22 December 2009

Drove the E up to High Point

Thanks to Chris for letting me charge up on the way to the very northern tip to New Jersey today.

Voltage at Chris' measured about 122 in the morning and 117 on the way home after dark.  No trouble charging either time.  Of course, the car would see about twice those numbers, or 244 to 234 volts. Well within ANSI specs.  Unlike at my house which  is often 253 volts and the Mini E won't fast charge.

By the way, the skiing was lousy, too many bare spots from all the wind.

21 December 2009

The cavalry is on the way

I heard today from Clean Fuel Connections that the entourage from Mini will be visiting me in early mid January.  Not just to diagnose the charging issue but with a potential work around and possibly a modified (fixed?) car to see if it has trouble with my high line voltage.  I suspect mine is the highest of the bunch who are having trouble, which apparently numbers six.  Not bad, 1% of the total.  This is the sort of problem that needs to get figured out in the prototype stage.

Now, if I can get really technical for a moment - Mike, you might remember this from your Electrical Engineering days.

You don't need a 15 kilowatt transformer to lower the line voltage to get around this problem.  A transformer with a 12 volt 60 amp output could do the job theoretically.  Wire the output in series with one side of the 240 volt line but opposing phase (symmetry is not needed) and you drop from 240 volts nominal to 228 volts.  That should be a MUCH cheaper transformer.  I'll post a schematic in a few days maybe but I doubt something like this would comply with the electrical code, so it is academic.  Maybe try it on the 4th of July so that if it explodes, it would seem to be on purpose.  Ha ha.  Maybe I can find a picture of the cavalry with some explosions?

20 December 2009

You can charge any car your want at Tom's restaurant

OK, this is about line line voltage, but not mine!

We visited Mike Graham today. His line was 244 point something volts which is well within ANSI specs, although on the high side. I could charge #458 fine, but Mike's #269 would not charge. So the over-voltage threshold in his car has drooped even lower than mine in the cold, but his line voltage is also lower than my 253ish. So we have similar problems at our respective homes.

Tom's restaurant is nearby Mike's home so we went over there to finish charging and get some DELICIOUS food and peek at his line voltage. As Tom told me, since he has industrial 3 phase power, the single phase feed is lower with a nominal of 208 volts. (For the geeks, that is 120 volts times the square root of 3, for 3 phase.) Tom's line actually read 204 point something today. And of course, no trouble charging. No one ever has had trouble charging at Tom's restaurant. Because the voltage there is lower.

Mini, are you taking notes? There will be a quiz when you make your house call. I am handing you the answer to this charging problem on a silver platter here.

Oh by the way, this long drive in a BIG blizzard. That traction control is Great. The heater is not dependable. Goes cold now and then. Need to leave the fan up on three.

18 December 2009

I hope this is the last post about line voltage

Tonight I read the power line at about 250.1 volts and the Mini E would not even blink. Had to even dig out the clothes iron to get the line voltage down to 248 before she would charge.

It is nice to know what is going on finally with this charging problem. But it is rather frustrating to deal with and not without risks. I think this is the end of the data collection for me.

I think I got word through the ramparts and into engineering at Mini, although there is no confirmation. Had to use the back door to get someone to listen, but that is the way it is most places, including where I work. It is not an ideal situation, but the engineering department can get flooded with noise and distractions if the guards did not keep the Visigoths from the door.

Last night it worked

We got home with temperatures in the low 20's, line  voltage about 252.6 and of course the Mini E won't charge, not even one blink or click.

After turning on almost everything in the house, the line voltage drops to 250.0 and bingo!  It started charging and did not stop.  I was expecting to need to get it down to 249, since recently 249.2 was not far enough down.  Sounds like 249 to 250 is the threshold when temperatures are a bit below freezing.

I heard from a reputable source that the AC Propulsion safety limit (maybe there is another more official term?  Jason, can you comment?) should be 264 volts.  That gives plenty of margin past the ANSI spec of 252 max.  To bad it doesn't hold in the cold.

