I V Watts.
Hi RB, That chart relates to testing open circuit (no load). Believe me, a battery measuring 10.5V open circuit is flat.
The reserve capacity test and AH rating is done loaded and based on how long the battery can supply an amount of current whilst maintaining min 10.5V. That is a different thing as the battery voltage drop is due to load/time rather than state of charge. Likewise the CCA test is done loaded down to just 7.2V. I found this explains it well, http://www.aa1car.com/library/2003/us20310.htm
I took Gerryn to mean the load would remain constant W, as it would with an electric motor or the solenoid coil, and so the current would increase with lowering battery voltage.
Yes, omhs law. R= v/i, if R remains constant, as v falls, i falls.
In your example of the battery powering a bulb, the resistance does not remain constant as the voltage falls. As the bulb filament cools with lowering voltage its resistance falls as well. Whether the current goes up or down depends on the temperature versus resistance characteristics of the filament.
Alan
Hi Alan, yes as you and RB say about Ohms Law but we don’t know the nature of the load.
I think Gerryn must either be thinking of
a constant load such as an electric motor would provide, in which case the current would rise with lowering voltage, Watts Law,
or a very much heavier load involving resistance losses in cables which wouldn’t matter here at mA levels,
or again a heavy load involving the internal resistance of the battery such as tested during CCA tests.
Whichever, at this small drain there will be very little difference in the number of days taken for the battery to go flat and it is not an issue.
Smiling - - I really got your brains working there didn’t I
Yes my suggestion it was Ohms law that ruled was incorrect as you all pointed out. However, I did give you a clue earlier on this thread, when I said that the PCM and other things were permanently connected to the battery. The other things are - the Immobiliser and the alarm. We don’t have internal info on any of these, but modern technology suggests they are all solid state devices, we know the PCM is so it follows that so is the immobiliser, if you recall, it has to be coded into the PCM, and no plain electrical device can do that. In addition, when the alarm is set on my car there’s a constant flasher light on the radio, and also on the instrument panel - these flash at different rates which suggest they are on separate flasher (solid state) circuits. Anyone who has built a simple flasher circuit will know it’s standard practice to rely on the combination of a charge resistor and a capacitor to drive a transistor which in turn provides the output pulse to flash a lamp, or a relay - these are running all the time the alarm is on. Minimal cuurent - yes, but they still are part of the load.
Why is the PCM permanently connected? - To retain the temporary registers that relate to a few key elements, which in turn relate to the way you drive the car. Any computer has a scan, that moves right through all stages in turn - inputs, program and outputs, any changes along the way are stored and transferred to outputs as the scan goes through them. In the event of battery disconnection, then there’s a small backup battery which tries to maintain temporary registers, as it discharges, the PCM reverts to default settings. This also is why you can clear a fault stored in the PCM, you dis the battery neg, and depress the brake pedal (brake lights are also permanently connected to the battery) This discharges the backup battery so any fault codes also in temporary registers revert to default (e.g. - off) Now - I can’t prove this, but having programmed PLC’s for years I know this happens on them, and it happens on computers too - if your backup battery goes, a lot of basic info is gone too, with the exception of those details which are written into permanent Bios (Eprom) memory. In later years, PLC’s have been designed with Eprom storage throughout, so they have no need for backup, but AFAIK this hasn’t yet transferred to PC’s YMMV.? If there’s a programming fault in any device, then the program will stop till that error is cleared, which is why a recent comment about the “known fault” in a 2005 Mk2.5 PCM didn’t make sense. That’s also why any duff software on your computer can ‘lock up’ - the computer cannot continue to run that program.
I would add - I’m no solid state expert. nor a software genius - I have my own ‘Eprom’ which permanently registers details about anything I observe, though with the passage of time, some of those entries get a bit blurred! - You could say my backup is running down!
However, let’s take the figures we know and multiply them over a two week (14 day) period. - We know the normal dark current (in the car - not the battery) is .020 per hour, per day that’s 0.48 amp, a week is 3.36 amp, per fortnight is 6.72 amps. That’s over 20% over the fully charged capacity of a 32 A.H. Panasonic (OEM) battery or the Westco at 31 A.H. My argument is that with the constant load of the PCM - even with the immobiliser and alarm not operative - supposedly with a current drain of 0.015A, (5.04 amp after 14 days) will still drag the voltage down below 12.5 volts. Remember we aren’t just carrying the load of the temporary registers, we are also keeping the backup battery charged.
