Automotive electronics - Honda charging system

(Personally I'd cobble up my own regulator ;-)

US 'CAFE' fuel economy standards are forcing car manufacturers into Rube Goldberg 'solutions'...

Recently was provided a 'loaner' of a new Infiniti Q50 while my ancient, 'real car' (12 year old) Q45 was in for service.

This Q50 had only a 2L 4-banger engine (provided by Mercedes no less).

At each traffic light the engine is STOPPED with a jerk, then restarted... feels like a diesel start (compressed air ?) :-(

What a disconcerting piece of crap!

(I informed the dealer I wouldn't be buying any Infiniti product until they came back to providing a honking V-8 again, like my Q45. He sighed, said he agreed.) ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 

             I'm looking for work... see my website. 

     Thinking outside the box... producing elegant solutions.

There's a 'little' TC issue that must be met to ensure proper charge...

Also see my patents about alternator regulators. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 

             I'm looking for work... see my website. 

     Thinking outside the box... producing elegant solutions.

Thanks very much. Well I don't see anything on your graphs anywhere near Honda's 12.4V. It looks like you have an absolute minimum of 13V even at high temps. And I suspect if my Honda followed that, there would be no problem.

Charging voltage aside, is the "off" voltage of a fully charged battery also temperature dependent - assuming you give it time to settle? So would the 12.6V I've been using as a fully-charged voltage be lower at higher temps? Is there a graph for that?

Thanks Peabody and all others for your comments.

My problem with the 2016 Camry was probably the result of driving the car o nly once every week or two without any long trips. That assumes that the v oltage regulator has been working as intended. The charging system might h ave been designed assuming several trips per week and a mix of short and lo ng trips.

The parasitic current drain likely becomes a factor with infrequent use. I f the parasitic current drain was 0.02 Amps, you would have 14.4 Amp Hours per month. A fully charged battery in new condition could be as low as 42 Amp Hours.

The voltage monitor does sound like an excellent idea. Using the accessory socket, the voltage reading could be a little low in some cars depending o n the wiring. Other loads could increase the E=IR voltage drop.

The Toyota Manual includes a table showing battery voltage at different cha rge levels. The table shows 12.6 to 12.8 volts for Full Charge, 12.2 to 12.

4 volts for Half Charge, and 11.8 to 12.0 Volts for Discharged.

Those values assume 68 degrees F and engine off for at least 20 minutes to remove surface charge. Personally, I think you have to wait quite a bit lo nger than 20 minutes. The Toyota Manual also mentions that charge current is 5 Amps maximum. I wonder if that is a design goal under high electrical load at a specified engine RPM.

I've also heard mention of a headlight test with the engine off. After one minute, the voltage should be above 12 volts. My battery also passes that test with the high beams turned on.

I checked the specific gravity of my battery one more time and was surprise d that it now reads 1.295. That is apparently about as high as it can get. The Black & Decker 2A Charger/Maintainer continues to charge the battery very slowly even at the float voltage of 13.68 volts. It appears that one deep discharge did not damage the battery.

I have a similar problem in that my gym is .75 miles from my house. That's not really far enough to recharge the battery, even if the regulator is at 14.4V. So if I do too many of those trips, with no longer ones in between, then I'm going to have to use the charger to bring it back up.

I've never had a good feel for how long it actually takes to recharge fully after one engine start if the alternator is at max. My Civic starts very easily, so I wonder if it needs more than maybe 15 minutes to recoup. Could even be less I guess.

Yes, I'd recommend you get one (Ebay - $10 or less), if for no other reason than you'll never have to wonder what's going on. And it might even surprise you. The supply to that socket goes through the ignition switch, but other than that I don't think there's much to alter the readings. It's going to be wired pretty strongly because of what it's supposed to be able to do. My monitor reads 14.4V with the headlights on, and that's right on spec. And the monitor itself is digital, and doesn't draw any significant current.

