Parasitic battery drain problem

Richard Casto · **Posted:** Thu May 14, 2015 3:23 pm

Now that it seems Rodney has maybe solved his fan problem with his Ford, you can give me some idea on a problem I am having.

Background and problem...

* New to me 2005 Honda Civic
* New alternator
* New battery
* Factory radio, no alarm system. Completely stock electrical system as best as I can tell.
* Car was bought for my daughter who is learning to drive. She doesn't have her license yet, so car sits without being driven for six days of the week. On the other day, it gets a good hour or more of drive time.
* Battery drains after sitting a number of days. So you must jump it to get it started.
* I "think" this problem started after I did a bunch of maintenance items (timing belt, water pump, adjust valves, new A/C compressor, new Alternator, new PS pump, new plugs, etc.). However, I am having a hard time building a scenario that my work caused this. The best I can come up with is something with the alternator (bad diode causing drain, it is really not charging like I think so that plus the expected drain is killing the battery)

What I have checked so far...

* I need to check more extensively, but the alternator is putting out more than 13-14V so it seems to be charging the system. However, I have occasionally seen the "battery" light stay on for a few seconds longer than normal upon start up. At the moment I don't think it is the alternator.
* I have no load tested the battery, but this problem happened with the old battery and also with a new battery. I need to load test the new battery
* I have hooked up my multi-meter in Amp mode and have tried to see if there is excessive drain on the battery when the car is parked. I am seeing about 31mA of current when the car is at rest (lights off, doors closed, etc.). This seems high to me, but from a bit of Googling around, it seems to maybe be within the range of potentially normal.
* I have pulled all fuses (both under hood and inside) and I can't find a circuit that consumes a large amount of current. One does draw about 6mA and that is under the hood and labelled "Backup". However I believe that is for the radio, dash clock, etc. And that is not the problem.
* have currently disconnected the battery all week and this weekend plan to reconnect to see if the battery held a charge on it's own all week.

Questions...

1. Is 31mA out of the ordinary for a normal drain? As I type this, I realize I should measure my other cars to see what they draw.
2. Am I just not driving it enough? It will get driven more once my daughter gets her license.
3. Other ideas?

Richard

Cash Davidson · **Posted:** Thu May 14, 2015 3:33 pm

I have had similar problems on a couple cars. 31mA is not high and should not discharge the battery in a few days. I had a drain over 100mA and it took 3 days to kill a fresh battery.

Your approach is correct - pull fuses until the meter reads zero then focus on that circuit. Mine turned out to a 30A circuit under the hood that killed the radio, dash lights, door locks, dome lights, anti-theft sys, etc. I had to trace & isolate each of these and it ended up being the anti-theft. I left it disconnected.

Good luck!

RodneyWright · **Posted:** Thu May 14, 2015 10:14 pm

I have little to offer here. Sounds like you're attacking this right by checking the draw on the battery. Is there anyway you can start looking at the fuse box and looking at any draws there? It might point you to something. I say this because one terminals on the mustang fusebox was supposed to have no draw, but I was showing .02v, so you may have a vampire somewhere...

MikeWhitney · **Joined:** Wed Aug 20, 2003 9:15 pm **Posts:** 4724

- At 31 mA, a 60amp-hour battery (typical) would last 80 days. 31mA is fine.
- I would have said "defective battery" but if it's the same with old and new, we can rule that out (for now)
- Double check voltage -- after running the car for a while, it should be 12.5+ while OFF

- As long as you're seeing 13-14V while running, then the alt is fine.

So ... if we've essentially removed the battery and alternator, then I'd suggest checking electrical connections. I have had a couple of situations where a car would not crank but the batt and alternator were fine. It was always one of these:

- Make sure the engine ground is in place and looks good
- Be sure the batt terminals are clean and tight
- Check the starter itself for tightness to the transmission
- Check the starter power wire nut for corrosion and tight
- Check where the battery negative cable attaches (body? engine?)

If you find no problems, try a 3rd battery.

Jordan Normark · **Posted:** Fri May 15, 2015 3:32 pm

Funny, The first thing that popped in my head was my FIL's experience with his Honda Pilot a few years ago. He inadvertently switched on one of the interior lights in the rear. It was on all the time, so if you opened a door you would expect it to be on and would not think anything of it. Also, it was just dim enough that he never noticed it when in or around the car. I don't know how many mA the draw was, but it was enough to kill the battery in a day or two. I was surprised Honda would design an interior light/switch circuit without a timer.

Richard Casto · **Posted:** Fri May 15, 2015 4:19 pm

Thanks to everyone for suggestions (and Mike for the math on the battery. That was on my to do list.) I haven't looked at the car since last weekend and will play with it this weekend (check ground connections). To Jordan's comment/question about the interior light. The light is verified to be off. Regarding interior light user experience, I have only had Honda's for a number of years, so I have the reverse experience that I am just used to the fact that if the door is open the interior light is on. I "think" that is standard thing for all Hondas. You can however manually turn it on/off. Actually the interior light switch has three positions... Off, auto (door open = on, door closed = off) and On (no timer and will drain the battery). Other light issues to look for are something like trunk or glove box lights that might be stuck on.

