[Rusttree]

I know this concept has been thrown around a lot on this forum, but I'm going to be brave and bring it up again.

My plan is to place a fairly large (approx 4' x 1.5') solar panel on the front dash while the car is parked to take advantage of the intense Arizona sun out here. I figure I'll give it a go and see if it messes anything up in the IMA system. If the computer has a hard time dealing with it, then I'll be out $150 (assuming I don't permanently damage anything) and I'll figure out some way to sleep at night. In the mean time, if anybody has any knowledge about where the best place to hook up the leads into the battery circuit, I'd greatly appreciate it. Or if anyone knows a website that sheds some light on this subject (specifically hooking up a solar panel to the Insight IMA system), that'd be awesome.

Thanks!

-Dan

 

[Yves M.]

You'll need a DC to DC converter. To bring the voltage from around 12 volts to around 150 volts (not sure exactly). Maybe a search on the internet could bring something easy to be done (few resistors, condensators and a transformator).

 

[DonWatson3558]

Rusttree:

As you probably know, I have been exploring solar options. I have decided to start small, and learn as I go.

My first crack at solar will be to use a solar panel (with controller) to charge the 12v battery instead of the IMA battery. My thinking is this ... The IMA battery is resposible for keeping the 12v battery charged. Possibly (?) by keeping the 12v fully charged, it will pull less power from the IMA battery, and will thereby reduce the burden on the IMA 12v power drain. By reducing the power drain for the IMA battery, it may have 2 benefits: (1) give it a better chance of keeping SOC at full ... an interest of mine to help with car slowing down slopes, and/or (2) give the IMA system more power to propel the car. I will have to see what happens once I get it in place. The 12v charging seems a bit safer for now, although I would like to ultimately do as you have described and hook it to the IMA batteries. I have the panel and controller on order, adn should be here soon. Here is the panel I have purchased:

http://www.mrsolar.com/Merchant2/mercha ... _Code=MSOS

This is the charging controller (to prevent overcharging of the 12v battery):

http://www.mrsolar.com/Merchant2/mercha ... _Code=MSOS

Both of these devices should act as building blocks for improvements to the system at a later date.

An early expansion option on this would be to add a 3rd battery (LIth Ion probably) to sit between the solar panel and 12v controller. This might act as more than a trickle charger by providing a significant amount of stored solar energy to support 12v devices (stereo, a/c fan, etc.) while driving, and once again, releive the IMA battery from at least part of this responsibility. Of course, the Lith Ion battery would require it's own controller (different than 12v controller) to keep it from overcharging.

With regard to Yves M's note about the DC to DC converter ... Seems like you would certainly need a 12v to 144v converter IF you connect your panel directly to the IMA batteries. From what I have learned so far, it seems rather clear you should not do this. Rather, others suggested I hook the panel to the frontend of the IMA battery controller (which likely has it's own 12v to 144v converter), and also lots of other features such as battery temp sensors, overcharge protection, etc. I would suspect the regen charging systems in the car produce 12v power which is fed to the IMA battery controller. If so, you might be able to hook the panel at the same point. However, if the IMA battery controller is only "turned-on" when the car is started, you might be back at square one ... it may just ignore your solar power source while parked.

This idea is not my own. it was suggested to me in a private email.

 

[Rusttree]

Don,
I like your idea of focusing on the 12V battery in lieu of tackling the big sucker in the rear. From what I understand, if you could keep a substantial amount of energy flowing into the 12V'er, it would allow the system to concentrate mainly on recharging the IMA, rather than using its excess energy to keep the 12V electric system running. I'd have to run the numbers to see if it's substantial or not. Anyone have an idea the average load the 12V battery bears to keep the electric system running (assuming normal driving conditions: radio playing, fans, etc)?

In terms of charging the IMA, I had a similar thought about introducing the charge at the beginning of the MDM. But there is one extra complication here. Apparently the IMA Generator makes AC current. I suppose an DC-to-AC converter could be introduced, but I'm guessing there might be a problem with frequency and phase interference with the two voltage sources (one coming from the IMA Generator and the other from the solar panel). And, as you noted, there's a high probability the MDM will shut down when the car is off.

So that puts us back to using a DC-DC converter to bring the solar panel voltage up to ~150V and connecting that directly to the 144V battery. Actually, there's a document out by the NREL that says the actual voltage ranges between 139 and 158. Thanks to Yves for sending me the link for this:
http://www.eere.energy.gov/cleancities/afdc/pdfs/Thermal_Eval_Insight.pdf
As long as an appropriate regulator is in place, I don't see how this could be a major problem.

Thoughts?

