Honda Insight Forum banner

1 - 19 of 19 Posts

·
Registered
Joined
·
13 Posts
Discussion Starter #1
I've been working on putting together a charging system for my 2008 Civic. I've installed a homemade harness and I have an Bio Rad electrophoresis power supply for high voltage. For PWM I've tried a few things: shorting with various resistors, from 100 to 100k, an Arduino microcontroller, a 555 timer chip, and a 2-channel PWM generator with adjustable frequency/duty cycle.

So far everything I've tried has the same effect as when I short it with a resistor: the fan cycles on and off repeatedly. It still seems to be moving air fairly well, but I don't understand why it doesn't respond to pwm. I've tried various frequencies and duty cycles, as well as different methods on the Arduino for generating the signal.

I've read many threads here and elsewhere, but I haven't seen much specifically about signal for the 2nd gen Civic. Thanks for your help.
 

·
Premium Member
Joined
·
3,415 Posts
You need abut 100W to run the Fan at full blast, and it readily exceeds 2A at very low speeds.

That's the easy way. Get a 10A/12V PSU for $18, short the PWM wire to fan ground and run it full blast.

The cycling you're seeing is your power supply overloading, shutting off and restarting.
 

·
Registered
Joined
·
13 Posts
Discussion Starter #3
Interesting, I had assumed the 3 amp supply I had on there would be more than enough. I have a spare 450W ATX power supply which should do the job for the moment, I believe.

The power supply is another issue. I need to get it to stop throwing no-load errors or switch to something else. Based on other threads, it seems like putting a resistor or diode on the circuit might work.
 

·
Premium Member
Joined
·
3,415 Posts
You absolutely, positively need a diode in the line, or you run the risk of damaging the supply. They are not designed for inductive loads but pure resistive loads - they don't like it when the battery pushes back on them.

I actually put diodes on both leads because I feel better about it.

Which one do you have?

The HCH2 fan is a monstrosity. It's like a leaf blower... moves a ton of air.
 

·
Registered
Joined
·
13 Posts
Discussion Starter #5
I got the brand wrong before, mine is a VWR Accupower Model 500.

I have some 1N4007's here, though I'm not sure the best way to wire them in. Right now everything high voltage is wired with safety banana plugs on regular 120v power cords. I see that they sell fuse-style diodes which might be a good option.

Will the diodes help the no-load issue?
 

·
Premium Member
Joined
·
3,415 Posts
I'm not familiar with that unit.

They should be wired in the lead. On the (+) side, the stripe should be on the battery side. On the (-) side, the stripe should be on the PS side. If you get these backwards, you'll never get current to flow from the PS.

On my Thermo unit, I have to press and hold the start button while the voltage rises until the current starts to flow. Once I get flow, I can release the button, and it sustains.
 

·
Registered
Joined
·
13 Posts
Discussion Starter #7
Interesting. The button push trick doesn't seem to work for me. Previously I had managed to get it to charge a 24v battery for a while, but now it pretty much throws no load code right away. Maybe I could have it power a lightbulb at the same time or something.
 

·
Registered
Joined
·
13 Posts
Discussion Starter #8
The ATX power supply worked out well. PWM works too, although full-blast is fine for my needs.

I installed 1N4007 diodes inside of the electrophoresis power supply, right behind the banana jacks. I was able to charge a 24v battery at that point, but trying to charge the hybrid batt is still throwing codes.

I bought a 100w 25 ohm resistor in an aluminum housing, I'm hoping that will manage to fool the no-load protection. If not, I might have to order another supply.
 

·
Registered
Joined
·
13 Posts
Discussion Starter #9
Quick update: after attaching diodes inside the power supply (heat sinked with foil w/ toothpaste for heat transfer), and plugging multimeter into supply before connecting the battery cables, I was able to get it to charge somewhat consistently. Charge maxed out around 184, then leveled off to around 182.6.

Battery had only recal'd once recently and has basically been in good shape, but I did seem to get a significant gas mileage boost: 46 mpg on a 40 mile stretch, as compared to 37-40 average before charging.
 

·
Premium Member
Joined
·
3,415 Posts
How long did you charge and at what current?

That sounds like a low voltage to me. My HCH2 packs routinely break 190V @ 350mA

Recals kill your mileage, so you should see a significant improvement.
 

·
Registered
Joined
·
13 Posts
Discussion Starter #11
I probably should've kept more careful track during the charging process. Initially I was having to start and stop the power supply several times before it would properly kick in -- it would often power up, show a low voltage like 45v, then throw a load error. Later on it got more reliable for some reason.

As I recall, I charged @ 500ma for a couple hours which brought the battery from 167v to around 178v. Then I let it run overnight @ 350ma, though it killed out at some point. Then from 7am to around 3 or 4 pm, when I decided too much sun was hitting the car so I cut it off. Then from 6 pm to 2am. So around 24 hours total, spread over about 32.

It reached the low 180s pretty quickly, I'm not sure why it topped out where it did. I might've still been at 500ma when it reached 184 (maybe 185), which might explain why it receded down to 182.5 later on.

Throwing together a discharger now, though I don't intend to use it till the battery gets pretty bad.
 

·
Premium Member
Joined
·
3,415 Posts
Again, 184 is low for a peak in my experience. I have about a dozen of these packs, and none have peaked at less than 190V @ 350mA. I would assert that your pack is not fully charged.

There is no reason to wait on a discharge. At a minimum, you should discharge to 158V from full while measuring current and get a baseline capacity on the pack.
 