16 December 2009

No dice tonight

This was a test of charging tonight. I get home at zero range (although the gauge floats back up to 11% when I turn it off) and temperature just below freezing. This is usually the worst combination for getting the fast charger to work. I measure 250.8 volts, not as high as usual, but 1.6 volts higher than last night which was warmer. So I am not surprised when Mini E #458 won't even try to charge from the fast charger. Note that 250.8 is within the ANSI standard.

I turn on all the big loads in the house. Space heaters, toaster, the few incandescent lights that have not been switched to CFLs. Voltage drops to about 250 and when I plug in the car, it blinks a few times but then stops. Well at least it tried. It noticed the difference. Usually, if it won't even blink on the first try, there continues to be no response for 30 minutes with the big wall box "charger".

So I turn on everything in the house, even open the door of the fridge until I hear it start. Voltage down to 249.2, like it was yesterday (except I didn't have to turn on anything yesterday to get down to 249.2, maybe the neighbors had something big on yesterday). The car blinks 8 times, more than last try but then it stops. Still, this makes sense, it is colder tonight by maybe 15 degrees even though I got the voltage down to yesterday's value.

OK, put it on the 120 volt cord, I am out of time tonight. At least it only needed 10 minutes on 120 volts before it would switch to the higher voltage fast charger.

I have an 8000 watt heater I have not used in a while since the sauna turned into storage space. I'll have to clean it out so I can turn it on. An 8 kilowatt load should pull the power line down by several volts. Best part is, it is just one switch, not 50 things all over the house.

15 December 2009

Voltage tolerance in North America

I did find a spec for AC power line tolerance in North America, 114 to 126 volts. So technically my power line at home which is usually at 126.5 volts is too high, but I say AC Propulsion needs to be more tolerant in cold weather right after driving.

The Wikipedia link is here. The excerpt follows:

"In the United States[3] and Canada[4], national standards specify that the nominal voltage at the source should be 120 V and allow a range of 114 to 126 V (-5% to +5%). "

[3] ANSI C84.1: American National Standard for Electric Power Systems and Equipment—Voltage Ratings (60 Hertz)

Tonight my power line read below 250 volts, 249.2 to be precise. I had no trouble charging, it started immediately and kept going. I cannot remember the last time I did NOT have trouble with the fast charger. So this supports my theory. Then again, it was not so cold tonight. Tomorrow will tell. But maybe, just maybe Wile E. Coyote (my alter ego in this context) finally caught a valid theory.

Who knows why my line voltage was below 250 tonight. Coincidence? The idea that winter voltage is higher because all the Air Conditioners are off is interesting, but it does not explain why waiting will usually get the car to accept the fast charge. Then again, neither does my theory unless it is somehow related to battery voltage. And besides, I bet the power company regulates for load conditions.

I know waiting usually will get it to start charging again. But a couple times a month I take the boss out dancing when I get home from work with 1% charge left. If it doesn't start charging right away, we have to take the gas car. And she doesn't like the gas car anymore!

So you see, Mini, I had to figure it out myself. I couldn't wait for the house call. Because tomorrow night is dancing night. And it is very important to keep the boss happy! (That would be my wife.)

13 December 2009

I finally caught the culprit

How many Mini E drivers have this problem charging from the 240 volt wall box in the cold?  Mostly in New Jersey?

Here is what I am pretty sure is happening:

The AC Propulsion electronics (the big brass box under the hood) checks the voltage when you plug it in.  If it thinks it is too high, it won't charge to protect itself, or it will start charging and then see the problem and stop.

Something about the colder weather (maybe combined with a low battery charge) is making it judge the AC line voltage (mains voltage) too strictly.  I typically have over 126.5 volts at the regular outlets around the house.  So at the Clipper Creek charging interface (or wall box), it sees twice that, which is over 253 volts.  (Please Please DO NOT try to measure this directly if you are a civilian.  There are safe meters such as the Kill A Watt which will measure a regular outlet and you can multiply that by 2 for an estimate.)

