Parasitic battery drain puzzler

I have a 95, 3.8 Taurus with EATC, analog instruments, auto headlights with delayed exit light turnoff, remote fob, keyless entry.

After enough puzzling and getting tired, I decided to clear my head, rest and hopefully have more collective thought from ya'll and hopefully shorten my troubleshooting.

I have a brand new battery and am loosing charge. Checked alt and then the underhood fuses. The "headlight" circuit was still active. Thinking disconnecting the headlight switch and the auto headlight switches would do it, I came to learn FIVE of the I/P fuse circuits are on the "headlight" main fuse. Continuing with my test light at the battery negative and clamp, I found one, no then two circuits at fault, depending on what order you pull them and reinstall them. I have yet figured out how fuse circuits 8 and 16 would be related to influence each other so that one disconnected but not the other will not open the drain. For example, the test light won't go off when I pull 8 alone, but will when I also pull 16. If I just pull 16, nothing happens until I pull 8 second. It's bizarre. I was hoping this would point me more quickly to my problem. I may have to ignore this phenomenon for now. Fuse 16 only has the EATC on it as I don't have the digital instr cluster. Fuse 8 has the courtesy/inside lights, radio memory, clock memory, keyless entry, power mirrors, illuminated entry module. TTBOMK, everything is working. I will have to double check if the problem doesn't get solved soon. Thanks.

Have you recently replaced the Digital Transmission Range Sensor? I seem to recall some time back another poster having an issue with the radio and EATC and it turned out to be a bad DTRS. I don't recall details though. I also seem to remember not being able to make sense of this since the DTRS didn't seem to be in the circuit for these components. I don't believe that poster had tried narrowing it down to a circuit or pair of circuits. Maybe try unplugging the DTRS and see if that has any effect on the current drain.

I also see that the instrument panel dimming rheostat is on instrument panel fuse #8. That would affect both the EATC as well as the radio, you may consider unplugging the headlight dimmer and seeing if that removes the parasitic draw. If not, then try pulling the radio and disconnecting the connector with circuit #8 in it to see if that fixes it, then the EATC.

You're right, there's nothing obvious to explain this one....

-Rod

As I looked further into the wiring diagrams, it appears circuit 8 only goes to dimming rheostat for the courtesy light portion and not for the illumination dimmer, so that wouldn't be a practical item to disconnect for troubleshooting.

-Rod

Uhhhhhh, a "digital" trs? Didn't know there was such an animal. I just thought there was a trs. Haven't changed it but will see what disconnecting will do. I can't make the connection though how it relates. I hadn't seen it in the diagrams, not to say I couldn't have missed it.

When I first investigated this, as I posted, all of the headlight stuff was and still is disconnected, so that would include the dimmer too. The dimmer is next to but not part of the auto headlight switch, but it controls both of the interior and dash lights. I thought disconnectiing the dimmer was practical as it was one of the easiest ones to disconnect! :D Yeh, I too see the two circuits aren't electrically connected and am not seeing anything that would join the two circuits in any of the diagrams. :banghead :screwy :runaround

Functionally, I haven't noticed anything not working, but I should more deeply investigate that. FWIW, I noticed that with the ke and the auto lamps, the kem takes the place of the iem. Could this be related to the replacement PCM? Could a dead digital clock do this? At this point, those things didn't appear to me to connect the two circuits, but at least the clock is part of one of the circuits.

FYI, this a station wagon.

Right, when I read about some of the past issues that were fixed by the DTRS/TRS I too didn't see a connection. Still don't. Some manuals call it the Digital TRS, others just the TRS. In this case, either are correct since there are discrete (digital) steps rather than continously variable (analog, like a TPS).

I wasn't sure from your first post if you disconnected all the headlight stuff or just discovered the additional circuits as you were looking at the diagrams for the headlight circuit. That rules one more circuit out then.

Help me out, what's the KEM and the IEM?