·
Registered
Joined
·
13 Posts
Discussion Starter #13
I'll try again sometime soon when it isn't so hot. The battery was being a little weird today, and dropped to 2 bars a couple times. Hopefully I didn't mess anything up by accident. It was about 95° though today, so maybe that was part of the problem.
 

·
Premium Member
Joined
·
3,415 Posts
I'll try again sometime soon when it isn't so hot. The battery was being a little weird today, and dropped to 2 bars a couple times. Hopefully I didn't mess anything up by accident. It was about 95° though today, so maybe that was part of the problem.
Hot is what you're not used use... 95°F is nice.

I re-read your post. Are you saying 500mA of charging for 24 hours over 32 hours?

If yes, that was actually too much charge. At 500mA, you're pretty much at 0.1C, so you should have limited it to 16 hours.

Two battery bars is almost always indicative of a negative recalibration. You typically shouldn't be having those immediately after a grid charge.

What was the condition of your battery before you attempted grid charging?
 

·
Registered
Joined
·
13 Posts
Discussion Starter #15
Nope, sorry -- 500ma for a few hours, then 350 for the rest.

The battery was fairly good so far. I saw 2 bars one time, but in general I usually have juice to burn when I need it.

When I bought the car in March the battery had just been refurbished. Fortunately it had a 3 month warranty, as it went pretty bad pretty quickly -- a few seconds of assistance getting onto the highway and then recal. They claim it was caused by a bad fuel air mix sensor, so they replaced the battery with one from a junkyard and charged me $450 for a new sensor.

I'm sure 95 is not bad in Phoenix, but it's rather humid here as well and my central air is out. Looks like this week will be better, though, so I'll probably give it another try.
 

·
Premium Member
Joined
·
3,415 Posts
Okay, good. It was probably about right.

Your story is all too common. I'm sorry. You were swindled. The upstream sensor is expensive ($200 for the Denso part), but it's easy to replace. It has a specific DTC, and there is NO way it could contribute to a battery code in any way. HOWEVER, if the junk yard battery is from a lower mileage car and is unmolested by a "reconditioner", then you may be way better off than you were before the battery exhibited problems.

The 2006-2008 Civic Hybrid batteries are the second worst hybrid battery ever made with a failure rate of about 1 in 6 in a 5-7 year timeframe. There's a reason there was a class action lawsuit and warranty extension. They are second only to the "improved" 2009-2011 Civic Hybrid battery - which has a failure rate of 30% in 3-4 years. The only upside to the 2009-2011 is that Honda extended all states to 10 year/150K mile warranty before they got sued again. :)

A single recalibration isn't a concern; however, your next one is an alarm bell indicating that you need to immediately conduct a full reconditioning effort.

I personally have 132 subpacks for the 06-08 HCH. Out of that 132, I have 4 good ones that would actually work well in a car. That should give you an idea of what you're up against.


BLECH! 95 with humidity is horrible. I'll take 113°F in AZ with our typical low humidity than mid 90's with humidity. I come from a long line of heavy perspirers... it's not pretty when the humidity climbs.

Good luck,

Steve
 

·
Registered
Joined
·
13 Posts
Discussion Starter #17
Yeah I didn't really think their explanation made sense. I would've rather replaced the sensor myself, but they'd already ordered it and I figured they weren't too thrilled to have to pay for another battery.

I finished my discharger and drained battery to ~157v. Very stupidly blew up my multimeter and scalded my hand with molten plastic in the process -- I put the red plug into DC10A by accident. Still works, surprisingly.

Charged for 30 hours and got it up to at least 189v for 6 hours. Hopefully that did the job, I could've let it go longer but I neglected to wake up when the charger's timer cut off at 30h, and once I woke up, I thought it would take too long to get back to max voltage again.

Running good so far, but only driven ~ 150 miles.
 

·
Premium Member
Joined
·
3,415 Posts
Welcome to the He-Man-hurt-yourself-doing-something-stupid club. I renew my lifetime membership several times a year.

I've done that too. Exciting!

You probably cooked the 10A circuit. I'm surprised the leads didn't fry. You might be able to replace a fuse.

I would expect everything on the non-10A circuit still works provided no adjacent circuitry popped.

30 hours is more than enough at 350mA. The HCH2 has lower capacity, and 24 hours is usually sufficient. Unlikely you did any damage. If anything bad happened, it's likely only a little bit of reversible capacity loss.

Getting back to max voltage again doesn't really matter, AND it will go a lot faster than you think. It's Time * Current = input mAh. If it sits for a few yours, yeah, you'll get a little self-discharge, but it doesn't really matter. Again, 30 hours is more than enough.

What was your discharge current?

Did you go lower than 157? If not, I recommend you do so.
 

·
Registered
Joined
·
13 Posts
Discussion Starter #19
For discharge I used two 60w incandescents, then for a short while two 23w flourescents since I didn't have low watt incandescents in the house (only one lit up). At one point the multimeter read ~105v, but I'm not sure why, as when I checked it shortly thereafter it was at 157.

Performance was dodgy for a few days after charging. No recals but lingering at 4 bars without properly using the juice to accelerate. Voltage was still showing 180v a few days after, and now it seems to be better. My best guess is that I inadvertently imbalanced the pack worse than before, but I don't really know.

I bought a couple 25w incandescents and I suppose I'll probably want something beefier than 60w too. I don't want to charge/discharge too often but I'll try again if performance gets dodgy again or maybe the next time I change my oil (has 30% left I think).

Thanks for your help, by the way.
 
1 - 19 of 19 Posts
Top