Normally I would guess the AC Propulsion box is happy if the charging voltage is below 260 volts, but in the winter that threshold has slipped down about 3% and it now thinks 253 volts is too high.  Well, our local power company PSE&G does a very good job keeping the voltage up in central New Jersey.  I don't think there really is a problem from the power companies' point of view until the voltage gets to 128 volts, but I am not having any luck finding a specification.  I just remember that from somewhere.

I think I can prove this to Mini, although they can probably verify it themselves.  Too bad, I was looking forward to the promised house call.  But I am super busy at work right now so I probably would not have time to meet them anyway until January.

If you are one of the Mini E drivers with this issue, try turning on everything in the house before you plug in your E.  The extra load should pull the voltage down a bit and might be enough to make the car judge the voltage as safe.  Don't forget to turn things off again as soon as the car has been charging for 15 seconds, especially the stove or the dryer.  I even ran the Microwave but be sure to put at least a cup of water in it, they don't like to run empty.

I switched back and forth several times, and the car would always continue charging when the line voltage was at 240 or slightly below, but back up at 253 volts it would either not start at all or only blink a few times and then stop.

After all the theories that have come and gone, I am going to be really embarrassed if this gets disproved.  But I am pretty confident this time.  The real test will be tomorrow after the long commute to and from work and the low battery charge.

12 December 2009

Throw out all the theories!

I cannot count how many theories have gone back and forth about what might be behind this charging problem.  Several have been ruled out.  I don't remember who first mentioned the cold, but I do have some evidence on that.  The first time I remember having trouble charging was Sept 25, which was also the first day that the ambient temperature was below 60 degrees when I got  home.  (See, those tedious trip logs do come in handy.)

It was also the first time I ran the battery way below 0% to see what would happen.  (That is how I remember the date, I posted a blog entry about it here.)

The only things I can say with confidence is that if battery charge is low on a cold day, I have trouble charging at 240 volts from the wall box in my garage.  Pretty much every other theory I have examined seems to be full of holes.

The last couple days were about as cold as I have recorded this season, with low twenties yesterday.  I was driving loaner Mini E # 008 until this morning, and I had more trouble getting #008 to charge than I ever had with my #458.  This morning when I got #458 back from Princeton Mini, I had just as much trouble as with #008.  The common element is temperature around freezing.

And both cars charge fine at 120 volts.

So when the engineers from Mini make a house call, we need to pick a cold day and I need to run the charge down low on the car.  Heck, they can bring #008 by again if they run the charge down, then we have two chances to figure out the problem.  Because once the car starts charging at 240 volts, the problem disappears until the next drive cycle.  And you can fix the problem if you cannot reproduce it.

On the edge of the qualifications

Some readers of this blog may get an incorrect impression from the lengths I go to when commuting in the all electric Mini E.  Those who are also Mini E drivers might understand the assumptions I am making in writing this blog, such as what the headline photo implies.  I have carelessly aimed this blog at fellow Mini E drivers.  For those readers who don't have their own Mini E lease, let me set the context.

When Mini started taking applications, it was clear that I would not be considered since my normal commute was 150 miles per day, and on the outside they were accepting people who drove at most 110 miles  per day.  Even then it appeared that they would only accept my application if they were desperate.  I worked with Google maps for quite a while and found a route to work on back roads that was about 110 miles round trip, and claimed that on my application.  It turns out that route is not passable by car, the best I  have done is around 116 miles, but normally 120 miles to save time.  I argued strongly with Mini that this would be fine since I could charge the car at work and Mini finally agreed to lease me an E, after strenuous pursuit on my part.  (I recently made the round trip without the benefit of charging all day at work, see link here.)

Being an engineer, I want to  know things, such as the effect of cold weather on the range.  In isolation.  That is, I want to know the effect of cold weather, WITHOUT the effect of the heater drawing down the battery.  And since I am already way out on the edge of the range limit, I go to perhaps ridiculous lengths to get this data without clouding it by other variables.