I've also thought about the charge lost in starting. My car starts in a se cond or less in the summer, so I estimate that to be no more than 200 Amp S econds. A 42 Amp Hour battery would equate to 151200 Amp Seconds although you would have to derate that value when discharging at high current. If I 'm doing the calculation correctly a single start might consume less than 1 % of battery capacity at full charge.

It sounds as though your wiring is fine. I'll probably try turning the fan on high to see if the voltage reading changes. Right now I'm considering t he INNOVA 3721 Battery and Charging System Monitor available on Amazon.

I just measured my battery voltage 20 hours after disconnecting the charger /maintainer and it read 12.82 volts.

I've never actually measured that. The graphs show the drive voltage necessary to supply current TO the battery, so I suspect, but don't know absolutely that the battery should show voltage equivalent to that IF at the proper temperature. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 

             I'm looking for work... see my website. 

     Thinking outside the box... producing elegant solutions.

Doesn't to me. It seems the charging system is trying to be too cute to save .001MPG.

Here is some information on adjusting battery charge voltage relative to te mperature.

"For every 10 degrees above 80F, subtract .028 volt per cell from the charg er voltage setting."

"Example: A 12 volt battery at 90F. The recommended charging voltage (at 80 F) is 14.60 volts. The adjusted charging voltage is 14.60 - (6 cells x .028 vpc) = 14.43 volts."

The next link mentions 12.4 volts as a key level to maintain battery life.

tic-draw

"The key to long life for any battery is to make sure battery voltage doesn 't drop below 12.4 volts. The combination of a significant parasitic draw a nd long periods of non-use can easily cause voltage to drop below 12.4 volt s. When that happens, sulfation begins to diminish both capacity and perfor mance. That is why vehicles that only see weekend or occasional use (this i ncludes boat owners) should have their batteries maintained with a quality battery tender or maintainer."

I've seen several threads discussing the time to recharge a battery after s tarting and estimates range from 2-30 minutes.

One method is to first estimate the cranking time and amperage. Assume tha t you crank the starter for two seconds at an average current of 250 amps w hich represents 500 amp-seconds. The charge current is hard to estimate bu t might be in the range of 5-30 amps.

If we use an average charge current of 5 amps, we could replace the 500 amp

-seconds discharged from the battery in 100 seconds if no energy is lost. The efficiency of a lead acid battery is considered to be 50-90% with the l owest efficiency seen near full charge. If we assume 50% efficiency, the re charge would take 200 seconds or a little over 3 minutes.

It sounds as though the Honda Dual-Mode Charging System is not keeping the battery fully charged. It's possible that the battery stays in a range wher e the state of charge is perhaps 60-70%. In that case, the battery is proba bly never recharged to 100% of capacity.

There are probably many other factors involved. Here is an excellent articl e that discusses surface charge in a lead acid battery that is worth readin g.

"The first thing to remember is that the process of storing energy in a bat tery is a chemical reaction. The idea is to turn as much as possible of the lead sulphate (which form the plates of a flat battery) into lead and lead dioxide (which form the plates of a fully charged battery). But this chemi cal reaction is actually quite slow."

"Initially the chemical reaction only takes place on the surface of the pla tes, where they are in contact with the electrolyte. It takes some time for this chemical reaction to start to penetrate deep into the plates. This is effectively the "surface charge"."

"It can mean that a very low charge current is going into the batteries bec ause the surfaces of the plates are fully charged. But wait some time, and the surface of the plates will become slightly discharged. In effect they h ave discharged into the "deeper innards" of the plates. This is effectively what limits the speed at which lead acid batteries can be charged."

Thank you for the excellent and very interesting posts on this topic. I also have a chronic problem with keeping the battery charged on short trips. I use a Scangauge to monitor the battery voltage and compare it to a 5-digit voltmeter plugged into the 12 volt accessory socket. The Scangauge reads about 0.1 volt low.

