries a few times before using them in competition or did you use them fresh?

-Nick

[Jon Stratis]

I know it doesn't exactly answer your question, but check out your local Batteries Plus. They usually don't have a lot of variety, but they'll likely have something that's on the approved list. Also, if you register with them as a corporate account (all it requires is a phone number, and they've never called me) you get a rather large discount... I want to say it was almost 3 for the price of two last time I went! At any rate, that should save a little bit on shipping!

[lpickett]

I used YUASA or Power Source at work for emergency lights, emergency notification systems and fire systems. They would perform better than some other brands we used. I never seemed to have any issues. Even under extreme loads, they did not have to be replaced as often as I thought.

[ChuckDickerson]

Thanks for the info so far guys. For clarification by "best" batteries I'm not necessarily looking for cheapest. As with anything, I know some batteries are better quality than others but I don't really have enough experience with different battery brands to know which are "high quality" and which to avoid. My philosophy is EVERYTHING on the robot starts with the battery so why risk using anything less than the best quality. We all pay a lot of money per match to play in FRC and I have seen may teams loose many matches because of battery issues over the years. Why risk it with el cheapo batteries? I just don't know which on the list have the best reputation for quality.

We have always gotten our Enersys batteries from Portable Power Systems: http://www.portablepower.com/12_volt_10_19/ENERSYS_Genesis_NP18-12B

I'm not saying at all that we are unhappy with the Enersys batteries and in the absence of info as to a better option I would likely buy more Enersys from PPS tomorrow if I had to. I'm just wanting more info from the collective FIRST community to make a more informed purchasing decision between now and whenever the time comes. I figure everyone would like the benefit of knowing what batteries seem good and what seem not so great. With a few thousand teams using probably tens of thousands of different batteries now there has to be some good and bad recommendations especially given how opinionated folks on CD can get. ;)

So, I have one vote for MK from Allen so far...

[AllenGregoryIV]

Did you use the batteries that weekend? Did you cycle the new batteries a few times before using them in competition or did you use them fresh?

-Nick

We used them right away and I don't remember having any problems.

We have been testing batteries using a CBA III this summer and plan to continue during the year.

If anyone is interested here are the first test results, these aren't that useful until we start seeing what happens to them over time. We also didn't run baseline tests on them when they were new either.

https://docs.google.com/spreadsheets/d/1MJmf66L0cBPzfb99ys-usODANxhVRt1ovwg0_mqTdVg/edit

These are tested at a 7 AMP drain and a cut off voltage of 10.5 Volts.

Here is a sample graph from one of the the battery test.
https://docs.google.com/file/d/0B0dr40Yt8DLZR0FXWDVYNF94THc/edit

[Al Skierkiewicz]

For those of you that are using the CBA, I recommend that you test at 7-10 amps and 8 volt cutoff. This will give you curves that match the manufacturer curves for most batteries. The CBA has a top current draw of about 7.5 amps, the CBAIV will do a little more. It helps to test batteries once or twice a year and use the overlay feature to see the difference in tests.
BTW, let West Mountain know you are a FIRST team, I have been trying to get them to be sponsors for a couple of years. They are located in a Milwaukee suburb.

[Ether]

we have a nice battery testing setup including a West Mountain Radio CBA with dual 500W amplifiers

What sort of tests were you planning to do with those amplifiers?

[ChuckDickerson]

What sort of tests were you planning to do with those amplifiers?




The CBAIII we have can only sink ~7.5A @ 12VDC. We have/do run tests at 7.5A like Allen and Al suggest above but also run tests at higher current draw (50A/75A) using the dual amplifiers. FRC bots pull a lot more current for a short 2 minute match than the 7.5A tests simulate. We like to stress test our batteries under conditions that more simulate FRC bots. We have identified under-performing batteries this way.

[Ether]

We ... also run tests at higher current draw (50A/75A) using the dual amplifiers... We have identified under-performing batteries this way.

What I understand you to be saying (please correct if this is not the intending meaning) is that you have identified under-performing batteries which would not have been identified with the 7.5A test alone. Is that right?

[ChuckDickerson]

What I understand you to be saying (please correct if this is not the intending meaning) is that you have identified under-performing batteries which would not have been identified with the 7.5A test alone. Is that right?




I suppose it would be unfair to say that they wouldn't eventually be found to be under-performing with a 7.5A test. We have had a few batteries that seemed OK testing at a lower current draw (~7.5A) that dropped off rapidly at a higher current draw (~50A). I will try to pull up some CBA data/plots this evening and compare.