One thing that hasn’t been reported back yet IIRC is the charge voltage from the alternator - is this functioning properly? That should be around 14.5 volts, but we also don’t know what running periods (every forthnight) are? - Does the car get a good run, or just a run-around? We need some positive readings here, or the subject will go on for ever. If the actual driving time is short, then the battery will never get fully charged - it’s on a ‘downer’ every week. If the charge voltage is low, then we could be trying to diagnose a problem that could be down to other things, like a slipping drive belt, or even a faulty alternator. Another thing that’s being overlooked on this thread - most vehicles have a 45 A.H. Battery, some even larger, I believe I saw a 75 A.H. battery on something in the past - and it wasn’t a bus! When I first checked the car over, after driving it from the dealer- I was surprised to see such a low capacity battery. The whole point is an MX5 was made to be driven, so storing it for extended periods isn’t going to do the battery any good over a period of time, if it never reaches a fully charged state, then it will adopt a smaller voltage, even batteries have memory (of a sort) ! That theory backs up my own experience, as after ten years, the car has around 36K on the clock. - Mainly down to a lot of medical problems since the end of 2008. Apart from short trips (around 5 miles round trip) the only longer ones have been to Chatsworth for the National, and a trip up to Doncaster (about 80 miles round trip) to pick up some cable. That’s mainly why I’ve now bought a new Panasonic from a local dealer, it’s lasted ten years, so it’s time it retired - - I wish I could! There’s a definite inmprovemnt at startup, but |I’m not going to test it without a tricklecharger for another three months - that’s asking for problems. Cars that start well after a two week holiday abroad are usually driven in the week, or achieve good mileage over a week end, which is the best way to keep the battery healthy IMHO. We need food, and so does a battery. Nuf said.
Question for Touranman - are you the person who posted a comment on MX5 parts that a Westco battery was “unfit for purpose” ? - I really think you should withdraw it if so - lots of members have the Westco - and they don’t have any problems. I opted to replace the OEM with another Panasonic, after ten years of ‘aggravated service’ it was still good, but I felt that that length of service warranted a replacement of the same make. If it’s good enough for Mazda, then it’s good enough for me. Whoever made that comment, then it didn’t help me make my mind up, but I felt it wasn’t warranted either.
Gerryn,
No, it wasn’t me who posted the comment re. Westco. At the moment there is no evidence to suggest that the battery is at fault.
Thank you for all your other comments. I am currently undertaking a series of voltage measurements on the battery, before trickle-charging, after trickle-charging, with engine running and alternator re-charging, after a few days standing etc to see if this yields any information on what is causing the problem.
Dealing with the hard facts, I think I can say that the following statements are true:
a) the ‘dark’ current with the boot lid closed, alarm and immobiliser off, is < 20mA, which from other comments appears to be normal and within Mazda specification
b) several responses indicate that people can leave MX5’s unattended for longer than 2 weeks (in some cases months) without the battery going flat.
Based on this the next avenue of exploration should be to confirm that the battery is still sound, and is being charged correctly - as you and others have pointed out.
This is why I am undertaking the series of voltage measurements mentioned above.
This is the first time I have used the MX5 OC forum and I have been amazed and very grateful for all the responses and help offered. I will report back when I have more data.
Another variable in the equation is that the dark or back up current is measured with everything off, if the car is then locked and the immobiliser/alarm/central locking etc. are in use then the current draw will be somewhat greater than 20 milliamps so to calculate how long the battery will last the current draw needs to be measured with the alarm etc. on. As Albert Einstein said, In theory, theory and practice are the same, in practice they are not! I connect a maintenance charger to my battery once a week when the car is not in use. trying to calclate how long the battery will last involves so many variables only an approximate value can be obtained, the condition of the individual battery will make a difference.
Regards Geoff Peace.
If you are using your car for work each day then I do not think you have a problem. It will be charged every time it is driven.
Regards Geoff Peace.
To be honest - I’m surprised you’ve stayed with it! Differences of opinion between us geeks can be frustrating, so on behalf of everyone, I apologise for our meanderings! I also offer my personal apologies for mentioning that post on MX5 Parts, though the circumstances were similar.