-Rod

Still no joy. :banghead: :runaround: :screwy: :shakehead :mad: :crying: .
I've eliminated the remote keyless entry module/illumination entry module, rkem/iem, the dead clock, and the interior lights as best as I can tell. I can't be 100% sure that I got all of the hidden lights if they are out since I can't see them! What hidden lights I remember are working. Suggestions where to look would be appreciated. I pulled the trs connector and even the connector next to it, which I guess is for the trans sensors and such, just to have no doubts. Maybe I'll even take a tire off if it will help!?! :screwy: :shakehead :banghead: :runaround: :mad: :crying:

So what remains to rule out on fuses 8 and 16 are the eatc, the pwr mirrors and radio memory.

I probably should have asked these questions sooner:

Did you replace the battery because it didn't hold a charge, or did this just happen after replacing the battery (ie: maybe you got a bad battery with a high internal resistance)?

Are you determining there's a draw because a test light in series with the battery glows, or are you using a current meter. If a meter, what is the current draw, and does the current draw drop after a minute or so? If a test light, can you get a current meter and post the current draw in amperes or milliamperes?

Just for fun, try disconnecting the alternator and re-checking the current draw.

When you're checking for current draw, you don't have a hoodlight that's on or a door open, do you? Can you verify the trunk and glovebox lights are not on?

It's not all that bad to disconnect the radio and EATC to determine if they are the source of the draw. You may want to unplug the radio and see what the draw goes to, but first I'd suggest knowing how much current draw you have.

-Rod

I'm going out there now to pull the eatc, radio and maybe the pwr windows. I already said in my first post that I eliminated the alt which I actually suspected in the first place. The battery was replaced as it would no longer hold a charge and it was quite old, so it's replacement was expected. I do not know if there was a parasitic drain prior to the battery replacement, but at least twice the new battery was down to nothing. I can recharge it to normal and it will hold a charge with the battery disconnected. How would I figure out if it is internally defective? By it's voltage?

Well, I thought I said it before, maybe not, but I am using a test light in series with the battery between the negative post and clamp. I think when I put my amp meter in place of the test lamp, I measure around half an amp. I'll have to check that again. No, no doors open, pulled hood bulb, glove box off, no trunk - it's a sw (boy you're loosing it), interior lights off. The test light glows about half bright. I did make the mistake of having a door or something on that had the light really bright and me stumped why so many power distr. fuses were involved. Was tired. Would it be meaningful to measure the voltage drop across the test light? I think it was around 9v with the battery 12.5 to 13v.

The voltage drop across the test light could be used to determine the current draw, but one would need to know the impedance of the test light to make the calculation. It seems easier to just use a current meter in place of the test light, although the test light is a better visual indication of when the draw goes away. It will be interesting to see what happens as you disconnect the EATC, radio, and possibly the power windows. If you find the radio is the problem, I have the one I removed from my 1993 SHO JBL system years ago. If you don't have the JBL system though, that may not do you any good as I don't believe it has an internal amplifier.

Oh, do you know if the car ever had an aftermarket alarm system installed? Conversations for these types of goofy electrical drains often start with, "Well, I was digging around under the dash and found someone had installed a Brand X alarm system and removed the fuse. Does anyone know where I could find an owner's manual for this?"

As far as losing it, yeah, I probably am. I've spent much of the day working with wood stain.

-Rod

So, if I measures the ohms of resistance of the test light and use I=V/R, I could calculate I and compare it to the measured I. Just to be sure, V is measured across the test lamp. I guess I could also calculate the current that is drawn by whatever is draining it by measuring the resistance of the drain and the voltage across it.

Well, I pulled everything that I know of on the suspect circuits. At first I thought I had it narrowed down to the radio at one connector as the test light dimmed at least by half. Nope, couldn't reproduce that later. ERRR. I eliminated the pwr mirrors (not the windows, they're not included) and all lights that I could find including out bulbs taking them out.

I was able to duplicate the bizarre electrical draw involving the two fuse circuits as I had described in first post. The EATC and the RKEM are somehow affecting each other. I was able to eliminate every single current draw. Without the EATC and the RKEM, the total remaining current draw is around 2 mA. Most of the time, the eatc and rkem draw around 6 mA. But, earlier, before I started pulling the connectors at the back of the radio and eatc to see which was the problem, when I would connect the meter at the battery, I would get current to start at 2 mA, I assume from the other normal current draws, and it would climb. It maxed out at 18 mA total or 16 mA I assume from the parasitic drain. How in the world are these two things interrelated? They're on different circuits and at opposite ends of the car?!?