I'll have to look again at the battery test files but I don't think I usually test them to as low a cutoff voltage as Allen (10.5V) and AL (8V) suggest. IIRC I usually set a cutoff voltage of 11.5V.

[AllenGregoryIV]

I'll have to look again at the battery test files but I don't think I usually test them to as low a cutoff voltage as Allen (10.5V) and AL (8V) suggest. IIRC I usually set a cutoff voltage of 11.5V.

I'm interested to know why Al tests all the way down to 8V. We went with 10.5V as that's about as bad as we have ever had a battery come off the field. We don't compete with voltages lower than that (no load voltage) so we didn't feel it necessary to drop them that low. Maybe we should be lowering the cut off voltage.

[nighterfighter]

I'm interested to know why Al tests all the way down to 8V. We went with 10.5V as that's about as bad as we have ever had a battery come off the field. We don't compete with voltages lower than that (no load voltage) so we didn't feel it necessary to drop them that low. Maybe we should be lowering the cut off voltage.

In 2009, our battery voltage would drop down to 7V after a heavy match, and under load, during the match, would display around 6V. We came very close to a brownout.

Of course, we were using 9 or 10 motors that year, and they were always on 85% of the time.

But its definitely possible to have the robot operate at dangerously low voltages, so it might be relevant to test the battery down to low voltages.

[ChuckDickerson]

I'm interested to know why Al tests all the way down to 8V. We went with 10.5V as that's about as bad as we have ever had a battery come off the field. We don't compete with voltages lower than that (no load voltage) so we didn't feel it necessary to drop them that low. Maybe we should be lowering the cut off voltage.

Basically the same philosophy with us. I think the 50A test cutoff at 11.5V usually take 6 or 8 minutes or something like that. Again, I'm pulling from memory right now so let me look at the data and verify but the tests go much faster and, well, you know how many batteries we usually bring to competition so it really speeds the testing process up as well. I know there will be comments about this process being harsh on the batteries and that manufacturers test at a lot lower current draw for a lot longer period of time but we aren't using these batteries like that so my philosophy is why test them like that? We stress a battery for 2+ minutes through a 120A breaker. 50A or even 75A draw for a few minutes seems reasonable to me. I actually wish I could figure out how to program the CBA & amplifiers to cycle the current draw to like 100A or 150A for a short time (seconds) then draw less and sort of simulate a 2-3 minute match. That would seem to be a better indicator of how well the batteries will hold up the way we use them.

For what it's worth we used to always charge our batteries using the 2A setting on our chargers. A couple of years ago (through our battery testing) we found the batteries seem to do better charging them using the 6A setting. We just got some of the Autometer BusPro-660 chargers which are 6 bank 5A chargers designed for SLA. I hope they work out well. I guess we will see this coming season.

All this battery testing discussion is great and useful but I really hope the collective community can provide some good insight into what their experiences were this year (good or bad) using battery brands other than the Enersys.

[ChuckDickerson]

In 2009, our battery voltage would drop down to 7V after a heavy match, and under load, during the match, would display around 6V. We came very close to a brownout.

Of course, we were using 9 or 10 motors that year, and they were always on 85% of the time.

But its definitely possible to have the robot operate at dangerously low voltages, so it might be relevant to test the battery down to low voltages.

Do you typically run a fresh fully charged battery at the beginning of each match or is it possible the batteries had been used in one or more matches previously?

Our team rule is one match per battery per day. We put a freshly topped off battery in the bot at the start of every match and don't use it again that day if we can avoid it.

[nighterfighter]

Do you typically run a fresh fully charged battery at the beginning of each match or is it possible the batteries had been used in one or more matches previously?

Our team rule is one match per battery per day. We put a freshly topped off battery in the bot at the start of every match and don't use it again that day if we can avoid it.

This was in 2009, during Lunacy.

We DID use a fresh battery every match, and still do.

One aspect that drained our battery, we had 2 large fans on the back of our robot. If we got into a pushing match (which, in lunacy, happened a LOT. And pushing matches draw a lot of current from the CIM motors), we would activate both of our rear fans, at 100% throttle, which caused a HUGE drain on the battery. However, it almost always would win the pushing match...

[AllenGregoryIV]

All this battery testing discussion is great and useful but I really hope the collective community can provide some good insight into what their experiences were this year (good or bad) using battery brands other than the Enersys.

From the inspections I did, I didn't see many teams using many batteries other than MK and Enersys. Most teams don't do any testing or even keep track if a battery is good or not. It would be nice to know if any of the other brands are better than what we have been using.