I had a 2001 Ford Taurus that insisted on keeping the alternator output at 13.5 to 14.0 volts. This failed to keep the battery charged and it would sulphate. After losing three batteries in three years ($450 total), I decided to remove the battery after each trip, bring it inside and hook it up to a lab supply that current limited at 5 Amp.

Obviously the battery was charged a bit on the trip home, but as I raised the voltage, very little current would flow until I reached almost 14.5 volts. Then the supply would rapidly current limit at 5 Amp. All this happened within about 0.1 to 0.2 volts. The current rise was very steep.

It only took a few minutes for the current to taper off down to 0.05 Amp where it would stay until the next time I needed the car.

My next car is a 2003 Ford Focus. I had to install a new battery shortly after purchasing the car. It has been happier with short trips but it has a very strange algorithm to charge the battery. In the winter when the temperatures are around -20C, the alternator voltage ranges from 14.5 to

15.0 volts. In the summer, it ranges from 14.0 to 14.5 volts. I haven't taken the battery out and put it in the charger, but so far the car starts in 1 or 2 seconds summer or winter. Even on the coldest days there is no indication the battery is low. However, when I measure the voltage using the accessory output after the car has sit idle, the no load voltage measures 12.1 volt in the winter and 12.3 volts in the summer.

I have turned off all loads such as interior lights, and I have added a separate relay that disables the door locks when I insert the ignition key. This draws 50 mA and I can see the battery voltage decreasing slowly, but it stops when I remove the key. So I know my low voltage readings are the open circuit battery voltage and not caused by an external load.

If you look at the various battery sites, these voltages indicate the battery has very little charge and should be sulphating. But it seems to work fine, so I'll leave it alone and keep monitoring the situation.

Thanks for the interesting information. I'll continue watching for your posts so see if I can find any information that could explain these strange readings.

That appears to be the case. And I think that's a charge level that would never give problems in terms of being able to start the engine. In my case, it maintained the battery at about 12.2V (measured in the morning after resting overnight), which I think is about a 50% charge, and I never had a problem starting - until the battery actually died, which happened at under three years. So while the Honda system may give them great mileage numbers, and work ok for starting, the lower average charge level results in increased sulfation that significantly reduces battery life.

Last week I took my Civic into the dealer and had a heart-to-heart talk with the service manager about a technical sevice bulletin that I was never advised about. It provides for flashing new charging parameters into the computer. They agreed to apply the TSB at no charge to me, and since then, the alternator has spent most of the time at

14.4V. But more important, when I measure battery voltage in the morning, it's around 12.52V, which I think is probably ok. Not 100% charged, but I think close enough.

I think a better approach, which may be in effect on newer models, is to allow the computer to completely disconnect the battery from the charging system after the car has been started and the battery brought back up to a full charge. Then you could run the alternator at 12.0V, or even less, but the battery wouldn't see that. That would be very efficient, but leave the battery fully charged, which is what you want.

Or even better, there's the new alternative car battery that's made up of lithium batteries, supercapacitors to provide the cranking amps, and electronics to charge the capacitors. I don't know if it's ready for prime time, but looks interesting, and, you know, the kind of thing we should be able to do now, with all the environmental and weight benefits.

The alternator voltage at 14.4 volts should make a big difference. I just f ound some information stating that you need 14.2-14.5 volts for a car batte ry to accept a reasonable charge current in amperes. That fits in with the information just provided by Steve.

I drove fifteen miles today watching my new INNOVA 3721 Voltage Monitor. T he voltage started out just below 14.2 volts and stayed above 14.0 volts fo r the next ten minutes. As the engine warmed up, the alternator voltage dr opped but stayed in the range of 13.5 volts to 13.95 volts.

I'll check the battery voltage in the morning to see how things look. As a reference point, my battery charger/maintainer brought the battery up to t he point where it read 12.82 volts after the surface charge had dissipated.