I had previously completely disconnected the rkem/iem two modules with no change, but now realize that nothing would have happened since I also did not disconnect the eatc.

I don't think I need a new radio, but thanks for the offer. I think the FSM has mine with its own amp. TTBOMK, there isn't any security system. I'll look, but I've not seen anything that would make me think there is one. I sure do remember those folks with crazy electricals with those am systems. I'll check the wiring for splices and connections at the back of the ip. Any suggestions where to look and what it would look like?

Now I'm really annoyed with Ford's sm. I looked up the fuses and got their list of things on them, and they were severly incomplete!!! ERRRRRGGG!!! I had spent so much frickin time on the two suspect items thinking they were the last things and now looking at the pwr distr. dia there may well as be a million more things to check!?! :banghead: :runaround: :mad: :shakehead :crying: Uuuuughh!!!!! Sighhhh. I think it would be easier to pull the fuses or the battery cable or charge it once a week at this point. I've got too much other s--t to do, and I'm too f--k n tired to fool around with this kind of s--t.

Well, I wanted to check things one more time before I posted this to see what kind of battery voltage remained overnight with the drain still there, and it didn't seem like it was drawn down much. I retested the amperage draw and used the test light, and it seems the problem may be gone for now. :shakehead:runaround: :mad: :banghead: :crying: So I don't know what to do for now. I did measure up to 4 mA total draw, much lower than before. I don''t know what to do from here, and I don't want to get burned by this later.

Well, 18mA would not be a big draw. The threshhold is around 70 mA before you need to be concerned. You mention you think the current meter showed half an amp draw. If the problem comes back, try using the meter.

-Rod

I used both to measure the current. Right now, it seems to be ok. It is drawing 2 to 2.5 mA when letting it sit for the day or overnight. I wonder how much current draw there was to make the battery so weak it wouldn't start, maybe sitting for a week. I may have measured it once just below 10 v when wanting to start it. I just hate for this thing to strand me at a bad time unexpectedly.

Are you sure that your current meter scaling isn't off by a factor of 10? 2.5mA doesn't even seem like enough for the radio presets to be retained, and certainly not the radio AND PCM. Now 25 mA I could believe....

It almost has me curious enough to go measure the draw on my Mountaineer to see what it has for current draw. Someone once warned me about experimenting with my own vehicles in an attempt to help others out though. ;)

-Rod

I'll measure it with two other meters, though I'm less confident in them than this one. It seems the problem has not surfaced for one day. Just before using the car today, I measured something really low, I think under 2 mA and the battery maintaned its voltage of 12.5 v, too. Previously, it would instantly bring it down 0.1 v from 12.5 v. I think it originally went below 10 v when left for at least a week.

When burning up your pcm......putting your meter on the battery, ONLY go between the neg post and removed neg clamp, not the neg post and the red post thingy. It'll toast your meter when your dead tired, nearly asleep. Just in case you stumble your way under your Mountaineer hood. :D Hmm, I warned you? Well, I don't think it's safe to experiment with it either, but measure it should be ok. Ok, now that I've warned you, what did you measure?

I'm gonna research more on what is a reasonable current draw. Oh yeah, I have a few more sacrificial vehicles that I can burn up.....test.....experiment..... measure. I'll report back with anything interesting. Thanks Rod.

If I hadn't stated it before, when there was a higher than expected amperage, it would start around 2mA and then would climb stopping around 10mA and maybe once at 18mA. This would take several minutes for it to level off. Any ideas why it's doing this?

If you have time, take a look at group 14 - battery and charge system in the FSM for 95 taurus. It provides some useful expected amperage draws. They say to use a test lamp, and if it glows, there is too much draw. Are they really talking about the typical automotive test lamp with a clamp on wire, sharp pointed tool with a neon lit handle? It's not what's in their diagram. Or, as in the picture, is it just an ordinary 12v bulb? What the heck is the shunt they suggest using and can I get one or make it or do I really need it? Is it a preset resistance that current goes through that you measure the voltage across it to give you the current? Why can't I do something like that without the shunt. When I measure between the neg post and neg clamp, it's about 11 v. If I then measure the resistance from the neg clamp to the pos clamp, can I use that info to calculate a meaningful amperage?