[MrBasse]

We have a stock of batteries from the last few years and have only just retired our batteries from our first season this past year. We had the option to get a few batteries from our local Batteries Plus store with an excellent discount and gave them a shot.

They were the Werker brand battery and we had some good experience with the brand through electric car racing. We used them all season and never had an issue, but they did test low in the available current department and no load voltage. They tested comparable to Enersys batteries that were two seasons old when load tested. Of our ten batteries that we brought to competition, the Werkers were placed at number 9 and 10 in our rotation behind every other battery we used. I would have to get into our shop to see the numbers, but I do remember them being comparable to a three year old competition used Enersys battery that had also been used for practice for two years.

They also seemed to not last as long by the ever so scientific kids driving until the lights go out method.

Again, we still used them and never had any performance issues with either of them in competition.

[ChuckDickerson]

I took a look at the most recent suite of CBA tests we ran on the 13 new Enersys batteries we got this build season (12 purchased new + 1 KoP). The test were run the week before our first event. We ran 4 separate 50A tests with cutoff voltages of 10.5V (same cutoff Allen uses). The 13 batteries ranged from ~3AHr to ~5.5Ahr. Each of the tests take about 5-7 minutes to complete.

We have an off season next weekend so maybe we should run them through again and see how they are holding up. I'll add it to the list of stuff to do this week!

[Hugh Meyer]

We test down to 7 volts. This will show if any cells are bad more clearly. See previous thread with graph.

http://www.chiefdelphi.com/forums/showthread.php?t=82258

-Hugh

[AllenGregoryIV]

We test down to 7 volts. This will show if any cells are bad more clearly. See previous thread with graph.

http://www.chiefdelphi.com/forums/showthread.php?t=82258

-Hugh

How would that not be seen on a test down to 10.5v, the amphr rating would be very small on that battery.

[BBray_T1296]

I'm interested to know why Al tests all the way down to 8V.

At some points during long practice we would get as few as 5V through during a high acceleration demand, and we never got brownouts either

[AllenGregoryIV]

At some points during long practice we would get as few as 5V through during a high acceleration demand, and we never got brownouts either

Yes but was the resting battery voltage actually at 5v? The CBA testing at 7A over several hours isn't going to be able to simulate a 200A draw on a battery no matter what we do.

[Al Skierkiewicz]

Allen,
I test down to the level shown on graphs just to prove that the battery matches published performance. However, not all cells are created equal. While the overall battery may perform well when fully charged, some cells may die off before others. Every cell that runs out of steam drops the output by 2 volts. See one of my early (search) posts with graphs done on a really bad battery. One cell runs out rather early and then another and then finally a third. That test is one of the reasons I like the CBA so much. A Battery Beak or other short testing method would not show this problem. It might show up as a change in battery impedance but I don't have any real data that I can quote for that condition.

[AllenGregoryIV]

Allen,
One cell runs out rather early and then another and then finally a third. That test is one of the reasons I like the CBA so much. A Battery Beak or other short testing method would not show this problem. It might show up as a change in battery impedance but I don't have any real data that I can quote for that condition.

Have you had a battery that only showed this behavior by taking it down past 10.5v? We've found bad cells before the voltage drops off rapidly and the batteries have much lower capacity.

Maybe I'm missing something but I don't think a battery with a bad cell would have a comparable AmpHr rating to other batteries for any sufficiently long test, am I wrong here? Is a 10.5v cut off not a sufficiently long test?

[wireties]

Our local Interstate Batteries store is a team sponsor. We've been using their batteries for 2 years now with no problems.

Each year we take all our batteries to the local auto parts store for testing and discard those that don't pass. And of course we always run matches with fresh batteries! But after reading this thread I wonder if we should do a more thorough test.

[MrBasse]

After using batteries for a parade on Saturday, the Werker batteries we have won't complete a charge cycle on our Battery Tender maintainer. They also are getting quite hot after charging at a 2 amp max charge. I think these ones are going off to be recycled. It's a shame they didn't even make it one full year.

[evanperryg]

6 during each build season.

Why so many every year? As long as you take care of them during the off-season, they should be fine for many seasons of use. My team still keeps around a couple of 2011 batteries because they actually perform just as well as the batteries we got this year.


But its definitely possible to have the robot operate at dangerously low voltages, so it might be relevant to test the battery down to low voltages.

To my knowledge, the lowest our DS has ever read in-match was 5V during a brutal pushing match. It is very important to know how your batteries perform at low levels.

