Measuring Initial Car Battery Voltage Help Needed

pblakeney · 11-03-2009, 11:15 AM

I am designing for the fun of it a dual battery system for a car that is automatic. One of the features that I would like to incorporate is dual battery charging. However, if one of the batteries is shorted ( I am assuming for now that it will have less than 10 volts if atleast 1 cell is shorted) then I do not want that battery to be charged because it would prevent the other battery from charging properly as most of the current will go to the shorted battery.

My problem is this. I want to look at the voltage on both batteries before the engine is running. As I understand it, if I wait until the engine is running I will be seeing the alternator voltage (not the true voltage of the battery). I was going to isolate each battery and use a comparator with a reference voltage of 10 volts (maybe even use a voltage divider to get it to 1 volt) and then compare the reference voltage to the battery voltage. I was going to do this for both batteries. I would then use the ouput of the op amp or comparator to bias a transistor which would be hooked to a relay that would engage and disconnect the battery from the charging circuit. Somehow I think I need something that would see the battery voltage upon an intial signal like going to the start setting of the ignition and then storing that value or hold that value to the op amp so that the op amp does not unbias the transistor once the car engine starts. Does anybody have any ideas how to do this?

Or, is there way to read the true voltage of a battery when the engine is running? I do not see how since the alternator and battery are in parrallel?

Bugbyte · 11-03-2009, 11:37 AM

How about measuring the current going to charge each battery instead? If too much current is flowing to one batt, then cut it out of the circuit.

pblakeney · 11-03-2009, 02:53 PM

Thanks. I was considering this. However, I also started to realize that output current increases with engine RPM and therefore the battery would naturally draw more amps (I think because the alternator would put out a higher voltage and create a greater voltage potential). Also I am not sure what a normal charging current is and if it varies depending on the batter size etc. I might be able to research what a limit is for too much current if the battery has a short. So many variables to keep track of but I think that you are on to something. I am just trying to figure out where to start on understanding how the alternator works and how it senses.

Bugbyte · 11-03-2009, 04:28 PM

I'd say find out what a normal charging current is and use that as a baseline. I don't think it's more than 5 or 10 amps. At least that's what my battery charger reads when I use it.

Try running a battery down and measuring the charging current to it?

justchat_1 · 11-03-2009, 04:32 PM

Most cars have fusible links inline with the alternator...if theres a short its not going to matter what control circuitry you have, the fusible link will have already melted and you'll be in a bad place fast. Since you can't detect a short until its occurring the only way to do this is to remove the fusible link which is INCREDIBLE UNSAFE.

I have to wonder what makes you so concerned a short will occur at the battery level?

pblakeney · 11-03-2009, 07:12 PM

I do not know what will actually happen if one battery shorts out. I was really not concerned that the shorted battery will draw damaging currents but rather that the battery will draw the other battery down and prevent it from charging so I was just trying to determine a means of detecting that type of event and preventing it from happening or to stop it once it has been determined that it is happening. I have had a couple of batteries short out on me and I could not even jump start the car. I had to remove one of the posts on the battery to bypass it. I then drove to walmart and bought a new battery right then. I remember thinking that when jump startring the car (when they are both in parallel) that the reason the car would not start was because the shorted battery is absorbing most of the current from the donor car and that there was not enough cranking amps available to start the car. I did not observe any damage to the donor car or mine because of the shorted battery only the fact that the donor battery could not start my car. Therefore I am thinking if the same thing happens with a dual battery system then the good battery will not get charged. This is only my theory and I am not saying that I fully understand what is actually going on. If anyone knows and is willing to share I would be so grateful!!!

pblakeney · 11-03-2009, 07:22 PM

I like your current monitoring idea. I will have to look more into that. I spoke with a gentleman today from (I won't say as I am not sure if this will violate any of the forum rules) but he mentioned that a peak recharge current with a drained battery could draw 40 to 45 amps at 14-14.5 volts. I know there is temperature compensation that affects the alternator output current and engine RPM etc but there must be a middle of the road limit I can use. He also mentioned that the voltage would be really low on the shorted battery like less than 10 volts and drop fast. If I could take a snapshot in time just before the engine is turned on and store that value maybe through a one shot circuit right when the ignition is turned on and that value is compared with a refernce voltage via an op amp and then that output is stored in a register and then if a certain bit is high then the relay that engages a battery will not engage. Something like that. I kind of have a general idea but not exactly how to make it all come together.

justchat_1 · 11-03-2009, 08:14 PM

Well i'm a computer engineer so I have a good amount of experience with this stuff....