Well, scrap everything. First, it seems to be behaving itself now for two days. Secondly, I must not know how to use my multimeter. Actually, I think something is wrong with it. I was getting false readings on the low amp setting and thought I didn't register anything or enough to use the 10 A separate jack hole setting. Anywho, basically, I'm measuring 90 mA at steady state. When first putting the test probes on it, the current can spike as high as 1.35 A. Just to verify the accuracy of my measurements, I also turned on the interior lights that drew about 2.5 amps. For comparison, I measured the current draw on my Windstar and found it to be 0.5 amps. The weird part is, I was basically unable to measure any (oc) or a meaningful resistance. When I could get something, it would keep descending or go all over the place when I changed tje tester range. I assume a capicitor is causing this. I just don't understand how you have a constant current draw with no resistance. Conversely, if I have V and I, shouldn't I have R? At this point I will let sleeping dogs sleep. Thanks Rod.

Well, you won't be able to measure series resistance using a meter in the system like that. The meter is trying to do the R = V/I calculation with it's own internal voltage and current source. However, since you have the meter in series with a battery, the battery is now providing a voltage potential difference from the meter's regulated voltage. The meter needs two parameters to remain known to calculate the third. In this scenario, there were two unknowns and your meter didn't know how to deal with it (voltage and resistance). In the case of making current measurements, the meter switches in a known low resistance load and measures the voltage drop across that resistor to calculate the current through the resistor. It knew everything it needed to to make that calculation.

It does sound like your current draw is a bit higher than typical in the Taurus.

The draw in the Windstar sounds about right, so I think your meter is working and you're making the current draw measurement correctly with the correct scale. But as you verified, the current draw you're experiencing is too low to be a light bulb left on or something. I think even the glovebox or trunk light would draw more current than 90 mA. That sounds like an active electronic device. Any chance there's an aftermarket alarm installed somewhere under the dash that you're not aware of?

What about the audio system, is it one of the uplevel models with an external amplifier?

-Rod

Thanks Rod for your continued response. Uh huh on the V I and R explaination. I'll take your word for it as I don't presently have the mental fortitude to follow it all the way through. I reread what you said about the meter not able to measure resistance in the system like that. Do you mean the resistance can't be measured with the meter at all with the type of circuit it is or where I have the meter while measuring current? To take the resist. meas, I put the leads across the neg and positive clamps. I surely thought that would give me the circuit resist. Then I would use bat volt for V when all hooked up. Right? My test lamp is 25 ohms. Can I use that for finding I?

Come again on the amp draw? The Windstar pulls 500mA and the Taurus pulls only 90mA and you're saying the Taurus' draw is a bit high? What about the Windstar? It however does not have a drawn down battery problem. I reread the FSM, group 14, battery and charging system, and found two threshholds. First, see if there is more than a 100mA discharge. Then the rest I will quote from it:

The current reading (current drain) should be less than 0.05 amp. If it exceeds 0.05 amp it indicates a constant current drain which could cause a discharged battery.....
If the reading is between 0.2 and 0.9 amp, a possible drain may be a vehicle lamp (glove compartment, engine compartment, luggage compartment, etc.) that does not turn off. If the cause of the drain is not a lamp, remove the fuses from the fuse panel one at a time until the cause of the drain is located. If drain is still undetermined, remove fuses one at a time at the power distribution box to find the problem circuit.

Done all of that. The 90mA is clearly in the range to be concerned, but the Windstar is more than 5 times too high. I can remeasure everything, but I'm gonna probably let this rest for awhile. What throws me off, is the supposed relationship between the electronic temp controls and the remote keyless/ light module. With them both disconnected, the large drain goes away. Just one disconnected, drain remains.

Too tired to remeasure the van current, but the Taurus remains at 90mA. The battery voltage also suggests the drain is affecting it as it is 0.2 volts lower than previously. I checked again for after market security systems. Nothing. No decals or old blinking lights. Any particular place they often are located? I looked all around the steering column with covers off and under the dash on both sides. Saw no wiring splices. All original. Radio is original with its own amp. Since the two components seem to be somehow related in this, I'm thinking there's a ground issue, but even that seems a stretch. That will require more FSM research and seemingly pointless searching. :screwy: :shakehead :banghead: :runaround: :mad: :crying: It might be quicker if I just installed switches to the EATC and RKEM/IEMs. at the loss of the RKE. It's not clear to me yet, looking at the diagrams for the EATC, what the power supply to it in question does as there are two. I imagine it provides memory power as it is on all the time, whereas, the other is on only with the key. I hate conceding defeat.. Swapping out the EATC and the KEM/IEM with used ones doesn't seem reasonable. I'd expect them to be expensive and just guessing sticking them and possibly ruining them too. I don't like my dwindling options.