If you would like to confirm your state of charge, you could check the spec ific gravity assuming your battery has removable caps. I'm using a hydrome ter that is at least 50 years old. My battery was slowly restored by the c harger/maintainer and the specific gravity leveled off at 1.295 which is qu ite high. Remember that sulfuric acid is dangerous and eye protection shou ld be used.

Over the last week I captured some data that should be useful when my car i s not driven for a week or more. The battery voltage drifted down from 12.

82 volts at a rate of about 0.015 volts per day. If that rate is maintaine d, the battery voltage would only drop to 12.5 volts after 20 days.

It does appear that my car is quite light in terms of charging. A battery w ill charge at 13.8 volts but a full charge would take a long time. I think the recharge time estimate for my car should assume a charge current that is only a fraction of an ampere.

No, technically the better approach is to charge the battery in the optimal way for battery life and not worry about saving a few drops of fuel.

But the car is now designed by politicians rather than engineers. m

Here is a new data point that might be of interest. Yesterday I reconnected my Black & Decker 2 amp battery charger/maintainer to top off the battery. I happened to notice that the voltage under charge read 14.24 volts which is actually a very significant number.

Charging at a current above a few amperes can only take place at 14.2 volts and higher which is consistent with my reading. It also suggests that my c harge current while driving will be very limited with the alternator output in the range of 13.5 volts to 14.18 volts. The only exception might be aft er starting the engine when there is no surface charge on the battery plate s. The voltage differential between alternator and battery would be higher for a period of time and I would expect the charge current to be higher and then decline as surface charge builds up.

It looks as though my drive on Wednesday did an acceptable job charging the battery with the voltage at 12.85 volts the following morning. The Black & Decker charger/maintainer charges to a somewhat higher level. I measured

12.88 volts this afternoon which was 22 hours after disconnecting the charg er.

I'll also pass along a link to an amateur radio website with a lot of usefu l information. See the quoted text below which should be taken in the conte xt of the 1970s.

"Since a 12 volt battery has six cells, any 12 volt lead-acid battery needs at least 13.8 volts to start to charge. This voltage will be enough to ful ly charge or maintain the battery on a trickle charge, but charging time wi ll be very long at 13.8 volts."

"To fully charge in reasonable times, alternator output must be 14.2 V to 1

4.5 V as measured right across the battery posts."

"Less than 14.3 would be a "weak" alternator or regulator."

"When running at a normal slow cruising engine RPM, the voltage across the battery terminals should stay over 14.3 volts even with full load."

Thanks for the new information and the link. It seems your infomation is very consistent about charging voltages. However, from your previous post:

"I drove fifteen miles today watching my new INNOVA 3721 Voltage Monitor. The voltage started out just below 14.2 volts and stayed above 14.0 volts for the next ten minutes. As the engine warmed up, the alternator voltage dropped but stayed in the range of 13.5 volts to 13.95 volts."

This is similar to the readings I got on my Taurus alternator. However, this killed the battery in less than a year. I went through three batteries before I finally decided that Ford had no clue about charging lead-acid batteries.

In your case, you measured 12.85 volts after your drive. I don't see how that can happen with only 13.5 volts to 13.95 volts from the alternator. That is what killed my batteries.

More mysteries to unravel...

I also was surprised at the charge level the day after driving the car. The battery which is quite new had been fully charged with the battery maintai ner. The engine starts in less than a second and the voltage rises to 14.18 volts after a cold start. The charge current could average 1 ampere over t he next 5-7 minutes which could restore the charge. However, I also had two warm starts on that trip and the voltage stays under 14 volts. A battery will still charge above 13.6 volts but at very low current. It may be that the trip was long enough to restore the charge lost in starting. There ma y be one other factor and that is the loss of surface charge when starting. The lower battery voltage at that time might initially allow a reasonable charge current to flow while at 13.8-13.9 volts. The key may be the diffe rential between alternator voltage and open circuit battery voltage.