 

[b1shmu63]

If you charge the IMA battery up without the computer's knowledge, it will be amazed to find an unusually high voltage on the IMA pack when it awakes. It will rightly determine that it has made a mistake in calculating the charge on the battery and will probably initiate a recalibration routine that will discharge the battery and charge it up from scratch. I've never tried this, but those who live in hot climates report recalibrations being triggered by higher than expected battery voltages due to heat. Force feeding the IMA pack would most likely have the same effect. I worked in power conversion for a decade and would enjoy a project like this if Honda was paying my salary, gave me access to their knowledge base, and if the Insight had a much larger battery pack. However, as things stand, the words warranty and lethal power come to mind. Surprisingly a modest 12 volt battery can hold more watt hours than topping up the IMA, so that is where I would do any mods to my own car. One scenario would involve super caps, rectifiers and a relay, but a simple aproach would be to just replace the power lost through internal leakage of the battery.

 

[Rusttree]

I've definitely had the same thought, but you also have to consider that the system is apparently able to handle a voltage change when the car is off. For example, if you're an idiot who goes to California for the weekend and leaves the headlights on in the Phoenix Airport parking lot for 3 days only to come back to a dead battery, as well as a AAA membership that mysteriously got cancelled and pay $50 to get it reinstated. Not... that I would know anything about that personally. But the point is it would seem logical the IMA computer would be able to handle a different battery voltage when it's woken up then when it went to sleep.

Also, I may be completely wrong about this, but it was my understanding that a fully charged battery is the ultimate goal of a recal. So if it detects the battery is fully charged, I don't see why it would try to change that.

 

[Yves M.]

My instink tells me that the recal is not triggered by a total voltage difference but by sub-pack voltage differences within the pack. Then to try to have all sub-packs with equal voltage, the recal is executed.

We have all speculated that a total battery voltage diff would have a recall but it was not tried out.

Maybe a test without the solar cost:
Buy a very small 12vDC to 120vAC inverter (like the one you put in the cigarette lighter) and put a bridge, capacitor and voltage regulator to 155 volts. Of course, while you have the IMA battery accessible, look at the voltage to see if we are in agrement to the voltage. Put that inverter into the IMA battery (probably where it meets the PCU) and hook it up to a 12V batttery. Once the voltage has rized (example from 145 to 155 volts), try out the car.

Maybe do it 3 times. If all is well, thaen persue further solar charging stuff.

The way I see it, a trickle charge of 2% of the battery capacity is OK. The IMA is 6.5 AH for 144v.
2% of 6.5 is 0.325A at 144v or 3.9A at 12V
(for a 12V source with a converter to 144vDC)

Therefore solar panel giving a constant 3.9 A would be OK as it would trickle charge the IMA. And this is for 24 hours a day. Since the solar is only a few hours, more amps can be put into the IMA for the few hours of sun there is.

What do you guys think of this. Any comments. I think that connecting to the IMA directly would be OK but a test is needed to prove/disprouve the idea.

 

[Foxpaw]

Charging the IMA directly is probrably not the best idea because to my knowledge that will force recalibrations.

However, you could probrably hook the solar panel up to the battery controller, to both power the battery controller and charge the IMA through the battery controller - this should alleviate the recalibration issue.

 

[Rusttree]

Ideally, yes, but again that depends on whether the battery controller shuts itself down when the car is off (the probable case). And since the whole purpose of this project is to recharge the battery when parked, that's extremely important.

Yves, that's a good idea to test it out first before buying the expensive solar panel. I'll look into those links you emailed me to build the voltage converters. Heh, looks like I have a reason to go back to Fry's Electronics for the 15th or so time since I moved here last month. Awesome store, we didn't have anything like it back East.

One thing I'm still a little confused about is the actual wiring of the IMA system. So far I haven't found a physical schematic that shows which wire is which. I'll probably be purchasing the eletrical systems service manual soon, but if anyone can help out in the meantime, that'd be great.

 

[Foxpaw]

It does shut down the battery controller when the vehicle is off, because there's no power to the battery controller. You can easily fix that by just powering the battery controller with the solar panel though - which should cause the battery controller to operate when the vehicle is off but there is power from the solar panel.

 

[Tim Maddux]

Don't forget this warning r.e. charging your 12v battery with a solar panel:
http://www.insightcentral.net/forum/vie ... php?t=1733

 

[Rusttree]

The thought crossed my mind, but I just assumed it would take too much of the precious available solar energy. Perhaps I should look into it still... it's probably nothing more than a 5V computer that eats a couple miliamps. The solar panel I was looking at recently claims it can produce 1 amp at 12V. That wouldn't be much of a sacrifice.

This project is getting more and more interesting.

 

[DonWatson3558]

Expanding a bit to my thought of using the solar to charge the 12v, I plan to start with a controller/panel alone, but once in hand, begin moving towards a 3rd battery pack idea. In my previous post, I mentioned the panel and controller I had ordered. I listed the wrong controller, this is the one I actually ordered:

http://www.mrsolar.com/Merchant2/mercha ... _Code=MSOS

This controller is the same as the other, but also has low-voltage-disconnect. Eventually, I will add the Lion Battery and a second controller appropriate for that battery.