So let me ask, how many other Mini E drivers have even gotten a range estimate such as the one pictured  at the top of my blog?  Over 140 miles is quite rare I expect.  Maybe most drivers rarely see over 100.

Let me ask how many Mini E drivers regularly commute over 100 miles per day?  I know there are some, but I suspect it is less than one percent, and most or all are in Southern California, not the frozen North of New Jersey and Pennsylvania.  And the one I read about has a 240 volt charger at work and at home.  I have only 120 volts at work.

So when I describe wearing 3 to 5 layers of clothes, preheating the Mini E with a portable heater, and drinking 5 cups of hot tea to stay warm on my long drive, this is  NOT behavior that one should expect is required in an electric car, any more that the terrain in the above photo is likely to be encountered in your daily commute.

But some people will always test the conditions out on the edge.  By profession I am a test engineer, and recounting these adventures is a testament to my personal curiosities, not to what the average driver of  an electric car might expect.

Believe me, on the weekends we drive the Mini E just like any car, with the heater blazing, maybe without warm coats in cold weather, maybe driving hard and fast.  And many days now  I have run the heater near the end of my long weekday commute, regularly bouncing the "charge meter" off zero a few miles from home.

But on the  edge I am very careful.  And the car performs amazingly well at the extremes if you are careful.  But you don't have to be anymore careful with an electric car than with the old "fossil fool" cars if you are not trying to reach extremes of range.

11 December 2009

Mini didn't find the problem, so they are coming to visit soon

So I have the snow tires and decals, but they did not find the problem that causes #458 to balk at charging from 240 volts.  So they say they will make a house call.

I'm not sure they understand that the car has to come off a long drive and be at low charge to have a high likelihood of not wanting to charge.  I'll make it clear when they come to visit.  Actually that might make it tricky to set up a visit with those conditions.

Michael with #249 says he has to wait between 10 and 60 minutes to get his Mini E to charge at 240 volts.  But this loaner #008 I have is worse.  For the two days I have had it, it won't even try at 240 volts when I get home, no blinking that stops after several seconds, nothing.  But of course it is happy at 120 volts.  And I have waited over an hour now.

I think the engineers from Mini might have a better chance reproducing the problem if they bring #008 over to my house after driving it to a low charge.

10 December 2009

Same problem charging, new car (temporarily)

Well, loaner #008 does not have the problem that my #458 has with the rear defroster killing the radio with static, but the charging issue is even worse on #008 than on my #458.

Otherwise it drives exactly the same. Gets more notice with those decals.

09 December 2009

I get #008 as a loaner? Not #007 ? But my name IS James!

So my #458 went in for 13,000 mile routine maintenance today, and Jimmy G at Princeton Mini managed to get me another electric Mini as a loaner. Wow. Very nice. But maybe there was a practical reason in addition to Jimmy being such a great guy...

You see, I have been having charging trouble with #458 as a few others in New Jersey have. It has gotten quite consistent lately. Most days I get home and the Clipper Creek wall box trips the Ground Fault Protection when I plug in to charge, and the red indicator lights up on the wall box. At the very least, the wall box stays on but the Mini E's yellow charging light stops blinking after only 6 seconds, 8 seconds at the most.

Tonight I got home with loaner #008, pull into the garage and guess what? Six yellow blinks and the car stops charging. It did it twice. Third time and it goes on to complete charging. That is a fairly normal scenario when my #458 has trouble, unplug and try again a few times and it is OK. Sounds like the problem is in more than one car, and maybe something about my place makes it act up? But what could it be? The line voltage does not drop much when the car starts charging, I have measured it. And the 120 volt slow charger always works.

There was one funny thing today though: When I got home from work and plugged in before going to the dealer to drop off my #458, I parked OUTSIDE my garage and did NOT have trouble charging from 240 volts for the first time in WEEKS. But the loaner did have trouble when in the garage.

This is getting interesting. I rained very hard last night and today, but the sun came out on my way home. However, it usually takes days of warmer dry weather for the charging problem to go away. (My theory is moisture in the wiring or in the battery modules is causing a Ground Fault Interrupter to trip.)