Do you typically run a fresh fully charged battery at the beginning of each match or is it possible the batteries had been used in one or more matches previously?
Our team rule is one match per battery per day. We put a freshly topped off battery in the bot at the start of every match and don't use it again that day if we can avoid it.

We do our best to avoid using the same battery more than once a day. Our battery box holds 12 batteries and uses a raspberry pi to automatically cycle through the batteries and charge 3 at a time, so we never have a problem with having enough charged batteries. Of course, there are times that a battery must be used more than once per day, and as long as it has had plenty of time to sit, it should be okay.

[AllenGregoryIV]

Why so many every year? As long as you take care of them during the off-season, they should be fine for many seasons of use. My team still keeps around a couple of 2011 batteries because they actually perform just as well as the batteries we got this year.


We like to run new batteries. One of the reason we are testing them more now is to see if we can judge the differences better and maintain them better. We also give away some of the batteries that we no longer think we need to young teams. $250 in batteries isn't that much for what could be the most important part of the robot, with out the battery nothing else works.

[BBray_T1296]

with out the battery nothing else works.

[tangent alert] You could say that about much of the robot though.

[AllenGregoryIV]

[tangent alert] You could say that about much of the robot though.

I'll admit I could have worded that better. What I meant is that it's the beginning of the power path so it affects the performance of all other systems.

[Al Skierkiewicz]

Allen,
The mechanics of this bad cell failure (in my best guess) is that through damage, either drop or robot hit, some of the plates in a cell become dislodged from the connecting terminals. As such the remaining plates still function normally but the cell has a lower amp hour rating as it will become depleted before the other cells. If we were to test at full current, it might show as a change in available current at 400-500 amps. When testing at the manufacturer's spec, you will see the battery run a normal discharge curve, in some cases more than hour, before the bad cells actually become evident. To visualize, say a cell has 10 positive plates and 10 negative plates. The battery gets dropped and one of those plates either cracks or comes completely off the output terminal plate. Since the plate area is directly proportional to the available current and the amp hour rating, any reduction will reduce both of these measurements. The CBA testing will establish a fixed current load that will calculate the amp hour rating without the need to measure max available current. Since both are affected by damaged plates, just knowing that a reduced cell amp hour rating is enough to declare the battery "practice only". Close examination of the battery case will almost always point to case damage from the drop or hit. If you search "AGM internal structure", you will find a variety of pictures and other links to get an idea of what could go wrong internal to these batteries when dropped.

[AllenGregoryIV]

Allen,
The mechanics of this bad cell failure (in my best guess) is that through damage, either drop or robot hit, some of the plates in a cell become dislodged from the connecting terminals. As such the remaining plates still function normally but the cell has a lower amp hour rating as it will become depleted before the other cells. If we were to test at full current, it might show as a change in available current at 400-500 amps. When testing at the manufacturer's spec, you will see the battery run a normal discharge curve, in some cases more than hour, before the bad cells actually become evident. To visualize, say a cell has 10 positive plates and 10 negative plates. The battery gets dropped and one of those plates either cracks or comes completely off the output terminal plate. Since the plate area is directly proportional to the available current and the amp hour rating, any reduction will reduce both of these measurements. The CBA testing will establish a fixed current load that will calculate the amp hour rating without the need to measure max available current. Since both are affected by damaged plates, just knowing that a reduced cell amp hour rating is enough to declare the battery "practice only". Close examination of the battery case will almost always point to case damage from the drop or hit. If you search "AGM internal structure", you will find a variety of pictures and other links to get an idea of what could go wrong internal to these batteries when dropped.

I understand that Al. My question is have you seen a case were this only showed on a test with a lower cut off voltage than 10.5V. We don't test lower than that, so I am wondering if we should every once in a while. Perhaps, if a battery has been dropped we should test it down to 8V just to be sure.

Part of the reason we don't test that low is to save time and part of it is because, though observation only, batteries that are drawn that low (always on the robot or powering an inverter) don't seem as healthy afterwords but that could just be observation bias.

[Al Skierkiewicz]

Yes,
The cell may be just shy of running out of energy at 10.5 volts. For that matter even the lower cutoff voltage may not show a partially damaged cell. The lower voltage will catch more defective cells than at 10.5 volts in my opinion. In some cases, internal temperature may exacerbate the intermittent connection. I had one battery that confused me for a while because when the cell got close to dropping off the graph, it suddenly produced a lot of "noise" over a period of 20-30 minutes before dying. These voltage peaks and dips were several volts. I came to the conclusion that the cell had partial damage and the noise was caused by the plate intermittently making and then breaking contact with the terminal plate.