Grid Charging

[JoeMultihuller]

I live at the end of a few-mile climb. By the time I park my Insight, my SoC is usually at half-charge unless I am really careful with the gas pedal not to go into Assist (easy to do, now that I have MIMA, but I find just the opposite because I'm trying to maintain a high mpg). I would like to be able to gently trickle-charge my Insight IMA battery overnight, even though NiMH batteries are nowhere near as susceptible to damage as flooded lead-acid batteries when left alone in a partially discharged state. I realize that I could fully recharge the battery while going down the hill, but with FAS it's so much fun to start off each trip with the first couple of miles pegged at 150mpg! Eventually where I hope this discussion will go is to a separate thread on battery pack modifications/expansions.

Ok, two primary issues with grid charging (ignoring personal safety-related ones for the moment):

1) Careful regulation and control over the battery charging current and voltage, with battery temperature monitoring a key ingredient. Rates of change of temperature and voltage also need to be factored into this.

2) Recal considerations.

Panasonic has some nice information about NiMH batteries on their website, which show the rapid voltage and temperature increase as the battery approaches full charge: http://www.panasonic.com/industrial/bat ... em/nicmet/

I haven't been able to find any NiMH battery chargers or specific regulators that operate in our Insight's voltage range. Anyone have any suggestions, other than designing it ourselves?

Regarding Recal - yes, we'll be confusing our SoC computer, but it usually finds itself within a few minutes of driving. Eventually we'll need to deal with it, but I don't believe it should hurt anything if we start off with a properly fully-charged battery. Can anyone comment on this?

Has anyone experimented with grid charging of our Insight battery?

Just thought I'd get the ball rolling out in the open. JoeS.

[nortoncomando]

This is the thread you MIMA owners should jump on!

Could you forget about your SOC display and just know your battery pack has been topped up with a gentle over-night charge?

Some elctronic shops biuld their own power supply test panels. A control chip is available that lets you dial in max volts and max current. This is what a 'smart charger' basically does. Of course inside this panel you need to start with a 175 VDC power source or so. You could have an external house current powered muffin fan at the cooling air inlet just while the battery is being charged so the only connection to the car is a low current/high volt / DC/ 2 wire connector.

This is beyond me, but Mike 'The Man' could do this on a paper napkin.

This is an area where MIMA could allow the Insight to be used as a plug-in-hybrid. Like those Prius pioneers are doing. Just use up your charge, gently, during the last 15-20 min. of your commute.

Any thoughts?

[Calpod]

There is enough energy on the Insight battery for about 3 miles, at about 80 mpg average, I don't think it's worth it, unless we had a bigger battery Pack

I'm looking into increasing the capacity of the battery pack, or increase the cooling capacity on the current pack.

But if you decide to try grid charging please keep us posted.

[Mike Dabrowski 2000]

Grid charging will be the next step to grid charged larger capacity pack. We will have both by next years tour de sol.
The personal safety issues need to be part of the design consideration when making connections to a 144V DC battery pack capable of > 100A. That will fry your A*s pretty quick.
Two people are working on this issue right now.
Armin had his grid charged in the storage bin on MIMA install day.
Sunhunter is working on a ground up design of a complete grig charging system, including a 12V aux supply for powering up the BCM and other circuits to correctly measure SOC while charging.
I will jump in soon once I get the MIMA stuff sorted out.
The other two guys will need to add the details of what they are working on, if they are ready to do so.

[Zircrose]

This is a simple circuit I made that should be able to charge the battery; I haven’t built it or thought of how to connect it yet, but this uses the 1/10th method, by charging the batteries at 1/10th the batteries capacity for 14 to 16 hours. I have made small trickle chargers for NiMH and NiCd battery packs using this method but never with such high voltages. I’m working on a more automatic charging circuit but until I find some more components able to handle the high power I’ll experiment with this.

I wouldn’t leave this on for more then 10 hours at 650mA and even then it would depend on how charged the battery is (like if it had half a charge I would only charge it for 5 hours or less). When first connecting it I would make sure the rheostat is set to the maximum resistance and once connected adjust it until I read 6.5V across the 10 ohm resistor and continue adjusting it until it’s charged to maintain a 650mA current. Just adjusting it to 650mA in the beginning and leaving it during the whole charge time wouldn’t really hurt it but as the battery became more charged the voltage across the resistors would drop lowering the current.

Feel free to point out any concerns or suggestions and if you do decided to use this remember that I am not responsible for any personal injury or damage caused to you, your property or to anyone else’s while building or using this circuit please take all precautions.

[Dougie]

There are a couple of problems not mentioned so far. First, the Insight battery (and the Prius battery, etc.) are not actually fully charged when the indicator says full. It's something around 80% charged at most. (And only discharged to around 20%.) This is to maximize the life of the battery. So you don't actually want to completely charge it, you want to get it to 80% charged.

Also, the SOC meter basically measures the current drawn from the battery to keep track of how full it is. If you don't charge or discharge the battery by sending electricity through the SOC system, it won't know how full the batteries are. So you would start the day with the SOC meter thinking they are, say, 50% full, while they're actually 80% full. After a long uphill run, the SOC system will think they're down to 20% and will shut down battery assist--but they're actually still at 50%.