Since the panel only provides a trickle charge, in order for the 12v to have a significant benefit, it needs more. The best the panel could do when the car is turned off, is top off the 12v battery. Once the car starts, far more power will be consumed from the 12v loads than the panel could every hope to make up for (directly). Therefore, the NIMH's will have to supply nearly all of the power while the car is running. By having a Lith Ion 3rd battery (perhaps mounted under the passenger seat), larger quantities of power could be accumulated when the car is parked. Once turned on, there would be a large store of energy available to directly feed the requirements of 12v battery for a good bit of time.

Depending on the size of the Lith Ion pack, it would be nice if it could handle "ALL" of the 12v load requirements of the vehicle without stealing energy from the NIMH batteries for 30 minutes driving or more. After that, it's back to a trickle for longer trips --- it may be a trickle, but I won't look a gift horse in the mouth on longer trips! For people making around-town trips, this might meet most or all of the Insight's 12v electrical component load requirements.

Net: Keep the load off the NIMH batteries!

 

[b1shmu63]

Rusttree, that panel you are talking about does not sound like a high efficiency one. Most lokely it is amorphous silicon on glass. These panels are cheap but are not as durable or efficient as other types. UniSolar makes really light ones and several other companies make high efficiency ones. Check out http://www.bitterrootsolar.com/solar/bp585.htm I'm thinking that you could get at least 3 amperes from a panel that size.

 

[Rusttree]

B1shmu, tell me what you think of this panel. I'm sure it isn't nearly up to par with the one you mentioned, but this one is significantly cheaper and I can buy it at the Fry's Electronics up the street. They actually sell single panels in the store, which are identical to this 3-panel set, but cost only $150:
http://shop4.outpost.com/product/3337682#detailed

It claims it puts out 15V and 45W. Is it safe to assume I'll get 3A out of it (15V * 3A = 45W)? Or are those numbers not as straightforward as they seem.

Another thing I was just thinking. The usable energy in the IMA pack is 3.8Ah. Let's say I have a solar panel that - for simplicity - is giving me 15V at 1A. Is it any more complicated then saying it'll take 3.8 hours to charge it (1A * 3.8h = 3.8Ah)? It seems like it should be much more complicated then that, but those numbers appear to work. It doesn't take the wattage into account, so that kind of bugs me a little. My intention here is to do some preliminary calculations to determine how much energy I can realistically pump into the battery over the course of an 8 hour workday. That way I can make a more educated decision on how high-end of a solar panel I need.

 

[b1shmu63]

Those panels are 15 watts each. they put out about 1 amp. They use single layer technology that is only about 5 percent efficient. I have one of thoise panels here. The glass is not tempered and will shatter into nasty bits if an acorn hits it! The same company ICP puts out a 30 watt panel that is the same size. If you boost the voltage from the panel using an inverter you will get about 0.1 amps out. Unfortunately if you try to run the inverter directly from the panel it won't work. You need a large capacitor or small battery to deliver the instantaneous current that the switching circuitry needs.

Kip

 

[Rusttree]

So how do the numbers work on a solar panel? I know that W = (I x V), but the current and voltage that the panel advertises doesn't seem to multiply to 15W in this case.

I'm going to pursue your capacitor advice. It shouldn't be too difficult to set up a circuit with a mechanical switch that lets the current fly out of the capacitor once it's charged.

 

[b1shmu63]

You just need to put the capacitor(s) across the output leads of the solar panel*. One of those monster capacitors that they sell for high power car amplifier applications should work just fine. :idea: Alternately you can go to a photo store and ask for about 20 of their used one time cameras. They just throw them out. Open them up. Remove the capacitors. Solder them in parrallel on a circuit board. Taddah! You have a monster capacitor for free.

*Use the correct polarity when connecting the electrolytic capacitors or they will surely explode! :shock: ( The small ones sound like a gun shot but once when production accidentally reversed a beer can sized electrolytic they had to clear out the building)

The panel may be spec'd open circuit voltage and short circuit current.

 

[Rusttree]

HAH! There's an interesting way to accomplish the same goal. Look at the mod underneath the CCD camera (about 1/3 down page):

http://arstechnica.com/reviews/3q00/honda/insight-4.html

 

[2flit]

PV module voltage for DC to DC conversion

If you are interested in charging at high voltages with a DC to DC converter.... You would be well advised to find a high voltage module. The efficiency of the conversion is related to the change in voltage. The greater the difference in voltages; the lesser the thru put efficency of the converter. This has been true so far for the ones that I have used for other purposes. There can be a rather big efficiency penalty for these conversions, be sure to ask what you can expect at power levels lower than the design specifications of the DC to DC converter that you are purchasing (E.G. at about 1/5 the PV modules Imp; current at max power point) Best of luck , 2flit