Canion Battery Readings

[siai47]

I'll have to modify what I said earlier about the canion accuracy. I only checked it with the L1. When I supplied L2 levels the canion was reading 6.8 amps and 360 volts. The voltage was accurate, however the Fluke ammeter was reading 7.5 amps. When charging with my Manzanita PFC-50, the canion was reading about 25 amps and the Fluke 31.8 amps. I didn't see the error at the low power level of the L1. It appears that the error is increasing and more noticable at higher currents. The fluke numbers seem more in line with actual levels going into the battery based on all the instrumentation that is attached to the PFC-50 compared to what the canion is reporting. There might be a tweek needed in the app--I know that's a constant thing on the LEAF app. I am still greatful for the work that has been done to get the canion app published and for the great help it is for those of us who are interested in this kind of stuff.

 

[offgridQLD]

Don't get me wrong its a great app and a donation will be made to express that.

I just think if the apps main function is to report data from the car it needs to be (reasonably) accurate or that function becomes pointless.

I think Voltage and amperage are the most important points as its the foundations for everything.

Kurt

 

[priusfan]

bonjour,
about canion's accuracy:
I tried to find infos when sniffing the CANBUS and comparing with the data provided by the service tool from peugeot.

yes, there is a problem with the amps but, in fact, canion reflects the same values as the servicetool I used....

next step will be to compare with the data from mut3 (the service tool from mitsu), and maybe it will be possible to find a better source for amps...

Xavier

ps: thx to generous donors.

 

[offgridQLD]

Yes if the amps reading were accurate then all the other data on the app will be accurate to like whr, regen and any other reading that uses amps in its equation.

On Monday I did a trip of 105km in 36 deg C weather with the AC on all the way with 3 people in the car. I did have one large hill to Regen down on the way of some 550m so that helped though at the end point I was showing 4 bars on the Imiev gauge and the app was showing around 35% SOC and a whr km consumption of 100wh- km. The Imievs gauge was showing I had consumed 12kwh and the app was showing a consumption of around 10kwh that's a big discrepancy.

I think the best bet is to work with know accurate data reference point. Perhaps from a calibrated clamp meter on the battery it could be a handy tool when developing the app in trying to work out what data feed is what.

If data isn't accurate or unless some kind of offset number given so you can calculate it to get a accurate number from it then the data is just a pointless number.

The app has huge potential perhaps it could be as simple as a small equation error or something small that's overlooked.

The amp numbers shown when driving like full throttle seem to mach the gauge on the Imiev. 50kw at full throttle 16kw at the first line and so on. Charging though the largest number I have seen is 1700w though it usually sits at 1.6kw with the odd flash as 1.7kw when charging while my charger is consuming 2200w from the wall.

Kurt

 

[PV1]

The Imievs gauge was showing I had consumed 12kwh and the app was showing a consumption of around 10kwh that's a big discrepancy.

One thing to remember is that each bar is not equal to 1 kWh. 8% of 16 kWh is 1.28 kWh, which is how much energy the top bar represents. I think the second bar goes at 86%, which is 6% (.96 kWh). There is a post in this thread with the percentages for each bar.

The app has huge potential perhaps it could be as simple as a small equation error or something small that's overlooked.

This is also quite possible.

On another note, does anyone else's 8th bar contain about 10 miles RR? I usually hit 40 miles RR on the 8th bar, and it doesn't go away until about 30 miles RR.

 

[siai47]

Perhaps it is the CAN information for the dash gauge that the canion app is using to report actual amps. Maybe that is what was "sniffed" when the app was developed. Someone printed a early post of how many amps each part of the gauge represented. The way my car is configured, I can see the gauge when charging and I am putting more current into the battery then the gauge indicates (provided the poster got it right). It's possible some of this information is buffered for display purposes---just guessing here. Anyway, an offset calculation in the app should be able to correct this inaccuracy or another look at the raw data to see if there is another place to pick up the data. If I remember correctly, there is a seperate CAN buss between the battery pack and the BMU (along with K line backup) that is independent of the vehicle CAN buss. The BMU reads the data from the current sensors and may be applying some changes before sending it on to the EV-ECU--just another guess. I am pretty sure the MUT III can see the raw current data.