It seems to me that any non-hack-job approach to charging must take these factors into account...

[b1shmu63]

First point, if you are going to do this don't expect Honda to warrant the battery. (Don't try this unless your warranty is expired)

Second point, charging the batteries without involving the BCM is the electronic equivalent of marital infidelity! (You might get away with it for a while, or not. )

Third point, As a technologist with 8 years experience designing power supplies up to 10,000 volts, I suggest that most people should not attempt this design. (Mike who has decades of industrial experience with digital control, would probably be an exception.)

Fourth point, as the sticker on the pack says, you could could kill yourself.

Fifth point, If you haven't done the other simpler mods and tips suggested on this forum, you should try them first.

Finally, as Yves once suggested, there is probably a better way to achieve the same end by using a deep cycle low voltage battery in a complimentary system. A typical boat battery has about the same power capacity, and is roughly 40 times cheaper.

Keith, if I was doing it and I'm surely not about to, I'd probably use a Triac instead of a rheostat or design a current regulating circuit with a high voltage transistor from a computer power supply, use an aviation style 1 amp breaker instead of a fuse, use a single auto transformer (properly grounded), use a much smaller capacitor, build the charger into the car where I could sense the battery temperature, voltage etc. See second point again!

[JoeMultihuller]

The points b1shmu63 makes about safety in this area cannot be overemphasized, so don't mess with this unless you know exactly what you're doing!

Zircrose, thank you for providing an input. Whereas a simple low-current input would certainly work if carefully manually monitored, what I am looking for is a fairly sophisticated regulator circuit which takes into account the specific characteristics of NiMH batteries. For example, if driven by a constant-current source, as the batteries approach full charge, there is a sudden increase in both their voltage and temperature - this certainly needs to be factored in as part of safety shutdown. Thus, in addition to any active driving circuitry, at least five monitoring parameters need to be incorporated into this regulator: voltage, current, battery temperature, ambient temperature, and time.

Note that our Insight battery is compartmentalized, with multiple temperature and voltage sensors. There's a whole computer dedicated to just battery management! I'm not (yet) up to hacking into it for our external charger. Whereas we won't be subjecting the battery to the high currents it sees while we're driving, our external control must still be very carefully implemented. I'm rather sensitive to this issue, having seen the results of a conventional battery blowing up -
http://www.katiekat.net/Cruise/KatieKat ... #Batteries

As a number of people have pointed out, a low-voltage battery is certainly a cheap storage medium. What I've already gathered are the ingredients for the front end of this charger: a couple of inexpensive ($9.95 on sale at our local electronics superstore) 400W 12vdc-120vac inverters. Put the rectified outputs in series and, voila, a high-voltage source driven off a cheap easily-recharged 12v battery (Mike D. had suggested this some time ago).

Now, b1shmu63, with your experience do you have any suggestions for off-the-shelf regulator circuits, or do we have to do this all from scratch? JoeS.

[clett]

Originally Posted by b1shmu63

Finally, as Yves once suggested, there is probably a better way to achieve the same end by using a deep cycle low voltage battery in a complimentary system. A typical boat battery has about the same power capacity, and is roughly 40 times cheaper.

I'd agree with that too! Steve Lapp has used this technique to great effect with his solar augmented Prius. The cells slowly charge the 12V lead-acid battery, which then injects its charge into the main battery while driving. Perhaps surprisingly, the computer seems to tolerate this quite well. It's possible that the Insight management could be equally benign....

There are loads of benefits from going with a 12V battery. First, you're not mucking around with the SOC of a very expensive battery only designed to hover around 60% full, and rarely reach 20 or 80% full. I think if you did cycle it between these two limits artificially day-in, day-out it would come to the end of its working life fairly quickly! Between these limits, the Insight battery only holds about 550 Watt-hours, which is only enough for about 3 miles electric range extension.

Putting in extra battery space would increase the electric range extension dramatically. In fact, lead acid is what the Priusplus amateur group used to give their first development mule 20 miles EV range. It was heavy but it nevertheless cracked 80+mpg. (The "pro" team used lithium-ion for much better results).

Assuming you go with a deep cycle lead acid battery, you could be looking at about 30-35 watt-hours per kilogram of battery. Thus for every mile of daily commuting range you want deducted from your gas bill, and placed on your electricity bill, you'd need about 160 Wh of onboard storage, which equates to about 5 kilos of battery.

For a 10 mile range extension, you'd be looking at about 50 kg of additional battery in the back.

Theoretically, assuming your daily commute is 20 miles, you could cut your gas usage by half using this layout.

[JoeMultihuller]

Clett, thank you for your comments. As I mentioned at the end of the first paragraph of this thread, I hoped that a separate thread on battery pack modifications and expansions would result. The last 100-yards of my driveway are extremely steep, but I've been using MIMA to pull the car all the way up it in second gear while keeping a very light throttle, thus showing that our Insight can successfully operate *almost* in EV-mode. I would expect thermal issues (motor and battery) to be major constraints. JoeS.