My mistake, I read the Windstar current draw as 50 mA, not 500 mA. Yes, half an amp draw is excessive on the Windstar. I wonder if you caught that with the accessory delay relay still energized. If you left the current meter in series for a minute or two, I suspect that draw would go down. Either you were reading the the accessory delay relay, or the capacitance of the system was still charging.

Speaking of system capacitance, that is half the reason you can't just measure system resistance by measuring across the (disconnected from battery) battery cables. Multiple loads on the charging system are active, not just passive. The system will have resistive, inductive, and capacitive components to it. There are numerous semiconductor devices (transistors) in the active components (such as voltage regulators, computers, radio, EATC, RKE, etc. -- integrated circuits). These devices will behave differently than a purely resistive component and may start to "turn on" due to the small current provided by the multimeter in the "Resistance" mode, causing the reading on the meter to jump all over the place.

The bulb in your test light is inductive, so it will change in "resistance" as the bulb filament heats up, and the amount it heats up will change as the current draw increases, etc. So using the DC resistance of a test light to calculate current is not really the best route.

I'm with you on the wiring diagrams not helping make sense of the current draw, but 90 mA isn't too far off. You may want to leave the meter in series with the battery for a couple of minutes and see if it remains 90 mA or possibly drops lower than 50 mA after everything goes into standby or shutdown mode.

-Rod

That sounds reasonable that the current draw with the Windstar may have not had enough time to get to the "sleeping" mope. The Taurus I have had the amp meter in series for a day, periodically checking it, with it remaining unchanged from 90mA. I'll revisit when Hanna and Gustav get out of here, and I have time to persue it more. Thanks for your time and explainations, Rod.

I am still concerned about the drain even though it still remains 90 mA as the battery voltage is being depressed from normal. Is there another accurate way to measure the amount of current drawn while the cables are connected? I don't have a clamp on meter that measures direct current, just ac. I'm suspicious that the current jumps when the system senses a powered circuit that maybe wouldn't energize with the meter inbetween. Some of my suspicion is raised by what I measured on my Windstar. On it, when first hooking up the meter, I get 10 mA for about a second, then it jumps to 450 mA. It stays there for quite awhile, perhaps up to 30 minutes, and then drops down to 250 mA. The battery does not get drawn down. It maintains 12.5 volts no problem.

Rod, you spoke of internal battery problems. How can I tell if I have that? It does measure around 12.3 volts. I measured the voltage between cells to be around 2 volts each. Interestingly, between the first cell and the neg post, I got around 1.8 v and 0.8 v between the last cell and the pos post. Any of that sound like I have a battery problem? BTW, there are 6 cells, so there are seven inbetween places to measure including the posts.

Rod, did you ever measure the "sleeping" current draw on your Mountaineer? BTW, do you know about the recall on it?

Interesting, I've never tried measuring between cells. Since the current flows "through" the battery, I'd expect all cells to be pretty even, but having never tried this measurement or really even thought about trying it, I'm not sure what a good battery would show. The only other really accurate way to measure the current would be to install a series resistor of very low resistance and high power rating (maybe 0.1 ohm, 10 W) and measure the voltage drop across it, then use Ohm's law to calculate the drop. That's exactly what your meter is doing though. And since the meter should be fused, it's going to allow as much current to flow as the fuse is rated for. If the car is trying to draw more current, it will blow the fuse and you'll know it. So if the meter's fuse isn't being blown, the car isn't trying to draw more current. Where you could run in to that scenario though would be when using a test light in series with the battery.