 

[offgridQLD]

To be honest I couldn't care less about the RR gauge its a total guessing meter with so many variables I just ignore it . How the hell is it going to predict the future based on the past:lol:. Give me pure whr consumption and whr- km that's accurate over turtles, bars and RR meters any day.

Kurt

 

[PV1]

RR gauge its a total guessing meter with so many variables I just ignore it .

Not to nitpick, but I find mine to be quite accurate. Granted, it seems to always shoot high, but I think this is because the last 2 miles to my house are coasting/downhill. I would prefer having readouts in the car of total available energy, consumption/regen, and distance traveled. But since the car doesn't have it, that is where CanIon fills in beautifully.

I traveled to brand new territory a couple of weeks ago, driving 55.7 miles one-way. I started with a RR of 70 miles and ended with 9 miles and 2 bars. This would leave the gauge 5.3 miles on the far side. Probably could've gone 70 miles if I run it until it stopped.

The return trip was 56.1 miles, starting with a RR of 69, ending with 21 miles RR and 5 bars. This used 48 miles RR to go 56.1 miles (due to 2 big downhill slopes on the return trip, the first one regen cut out because the battery was full, the second I regained a bar (!!) on the way down.) Sailing down a (rather steep and curvy) hill at 50 mph with little regen is quite an experience. :shock: Made me feel like I was driving an ICE pushing that hard on the brake pedal. Cranking the heat didn't help, just diverted some power away from the battery without increasing force.

CanIon was a big help for this 2-day trip, allowing me to run navigation on my everyday phone, have the trip odometer up on the car, and RR showing in the app (this is in miles, even with the rest of the app in km, pretty sure it just sniffs whatever number goes to the display.) I assume that cars set to km show RR in km as well in the app?

 

[Don]

The RR display is the only thing I ever use the multi-meter for and I find it quite useful - More so for me than the bars on the 'Fuel' gauge - I wish the RR was a constantly displayed number somewhere else on the dash . . . . and a little larger readout than what's on the multi-display

Sure, I'd love to have a SOC readout or a KWH consumed gauge . . . . but I don't think I'd rely on them in everyday driving as much as I do the RR display - When I know how far I have to go to get home, I concentrate on driving so as to minimize the negative effect on the RR gauge . . . . and I always make it

Don

 

[PV1]

I did some testing yesterday. I charged the car for an hour and a half and compared Canion to a kill a watt for stock level 1 charging.

The kill a watt showed 1.29 kWh, Canion showed 630 Wh "regenerated". This is a difference of 660 Wh. So the car, including charger efficiency loss, uses 440 watts while charging.

 

[offgridQLD]

I charged on Sunday at a public charger 3.1kw and canion was showing 2.5kw. So again 600w difference. At home I see 2.2kw at the wall and 1.6kw on canion again 600w.

Kurt

 

[cometboy]

New member to this forum but not new to EV’s (see http://www.mightyboyev.com). We have purchased a new Jan 2012 i-MiEV at a very attractive price – Mitsubishi Australia has decided to only special order i’s from this year forward and hence sold off old stock here at basically cost price with full warranty. The vehicle we purchased had 33km on the clock and basically had just been stored at a dealer and forgotten. We have had the little EV on the road now for a couple of weeks (great machine!), and I have just started using caniOn to gather some base data for future reference and any warranty issues down the track. The vehicle is used for relatively short distances on flat ground probably averaging 45 km a day. It is charged nightly on off-peak rates and most charges are complete full charges (so with the balancing carried out).

My first question is:
Should I be concerned about a 40mv difference on two cells as per the Batt Status graph? I cannot find to date any information on what is acceptable here? I’m sure many on this forum have collected Batt Status data and was wondering what the results show?

Any views appreciated....

Thanks
Bruce

 

[PV1]

Based on the voltage, this is near full charge. I would say do a few runs down to between 1/2 and 3/4 charge (8 to 12 bars) and charge back up fully (until the charge light goes out) to see if they catch up. Keep an eye on the cell voltages as you drive to make sure no individual cell drops dramatically in voltage.

While you're driving, the cell voltages will bounce around in-between refreshes, so comparing the voltages is best done when sitting at a redlight.

**EDIT** - One afterthought, do your charging with the cord included with the car, unless you need to charge faster. A slower charge rate should help the balancers work better. Although, I don't know if this actually does help, since the car ramps down no matter what you use.