Richard

DickRasmussen · **Posted:** Sat May 16, 2015 9:16 am

Richard Casto wrote:

Thanks to everyone for suggestions (and Mike for the math on the battery. That was on my to do list.) I haven't looked at the car since last weekend and will play with it this weekend (check ground connections). To Jordan's comment/question about the interior light. The light is verified to be off. Regarding interior light user experience, I have only had Honda's for a number of years, so I have the reverse experience that I am just used to the fact that if the door is open the interior light is on. I "think" that is standard thing for all Hondas. You can however manually turn it on/off. Actually the interior light switch has three positions... Off, auto (door open = on, door closed = off) and On (no timer and will drain the battery). Other light issues to look for are something like trunk or glove box lights that might be stuck on.

Richard

Probably not applicable but our 84 Nissan had a brake light switch mechanism with a plastic spacer between the pedal arm and the switch. The spacer failed after many years which resulted in the brake lights staying on and draining the battery. The way we parked the car at that time "hid" the fact that the brake lights were on.

Roger McDaniels · **Posted:** Wed May 27, 2015 9:12 pm

I recently had this problem with Ricky's van. I just had to keep pulling fuses until I tracked down the draw. Ours turned out to be the AC compressor clutch, which stays on if the AC button is pushed, regardless of whether the van is on or off. Now I just have to remember to turn off the AC before leaving the van. It's either that or actually fix it, and fixing it seems like more work than not fixing it.

Richard Casto · **Posted:** Thu May 28, 2015 12:29 pm

Update...

I haven't been able to spend a lot of time on this. I wasn't 100% sure that the battery was fully charged so I really made sure of that before doing any new debugging/testing. I didn't want to try to let it sit a week if it was not full. I also checked the charging voltage under different conditions. With everything on (radio, interior lights, headlights on high beams, A/C on, etc.) at idle is was marginal as it was right at 12V, but with the engine running at a higher RPM (such as driving) it was above 13V. And when when not on maximum load it was more than 12V at idle. So I continue to think it is charging fine "while being driven" and maybe also fine at idle as well.

I can't remember the exact order, but somewhere in this I let it sit a week without it being connected to the car. I connected it to the car and it started fine. I then let it sit with it connected to the car. Maybe about 4-5 days in I tried to start it and it started fine. I ran the car for about 5 minutes (mostly idle) and shut it off. I tried again about 3-4 days later and it was dead as could be and I was totally shocked.

At this point I am tired of messing with it. I have since charged the battery, but have left it disconnected from the car. I am awaiting a quick disconnect battery switch from Amazon so that I can turn it off when I am not messing with it. I haven't checked ALL of the connections Mike suggested above, but the terminals are clean and the ground strap for the engine is connected. One "odd" thing is that I don't know if there is more the one ground strap, but when I bought the car a strap (clearly a braided ground strap) that connects from around the area of the thermostat on the head was laying loose and not connected to the body. The car ran fine without it connected, but I did figure out where I think it connects and did reconnect it to the body (near the radiator). I can't imagine that would have anything to do with it, but i figured I would mention it just in case. The only connection I haven't really checked is on the starter (on my to-do list).

Given that it seems to measure a low current each time I check (low 30ish mA) and that draw can't be traced to any of the fused circuits I am still at a loss as to what is going on. My current wacko theory is that something wakes up and causes a higher current load. I am about one step away from doing some type of Arduino based data capture to see what the current load is over a period of a week.

Regarding the A/C switch idea. I think that is probably just a signal to the ECU anyhow as I think it can decide when to cycle the A/C clutch as needed??? So I suspect that is not it, but will check anyhow.

At the moment I have nothing pushing me to solve this as the car is mostly not needed at the moment, but the fact that I haven't been able to solve this yet is bugging me.

Richard

Richard Casto · **Posted:** Thu Jan 21, 2016 10:11 am

Hopefully my final update on this. Short version... Replacing the alternator seems to have fixed this.

Long version...

As mentioned above, the alternator that came with the car was replaced because the positive stud snapped off when I was removing it to get at fixing the A/C system. The first replacement was a totally new (not rebuilt) unit and "seemed" to measure OK when charging. But in the post directly above you see some indication from me that maybe it wasn't putting out a high enough voltage.

I had too much other stuff going on to deal with this for awhile. I ended up "topping off" the battery weekly with a weekend charging session to keep the battery from going dead. Eventually my schedule freed up and I spent more time looking at it. I then found that in some situations it was not putting out a consistent voltage that would charge the battery. So if I started the car up, it might show a good voltage and current, but if I left it running and checked later it would be lower than it should be. Not low because it didn't need to charge, but low enough that it wasn't maintaining the battery.

After some reading, I figured either it was the internal regulation in the alternator or maybe a problem with something Honda does (I suspect other manufactures do something similar) which is to use a electric load detection circuit that senses when there is load on the overall electrical system and then tells the alternator to increase or decrease the field voltage to adjust the output. The idea is that when the load on the system is low, it can crank down the field voltage to decrease the load the alternator puts on the engine (to help with fuel economy and probably increase alternator bearing life). Given the problem seemed to show up with the "new" alternator, I did't dig into debugging if the load detection signal (from ECU) was funky and just decided to replace the alternator again. So I put in another one (rebuilt this time) and it immediately showed a higher charging voltage than the prior unit. I also checked the voltage drop in both the negative and positive path between the alternator and battery to make sure I didn't have issues with the current paths (bad wire or corroded connection somewhere) and didn't find any problems. That has been about a three weeks ago and all seems to be good.

Richard

THSCC Forums

Parasitic battery drain problem

Who is online