The only internal battery problem that would not be intermittant would be excessive sulfate formation, but there's no real way to monitor that other than measuring the capacity of the battery (or load test). The measurements you performed would give indications it would seem to open or shorted cells if they happened to be open or shorted at the time of measurements. But, if they were open or shorted at the time, the battery voltage would have been something less than 12.5 volts I'd expect. Although your cell connected to the positive post would be close to being shorted.... A battery issue shouldn't affect current draw of the car though.

I'll try to measure the draw on my Mountaineer tonight using a series meter AND a clamp on DC current probe just to see how the numbers compare. And as far as I know the recall work has all been performed. Most recently was the cruise control wiring.

-Rod

What kind of internal defect or damage to the battery did you mean when you have mentioned to others about that posibility. Is that the excessive sulfate formation you mentioned? Or, is it something that would cause a more inconsistent battery problem? Thanks for your input. Unfortunately, I won't be able to keep you updated for awhile on my progress or lack thereof, as I'll be out of town for awhile and doubt I'll get to go online and won't be able to pursue this either then. When I do have more time, I'll go try measuring the cell to cell and cell to post voltages of other batteries for comparison. The down side to doing these measurements, the acid eats the silver coating on the probes, so sacrificial secondary probes may be a better way to go, like copper wires in aligator clamps. Live and learn. The low ohm resistor with high power rating makes me think of the shunt I mentioned from the FSM. What was neat about it was the resistance would produce a current and voltage drop of the same numerical value, avoiding doing any math.

The recall I heard about was announced in the past week or two, and I thought it was new . Didn't want you or your family to become another crispy burn victim of Ford's faulty electricals.

Alright, I took some current measurements on my Mountaineer which I don't have a problem with the battery draining unless it sits for a couple of weeks or more. For what it's worth, I do have an aftermarket alarm with remote start installed. These measurements were with the alarm in standby mode (passive arming). Initially I didn't trust the readings from the current probe since the full scale of my probe is 600A. I assumed that a reading of 200-300 mA was outside the reasonable resolution of the probe. So I installed a current meter in series with the negative battery cable and found that those readings are comparable to what the current probe was reading. It seems like 7 minutes should have been sufficient for the systems to go into standby mode, but I did not see a significant drop in current from the time the headlights went out (auto lamp delay timed out) to the time I stopped the trend plot (> 7 minutes later). The series meter readings were only plotted for 2 minutes. I also found it interesting that when I went to set the radio clock after having the battery disconnected, the time was still correct and stations were still there. The Distance to Empty and Average Fuel Economy was cleared out though.

http://i538.photobucket.com/albums/ff342/raschmidt/Automotive/current_profile.jpg

http://i538.photobucket.com/albums/ff342/raschmidt/Automotive/current_profile-2.jpg

http://i538.photobucket.com/albums/ff342/raschmidt/Automotive/Series_Current.jpg

As for the battery defects/damage, sulfation could be a problem that would contribute to battery capacity issues. The other problems would be open or shorted cells. These wouldn't be detectable unless they were a problem when you were making measurements. An open cell would cause the battery to not even crank the engine. A shorted cell would lead to low battery voltage, low capacity, and probably higher than normal self discharge. The cell reading you measured of 0.8 V would seem pretty close to a shorted cell....

-Rod

Impressive meters and graphics. Must have set you back a couple. It would seem I don't have an excessive drain on the battery when all is shut down with it at 90 mA and the van at 250 mA and yours 200mA to 300 mA. Since the van battery has a higher voltage and a higher current draw when in sleep mode, that makes me suspicious of the battery, especially with the one low cell reading. I will pursue this more when time permits. Thanks for your input.

Well, I did spend a fair amount on the test equipment, but I use it pretty often for the electronics repair hobby as well as these automotive issues.

I don't see how I could get down to the current draw that Ford says is normal or what I thought was normal (70 mA) unless my aftermarket alarm is drawing over 120 mA itself. But, since I don't have problems with the battery being drained, I think I'll leave well enough alone for now.

-Rod

Hey Rod. Ford's allowable draw is printed as .05 amp. But this can take from 30 minutes to an hour for the system to completely shut down. I allow up to .075 on our vehicles due to the extra equipment installed.

Hmmm, I don't know that I'll leave my meter connected and logging for an hour unless I start to experience a problem. But, that could certainly help explain the draw that Tripletdaddy and I are measuring. Thanks for the tip!

-Rod