What your describing sounds more like a dead battery then a shorted battery. Having a short in the system would most likely result in a fuse blowing...it would have to be a short between the battery and the distribution block to cause an actual battery short. Technically the short could also be between the battery and the voltage regulator but this would have fried the voltage regulator.

My guess as to your situation:
If a battery is run below a certain voltage (referred to as deep discharge) then supplying high alternator current will not recharge it (you need to use a special battery charger which strictly regulates a linear recharge). Should a battery sit in this discharged state for awhile it becomes unusable (don't remember the chemistry, think it had something to do with plates oxidizing?).

What your trying to do:
You need to take account of quite a few variables in your design:
1) When cranking there is a 100-300+amp load on the battery with voltages dipping as low as 6v in some setups.
2) After ignition voltage spikes as load decreases and then increases and the alternator tries to compensate.
3) Stereos, headlights (8-10A), A/C blowers(20-30A) can all cause sudden current draws which which are sometimes provided by the battery until the alternator can increase its output to compensate.
4) On some cars the alternator is deactivated at WOT relying solely on the battery for a few seconds
5) Charging current can vary greatly depending on temperature, battery size and charge levels
6) Alternator current can vary from~0-full rated current depending on load.

Your most likely going to want to use a microchip to provide the logic (anything using comparators or states is going to get expensive/complicated). There is also the potential of using a current regulating diode setup but again by the time a problem is detected the battery is already discharged.

pblakeney · 11-05-2009, 12:34 AM

Hey thanks justchat_1! I guess what I am trying to describe is the situation for example where when one moment you get in your car and the car starts fine and then drive to the store and then after you come out of the store the car won't start, no lights come on etc. It is like someone took the battery out of the car. I had this happen to me on three seperate occasions. However, nothing in the car was damaged. I am still driving the same car and it has not happened in the past two years now. It seems that car batteries these days just die without any advanced notice. A long time ago you could tell your battery was going bad by how sluggish the car started and the lights etc. I have heard many people with the same situation that I have experienced where it starts the car like a good battery one moment and the next time it is completly dead. I have had to bypass the dead battery just to get the car jump started on each occasion. That tells me that the dead battery is diverting most of the cranking current to the dead battery instead of having it available to start the car. In other words it is absorbing most of the current supplied by the donor car's system. I have never encountered any cables or parts being damaged during that process though. I don't think that it is a short in the sense that there is fault current occuring in the battery although I do know that there are situations that does occure, just not in the scenario that I am talking about. I think it is more like atleast one cell breaks down some how leaving a low voltage on the battery. So you are correct of the assesment of what is going on!

One of the things that I just thought up involved monitoring the voltage of the batteries when the key is in the ignition and turned to the on position and then look at the voltage level and use a latching relay configured so that a bad battery would not be placed in the charging circuit. I could unlatch the relay by going back to the off postion (using another realy) of the ignition. I think that this might work, I hope! I am still hoping others might offer some more advice on this subject.

justchat_1 · 11-05-2009, 12:45 AM

Is this project on the same car that was having the problems? If so, whats the year, make and model? Also, throw a voltmeter on the battery terminals when the car is running-whats the voltage?

What you describe sounds like a dieing alternator (not dead so it still charges on longer trips but on short trips it doesn't make up for the cranking loss and kills the battery). Could also be a lose connection from the alternator which does the same thing. Voltage should be pretty close to 13.8 but I can be more exact once I know which car.

Bugbyte · 11-06-2009, 07:18 PM

Yeah, sounds like a dying alternator. Check the voltage, it should be near 14.

User Name		Remember Me?
Password