My battery lesson...

[bahama flier]

After five years in my Cherokee Six, I thought I should replace my battery, you know just to be sure.

Looking at new battery prices, I went with an acid battery, yes cheaper.
I very soon started to have battery problems, they battery acid came out the drain tube and splattered the whole underside of my new expensive paint job. It had a bad cell, and I had a lot of paint peeling.

I purchased the battery from Aircraft spruce, they don't do anything but tell you to contact the manufacturer, I did and was told to send the battery to them (at my expense)and they would look at it.

I just eat it and got a gel battery, no more problems, been five years now all OK, lesson learned.

DON'T PUT AN ACID BATTERY IN YOUR AIRPLANE

 

[bgdriveway]

Any battery retailer will tell you to contact the battery manufacturer. That's normal. Sorry about your paint. It's hard to go wrong with a sealed Concorde battery though. They do cost more.

Instead of just replacing a battery on a set period of time, you can do a capacitance check. Some shops have a machine that puts a load on the battery for a specified amount of time. Usually an hour.

Here's how it works. Say the battery is rated for 12 Ampere hours. Meaning its supposed to be able to support 12 amps of a load for 1 hr. Or 6 amps for 2 hrs. The tester loads the battery to the amperage set and measures voltage drop. End point voltage (EPV) for a 12V battery is 10V while nominal voltage for it is about 12.8V fully charged. EPV on a 24V batt is 20V. The load is placed on the battery and voltage is watched until it drops below EPV. At that time, the test stops. The length of time will tell you how your battery is for its rating. 51 minutes (out of 60 min) is 85%. 48 min is 80 percent and so on. You can have a battery that is over 100%, but it's not normal for one that's been in service for awhile. Concorde says pull a battery if it's less than 85% and replace it. I don't remember if there's a legal requirement somewhere.

Here's the cheating way to do the test to get a rough idea where your battery is without the cool machine. Figure out how much load your nav lights/landing lights/ etc are. (Lights are a high load item, which is why I start with them.) Get as close a load to your battery rating as you can. Although some say that a 2 hr test with half the amperage of your battery's rating is more accurate. Turn the lights etc. on with a fully charged batt, and start keeping time. If you have a voltmeter installed you don't even have to go find one to measure the voltage. Keep track of the time and do the math when voltage gets below EPV.

I would pull all the other circuit breakers so that you only have the planned load. (You can't just count up numbers on the breakers and use that for your load. Breakers/fuses are to keep from burning the wires and should not be fully loaded to their max rating.) Just be aware of any other parasitic loads that you can't get rid of by pulling breakers and caculate for them.

Make sure to recharge the battery to avoid welding contacts on relays when you go to start the plane.

If your battery is in the 80+ range, I wouldn't worry too much about it. If you fly IFR and your battery is in the 45% range, I'd get a new battery tomorrow.

Most of this I learned from reading Concorde's manuals, readily available on their website or they come with a new battery.

 

[weirdjim]

One second order problem is that as the battery voltage goes down, the lamp current goes down.

I much prefer a 5 to 10 hour test as it negates battery heating as a source of error. Harbor Freight sells a nice little multimeter with a 10 amp scale on sale this week for $5. That way you know for sure what the load it.

Don't try to hit an even hour; try instead to keep it between 4 and 8 hours by turning on lights until the current flow is between 12 and 25% of the rated amp hour of the battery. For a normal 25 ah small aircraft battery this will be between 3.0 and 6.25 amps. It doesn't really matter where in this range it falls. Then take the measured current (let's say, oh, 4.15 amps just for giggles) and divide the rated current by this figure. I come up with 25/4.15 or 6.02. This is the number of HOURS that the battery should stay above 10 volts (and no, I don't know where this number came from or if it the same for all batteries, only gel, only wet, 25ah, 35ah, or what). 6.02 hours is 6 hours 1 minute 12 seconds.