Re measuring dark current, can you also check the current draw when the alarm and immobiliser are on please, I’m curious to know if it’s equal to mine. To obviate any spurious draw, if your meter leads are long enough, open the boot lid first (which way you like) and once it’s open, if you take a fairly broad flat screwdriver you can push down the boot latch, which switches off the boot (number plate) light. When you’ve finished checking, don’t forget to relase the latch before closing the boot lid, - press boot lid on the fob. or pull on the internal boot release - or you’ll get a horrible ‘Clunk’ ! However. as doing this puts the alarm in a ‘stage 2’ (which may draw more current than normal,) it’s really advisable - if the meter leads are long enough, to connect the leads, hold the meter outside the boot, and close the lid. At this point you can switch the alarm and immobiliser on, when you should get the total draw in a normal alarmed state. Make sure the leads aren’t near the latch, otherwise they could get trapped.
May I also suggest if you connect a trickle charger, that you close the boot lid while it’s working, even with the car in a garage, things can happen, if you have a dog or a cat (especially) and they get in the garage, you never know what they will do, or where they will go. Also, I’ve read so many times of things falling off shelves - one of the hazards of a garage, so closing the lid will negate any outside interference. As I think I said before, if you run the charger leads (the 12v ones - not the mains) to the right of the boot latch, then the rubber surround on the top edge of the boot will cushion the leads safely, so there’s no worries about worn insulation or short circuits in those leads. My car has sat for at least three months with the charger connected that way - and to my amazement no-one seems to have noticed, even my wife, who is meticulous to a fault - - - - I mentioned this to someone I worked with, and he said “If someone did notice it, what would they think?” and I replied “Hopefully, they’ll think the car is alive, and won’t go near it!” That’s always been a tempting thought!
We’ll keep an eye out for you report, and thanks for staying with us! I hope one of us can come up with a definite conclusion.
If I may be permitted a word of caution when testing. Ensure the meter is on the highest amp range, as when the immobiliser/alarm is set the indicator lamps will flash and around 7.5 amps will flow through the meter, only for a fraction of a second but enough to blow the fuse if on the lower ranges. Also some meters are not fused and even if they are the fuses are seldom standard and can be difficult to scource. I will try to take some readings on my car this weekend and post the results.
Regards Geoff Peace.
Yep, good point Geoff, I wouldn’t have thought of that and would’ve just stuck it on the 200mA range and “phut!” as you say. Blown a few and you can get the fuses at Maplins but a pain as you often have to take the case apart to fit it 
I have a Mk1 & a Mk2 and have the same problem with the 1998 Mk2. The 1 can sit all winter and start first time. The 2 stood for 1 week and is flat. New battery, charging fine (according to the voltmeter). Dunno what to do next really…
How do I get my multimeter in the charging circuit to measure the current? Obviously it has to be in series?
thanks.
It may still be that the new battery is not charged to full capacity. It can take hours on a run. It will start the car if only 1/4 charged capacity-wise but will then go flat after a week instead of a month. So give it a good long trickle charge for a day or two first to be sure.
Assuming it’s the key-off standing current you want to check (if it’s the load/charging current you’re after you’ll need a bigger meter capable of 100A+, typically a clip on inductance type).
Firstly, it’s very easy to blow one or both meter fuses doing this. Even switching a door light with the meter connected in the mA range may blow the meter fuse so take your time as it’s a pain to replace.
Set meter to its highest Amps range and put the probe leads into the correct Amps sockets (typically there are 200mA and 10 Amps ranges, black stays in COM and red moves from Volts to Amps, careful, there may be separate mA and A sockets. Go for the highest range first to avoid blowing the fuse on the low range)
Turn everything off inc boot light.
Remove neg/earth terminal from batt and clip meter in series between the clamp and batt post. (It doesn’t matter which colour lead goes to post or clamp, you may get a minus -Amps reading one way but the value is all that matters.)
Check Amps and for a more accurate reading remove meter and reset probes to lower range if the higher range check says it’s ok to do so Give it a minute or two to settle/alarm/ecu to set, and check the drain current, expect the order of 20 or 30 mA.
If I may be permitted a correction to the last line of Rich M’s post. The maximum drain/dark current is 20mA. my own car is around 14mA. 30mAs far too high. At the 20mA level, in theory, the maximum level of drain will take 0.5 Amp/hours per day from the battery. It does not sound very much, leave it ten days or so and the engine may well struggle to start. To reiterate a point I made earlier in this thread, the dark current reading, if it is to be meaningful, should, in my opinion, be measured with the doors locked, the immobiliser/alarm set, etc. All these will only add to the current drain. If the reading is taken under these conditions a more realistic and meaningful view can be taken.
Regards Geoff Peace.
Hi Geoff, Yes Mk2 manual says, max 20mA, so your 14mA is a good example of what to expect.
Mk3 says, average 30mA.