 

[siai47]

I would agree, just keep an eye on the cells that are low. It takes a long time to bring the cells back into balance but slowly they will. I bought a wrecked, almost new, I-MiEV that had been sitting for about eight months after the accident. When I swapped the pack into another car, there were two cells that were about as far out of balance as yours. It took a couple of thousand miles and full charges to bring them back into balance. The balancers aren't very powerful and it takes time. At least you are aware of it---if you didn't have the canion you wouldn't even know. I'd bet there are a lot of I-MiEV's out there in this state except the owners don't know it. I also own a LEAF which doesn't hold the balance as well as the I-MiEV and it seems to run normally.

 

[Malm]

If at 50% SoC the difference is lower, you have two unbalanced cells and there is no reason to be woried. As said, just keep charging it to 100% every charge, and one day they will charge as the others. But it will take a long time, long, long time.

 

[JoeS]

cometboy, congratulations on your new i-MiEV and welcome to this forum. Lovely clean Suzuki conversion you did! Looks like you're in SA(?) - we have a number of members in QLD, at least one in Tassie, one in the Melbourne area, and one in Canberra recently joined. You might update your profile to show your location as it's always fun to see regional perspectives from around the world. If you're like many of us, I think you'll find yourself enjoying your nimble little workhorse.

Thanks for posting your battery status and I agree with the above comments and hopefully over time the cells will balance out.

Of far more importance, now that summer is coming for you, is NOT to leave the battery pack in a fully-charged state. This is a radical departure from your experience with Lead-Acid and does require a mindset change. Here's a writeup from our paranoid Leaf friends - it provides a lot of hints about the care and feeding of Lithiums.

http://www.mynissanleaf.com/wiki/index.php?title=Battery_Capacity_Loss

Edit: Keep in mind that the Leaf has twelve bars whereas our iMiEV has sixteen bars at full charge.

Since you will still be attempting to balance charge your i-MiEV, you might time your charging s.t. you do it late at night or very early morning when it is cooler and then drive away very soon after she stops charging.

All the best to you.

 

[FiddlerJohn]

... Should I be concerned about a 40mv difference on two cells as per the Batt Status graph? ...

Cometboy,
I agree with the above comments. Balance first.

You battery's internal resistance is a good indicator of health. Full acceleration should draw 150 Amps and the pack voltage should drop around 15 Volts. That gives a battery internal resistance of 0.1 Ohms.

I measured a 10,000 mileage pack with a 14 volt drop, but my 23,300 mile pack has a 17 volt drop. I try to measure the Voltage drop with a 50% to 80% SOC.

 

[Malm]

"I measured a 10,000 mileage pack with a 14 volt drop, but my 23,300 mile pack has a 17 volt drop. I try to measure the Voltage drop with a 50% to 80% SOC." Lost confidence in SoC values determined by my i-MiEV, It jumps 5% up and down by magic (maybe because of changes in cell tempertaures). The i-MiEV has strange ways to measure his SoC :? .

 

[PV1]

Malm, if it is because of temperatures, then the SoC calculations are based on usable capacity, which more or less shows true SoC. Less energy is available from a colder cell than a warmer cell. This isn't as visible with lithium as it is with lead-acid.

Power output is also reduced for a cold (or hot) cell. This is why regen is reduced in cold weather. I noticed my regen is capped to 30 kW when cell temps are at or below 13 C. Above that, even at 14-15 C, I have full regen (up to 43 kW with brake pedal in B mode). That 13 kW extra really is noticeable. It had me worried when I first experienced the cap.

 

[offgridQLD]

I did find a little write up from a local owner of a Imiev regarding cell balance check during servicing. Though I question the level of technical understanding the owner has. So there could be some misinformation being reported. That said It dose make me wonder if Mitsubishi servicing actually have a off board balancing device to bring a way out of balance pack back into line over a short period of time.

I know I have had to do it by hand on a few small electric bike packs using a 50w light bulb as the little BMS shunt just didn't have the grunt to get the job done.

link to blog that got me thinking.
http://www.hinterlandtimes.com.au/2012/08/03/i-miev-blog-keeping-the-ev-batteries-balanced/

Kurt