Your battery will fall below the magic 10 volts in some time less than this unless it is brand new. COnvert the time that it stays above the magic number into hours and decimal hours. Divide this by 6.02 volts and this is the % of your battery's capacity.

Hmmm .... I smell a KItplanes article coming out of this discussion ...

.
.

Jim

 

[weirdjim]

DON'T PUT AN ACID BATTERY IN YOUR AIRPLANE

Ummm ...

They are ALL acid. Some are liquid and some are gelled, but they are all H2SO4 sulfuric acid.

Jim

 

[txflyer]

Beats me how someone can spend what we spend on these tubs and then put a Walmart battery in it.

Gel is the way to go. Less spillage.

After seeing what standard fluid batteries eat up on the farm, I want no part of that near my aluminum.

 

[bgdriveway]

So how much error is heating induced? I'm not worried about at few percentage points. If there is some legal requirement, then maybe it matters. It's just good to know that your battery is doing well or if it's junk you should know that if you are relying on it as a backup. Day VFR, ehh... Not a big deal typically. Depends on what you are doing.

Also, I don't have time, or want to keep running over to a plane across a 6 hr test. Get it done, and lets keep on moving. 6 hrs is a long time to be checking for when the voltage drops.

I don't know how big of a deal it is to keep bumping up the load on the checks. Yes it will drop, but I don't know if it's significant.

I believe the 10 as the EPV comes from the minimum residual voltage in each cell added up. NICAD's hold voltage really well then go off a cliff, figuratively speaking while lead acids do some sort of down hill slope.

 

[weirdjim]

So how much error is heating induced? I'm not worried about at few percentage points. If there is some legal requirement, then maybe it matters.

Does it matter "how much" error? I'm only a poor p1$$ant engineer that knows that battery capacity in elevated temperatures inside the case go to hell. "Go to hell" is a subjective term, and until some EE on a master's program that has more time than I chooses to do his/her thesis on the project, I can't say. Just "some" until then. As to a legal requirement, again, I'm only a p1$$ant A&P IA. If there is some legal beagle out there, please chime in.

It's just good to know that your battery is doing well or if it's junk you should know that if you are relying on it as a backup. Day VFR, ehh... Not a big deal typically. Depends on what you are doing.

"Doing well" and "junk" are subjective terms at best; pejorative at worst. All I'm trying to do is give you a number. You want to quantify the number, be my guest.

Also, I don't have time, or want to keep running over to a plane across a 6 hr test. Get it done, and lets keep on moving. 6 hrs is a long time to be checking for when the voltage drops.

Like you aren't going to be in the hangar for 6 hours for the annual? Or bring it home and put it in the garage? Just like water boiling in the pot, don't keep watching it.

You know, I'll bet some smart person who writes magazine columns can come up with a way of measuring when it goes past the magic limit.

I don't know how big of a deal it is to keep bumping up the load on the checks. Yes it will drop, but I don't know if it's significant.

OF course it is not. That's the way the battery is going to work in an emergency environment anyway. By the way, the limit on avionics working correctly is 11.5, not 10. Somebody is going to have to do the research to find out what the magic end voltage for all combinations of batteries is going to be.

I believe the 10 as the EPV comes from the minimum residual voltage in each cell added up. NICAD's hold voltage really well then go off a cliff, figuratively speaking while lead acids do some sort of down hill slope.

That's not true on NiCads vs acids, but believe what you will.

Jim

.....

 

[Henning]

Gel is yesterday, and easy to fry the battery, AGM is the way to go at this point

 

[weirdjim]

Gel is yesterday, and easy to fry the battery, AGM is the way to go at this point

Shut up and go away. We are talking gel and AGM as the same thing, or didn't you bother to read the whole thread?

Jim

 

[bgdriveway]

All I was trying to say is I don't see a 1 hr test being a problem with the heat. That's going to happen if you have an electrical failure and you are relying on your battery.