Mk1 will vary depending on imobiliser or not
I only gave typical figs off the top of my head, the point being that Tom’s looking for a reason for the battery being flat in a week instead of a month so if he finds an obviously high result such as 100mA or more he’d be looking for a fault such as failed diode etc.
If he doesn’t find a high drain then it’s back to looking at the charging system or checking if the battery was simply not fully charged when he left it.
What in the name of all that’s holy happened to this thread? 
I hate to correct Rich some more, but if you’re unsure of the current you’re about to measure, don’t use the very highest amp range of your multimeter (the 10A range) as this is usually unfused and will break for good if it turns out to be, for example a short to ground. At least on the 2000mA range you have a second chance…
I have had similar flat battery problems with both a Mk3 Sport and my current Mk2.5 when leaving them unused for several weeks. Do not get the same problems with either of my Fords which will start immediately after being stored for several months.
Hi Jon, most meters have a higher range of 10A, 15A or 20A so I would start there as in this instance we’re looking for a possible drain flattening the battery in a week not a few hours so it’s a safe call imho.
Protection will depend on the meter, I know I’ve changed both fuses in mine, see below, 0.2A and 20A fast blows,
http://www.digitek.com.hk/en/cpxx.php?id=719
Part of the problem is the battery capacity - most cars carry a 48 A.H. battery - I was rather suprised when I saw the MX had a 32 A.H. My 2.5 (2002) battery gave me similar problems after a few years, especially when I had to lay the car up twice over the last two years. While most of my battery experience was with lead acid cells, I have a feeling that if a battery is not maintained fully charged over a period of time, then it won’t retain a full charge, but tends to accept a lower one. As a lot of my driving these days is local (cost of petrol dictates that) then eventually I bought a new Panasonic a few months ago, but as it appeared to be fully charged when fitted (I checked the voltage) then so far it’s held up well - as it should. Batteries can be on the shelf for months, so as Rich M suggested, so a trickle charge to start with does no harm. You didn’t mention which type of battery you bought, but as the Panasonic lasted ten years on my car, I had no hesitation in buying another, despite the cost. You get what you pay for.
Couple of pointers on connecting a meter - 1) Can’t remember offhand if the Mk2 has the same dual purpose boot/license plate lights as the 2.5, but if it has, then the boot lid lock controls it. Once the boot is open, get a largish screwdriver (slot) and depress the catch on the lock - this disengages the boot/parking light - otherwise you’ll get a higher reading with the meter. Don’t forget to release the catch before trying to close the boot lid afterwards, either by the key fob or the manual release, then you can close the boot lid. If your remote locking system is more or less same as mine, then you can now lock and alarm the car fully - if you don’t activate the boot lock then the alarm is at first stage only, and may draw less current. but the boot/license plate light may be on, which will draw more current!
- Meter leads, presumably you have one red and one black lead - the black lead should go to COM and the red in the marked socket for current reading, if the meter has a 20 amp socket start there, if it’s only a 10 amp, then there. Disconnect the Negative battery lead, and connect the red lead to the disconneted battery cable, then touch the black lead to the negative battery terminal while watching the meter. If the meter needle shoots rapidly to the maximum point pull the black lead off the battery immediately (you’ve got a problem!) If it doesn’t, then as long as the reading is very low on that scale, then you can start dropping (moving the meter knob) the current level setting till you are able to get a decent reading. As has been said, the normal reading as listed by Mazda should be 20ma maximum. Most cheaper multi-meters don’t have a low enough scale, so you may have to use either a 100 or 200 ma scale, so each division mark on the scale will represent a fraction of the total. Count the marks between each number, and mulitply them by the number of numbered marks - mine has a scale of 1 to 12 (12 amps maximum) so each mark represnts 1/12 of the full scale reading depending on the current setting. i.e. - meter reading of 2 on my scale with knob set at 50 ma is 1/6 of the maximum reading = 16.66 ma. From this you can see it’s difficult to get an accurate reading, you just have to guess from where the needle sits between divisions that it’s about right, or it’s too high. Your reading will depend on the scale, you need to understand what each division means at the meter setting. Cheaper multimeters also have limitations, the tolerance (accuracy) can be anywhere between 10 and 20 %, and you have to judge accordingly. If you still have the manual (such as they are) then check what the accuracy (tolerance) is if you are in doubt. Also, check the needle setting is right on zero before you start checking, adjust the separate knob or the one just below the meter itself (usually a screwdriver slot) till the needle is right on zero. Adjust the latter carefully - it controls the spring on the needle, and can damage it if you get too vicious!