You mentioned something interesting when describing your power supply charg er with current limiting. If I understood correctly, the current declined a fter a certain point and then leveled off at 50 ma. I think that suggests that if you wanted to maintain a given charge current, you would have to sl owly increase the charge voltage as the battery charges.

It's possible you could benefit from the use of a Solar Battery Charger. Se veral are available on Amazon and they can operate on the dashboard or in a window. If you park outside in the sun it might be worth reading the revi ews to see what you can learn.

As I brought up the voltage, the current rose from a very low level, perhaps 100mA, then suddenly went into current limiting around 14.5 Volts. It stayed in current limit for only a few minutes, then gradually tapered off in a linear fashion down to 50 mA. Ther threshold is very sharp, perhaps 100 to 200mV.

This tells me the alternator voltage must be high enough to rise above the battery threshold in order to charge the battery. A lower voltage, even a couple of tenths of a volt, will do little to charge the battery.

Of course, the voltage changes with temperature. This is documented on a number of sites, but I believe the change is around -2mV per degree F. So on a hot day, + 30C, the threshold voltage will decrease, and on a cold day, -20C, the required voltage will increase.

The car makers have wildly differing opinions on what the optimum voltage should be. My neighour had a Pontiac Trans AM (I forget what year, perhaps around 2006), and her alternator put out 15.5V after starting.

Definitely on the high side, but it didn't seem to harm the battery. The battery was a GM side mount and the connection got corroded, so I was out in -30C weather more than once to try and start her car. Fortunately, she traded it in on a new Nissan and I have not been asked to start her car since. Horay!

I got the bug to install a solar charger. There are a number of problems. For example, the sun angle in winter is very low, so the solar array has to be tilted to get much energy. However, the snow fall overnight can be half a foot or more, which would cover the array. Trying to brush it off might mean damaging the array, since it is hard to find under the snow. I did order three arrays to try to capture as much energy as possible, but only one arrived. The others were lost in shipment.

The other problem is trying to capture as much energy as possible, then to transfer it to the battery. One approach is to charge a battery bank, then start a SEPIC converter to charge the battery until the capacitor bank decayed to some suitable voltage. I found that the local electronics recycle bin often contains discarded high power stereo amplifiers, which contain large value electrolytic capacitors. I had collected quite a few of these before I traded the old Taurus for a new Ford Focus. The Focus seems to be doing a bit better to charge the battery - a least it hasn't failed to start even in winter. So I am monitoring the situation and as long as the car continues to start, I see little reason to get excited.

However, now that the temperatures are starting to climb, I am getting a little nervous. The alternator is starting to put out 14.2 to 14.3V, which I believe is too low to charge the battery. Accordingly, the unloaded battery voltage is starting to decrease. It was 12.2v this morning. So I believe the battery is heading to an early death, which could be prevented if the alternator put out a higher voltage.

Now I am looking at making a simple circuit to intercept the PWM commands from the engine control, and inserting my own PWM command to the alternator to tell it to keep the voltage at 14.5, regardless of different loads such as air conditioning, headlights, and so on. Once the battery has charged, the drain on the alternator drops to insignificant levels, so the higher voltage should have little effect on the alternator.

I remember the old days when cars used a generator with brushes, and the regulator was some relays that you could adjust the voltage to whatever you wished. The battery would last for 5 or 6 years with no problem, even with the harsh winter. Now, the electronics have taken over, the the algorithms don't seem to have been written by someone who understands what they are doing. So I am interested in overriding their poor performance. But it is very hard to find any information on the PWM signals and how to insert a different command without upsetting the engine control software.

Sorry for the long post. There was a lot of detail to cover.

IOW large lead-acid batteries have very low impedances, which is more or le ss the point. The open-circuit alternator voltage will be nearly proportion al to rotor speed till iron and copper loss take over. That's why I suggest ed a smaller alternator pulley (upthread someplace.)

Cheers

Phil Hobbs