I live in an Apt. No garage, can't bring the batt home. When I'm doing an annual yes I pull the battery and put it on the tester. I'm not going to work for 6 hrs on an aircraft with the power on. As you are an A&P, you know time is money. And I don't have that kind of time. I was suggesting leaving the lights on for an owner who was curious as to where their battery stands with out making something super complicated. Maybe I didn't succeed.

Please enlighten me then on the difference between nicad's and acids. Yes there is some degradation of a nicad voltage before it drops off, but the drop off on a nicad is much steeper than a acid. This I have witnessed myself. As for coming up with something that will tell you when the batt has reached EPV, the call those things testers. I've used several different types.

I don't see your and my information being all that different, just that we are coming at it from different angles.

 

[weirdjim]

I submit for your consideration the attached two graphs.

Discharge 1.jpg is the discharge graph of a standard lead-acid battery. Note that at 1C (discharging it at the rated amp-hour capacity) gives you a battery that is "flat" (below 10 volts) in about 22 minutes, but it is relatively flat for about 8 minutes and then falls off rapidly. For a 25 ah battery, the discharge rate is 25 amps.

Note that at 0.2C that the battery doesn't hit the (extrapolated) 10 volt mark until about 4 hours. This is the way I like to test batteries. The current here is 5 amps. Still, this battery stays flat until about 1 hour into the test.

Contrast that with the nicad discharge curve. Note that it starts off life with a relatively sharp drop immediately (surface charge leak-off) and then slowly drops down until it has a rather soft fall to its capacity cutoff point.

I therefore rest my case that ...

• A lead-acid battery can be tested more accurately for rated capacity remaining at a 20% load (0.2C) than at a 100% load (1.0C).

• A lead-acid battery has a flatter discharge curve than a nicad.

Jim


.
.

 

[Henning]

Shut up and go away. We are talking gel and AGM as the same thing, or didn't you bother to read the whole thread?

Jim

Well, if you treat Gel and AGM as the same thing, good luck, it gets expensive.

 

[Dan Thomas]

AN AGM battery uses Absorbent Glass Mat to hold the liquid acid against the plates and therefore hold the bubbles of gas generated against the plates until recharging absorbs them. The gel cell uses a gelled acid to do the same thing, but its drawback is that it can develop more or less permanent bubbles against those plates and reduce the surface area, reducing the battery's capacity. Both types are sealed.

Ordinary batteries will make a mess and eject acid if they're overfilled. Way too many get topped up to the bottom of the split ring when they're just sitting there doing nothing, and that electrolyte expands once the battery is being charged (the bubbles) and it will overflow. The battery manuals will tell you to top it up while still actively charging, not at rest. If I check a battery's electrolyte levels in an inactive state, I won't put in any more water than it takes to bring the level halfway between the tops of the plates and the bottom of the split ring, and they never overflow.

Dan

 

[bgdriveway]

Ok. Fair enough on the graphs, (assuming the info is accurate, which I am in no position to argue one way or the other.) Maybe that's one more area that there's a lot of misinformation on that I have just now been corrected.

But even on the chart everything is hitting EPV before it should have if I am reading it correctly. That is, I would expect a 1C load to go 60 Min, a 0.2C to go 5 hrs, etc. And what is with the dotted line, I'm guessing is the EPV rising after longer tests? I am not following that.

Concorde does tell you in the manual to let the battery cool, 12 hrs I think, after charging it before beginning the test.

Here's the down side that I see to doing the test over the several hour period. Maybe you have a better explanation. If one was to have a alternator/generator failure in flight, I don't expect that many people would keep flying for another 6 hrs, or even 2 hrs typically. I don't see how testing for that long will help you determine that your battery will last a whole lot longer in the case of a real system failure. If there was a legal requirement to have a battery rated at 80 or 85% or any other fixed value, then yes, a longer test giving a better, more accurate number would be worth while. IF however, you want a decent idea, I don't see the one hour test being that much worse.