So could there be a difference parking outside on the blacktop driveway that was in the sun versus inside on the damp concrete? Maybe damp concrete is a better conductor than sunny bitumen and with damp tires, any leaking electricity could get to ground and trip the safety protection and stop the charging? I don't usually get home when the sun is still up but I left early today so I could get the loaner on time.

OK, I'm not going to do anything for a couple days so I can see if things with loaner #008 are exactly the same as my #458 or not. (Hmm, maybe this is WHY they loaned me another electric Mini? To see if it has the same problem with my charging box?) Over the Christmas break I should have time to make some measurements in the Clipper Creek wall box with an oscilloscope, current probe and high voltage differential probes. But it is too busy right now, and I hear that Mini has started sending engineers to the garages of some Mini E drivers that are having charging problems. So maybe they will get to it before I do. I hope so. I am an electrical engineer, and I have a healthy fear of high voltages. I would rather they figure it out, but I am getting excited about fixing this and I might not be able to wait...

I can't stop wondering what would happen if I dry out the inside of the Clipper Creek box with a hair dryer. I had to do that months ago with the little yellow portable charging box, and it worked perfectly. (For the post about that episode click here.) But even if something like that works, it does not give specific enough information on the root cause of the problem.

Or maybe I will park on pieces of wood to see if that insulates the car and stops the Ground Fault trip. At least that would not destroy evidence like the dryer might.

Ah, I know! Next time it trips and the red indicator lights up on the Clipper Creek wall box, I will measure from a metal tire rim of the Mini to ground and see if there is any voltage that could trip the GFI. Charging always works after letting it sit for ten or twenty minutes, maybe that is long enough for the tires to dry off? If this finds leakage, I bet the leakage it is out of phase with the hot line from the 120 volt charging cord. That would explain why it only has trouble with 240 volt charging, the only wire that is leaking is the one that is grounded on the 120 volt box.

This sure is getting interesting. The solution is so close I can almost smell it. Being an engineer is fun. The future is almost here!

Anyway, they probably didn't make a # 007 Mini E... Or a # 013...

05 December 2009

Hard data on the heater

Robert with #304 posted a great article on the heater in the Mini E.  It says the "PTC" heater (I think that means Positive Temperature Coefficient) can put out 3000 watts.

Mini said at first that the battery has 30,000 watt hours of usable capacity.  (Later reports indicate slightly less if I recall.)

So running the heater at maximum for one hour should consume 10% of the range.  I have never run it at full for more than a few minutes, I usually back off to half at that point.  But theoretically, this means the two hour driving range (usually enough to drain the battery fully) might shrink to an hour and a half roughly, if it were cold enough to run the heater at maximum the whole time.  At lower settings, the range reduction should be less than 20%.  Maybe much less.

I measured heater power usage at low fan settings and posted them way back in September here.  But this morning I measured again at the highest fan and heat settings with the windows open.

After 80 minutes at max heat the charge meter dropped 18%, which is a bit more than theory predicted above, but reasonable.  It was in the upper 40s Fahrenheit and I left the hatch open.  Temperature one inch from the heater vent was above 110 F with the fan at max, and the temperature was higher when the fan was slowed down.  At fan setting #2, the temperature went off my scale, above 130 degrees F.

I have stayed away from using the heat on long drives.  This is partly because of the reports other bloggers have posted about reduced range, but mostly because I'm from Minnesota where only the weak use the heater until the weather is below freezing.  Ha.  But I'm starting to think I can afford to use it more on my long commute.  These numbers are encouraging.

While the charge gauge is pretty reliable, I find it to be pessimistic early in a drive and I almost always have more range than it indicates.  I wonder if the calibration gets worse when the heater is on?  This might explain some of the concerns that seem worse than the numbers I took.

One technical point about a "PTC" heater.  It is designed not to overheat.  So if you run the fan on low and turn the heat all the way up, the resistance of the heating element increases so that won't over heat and burn out.  My measurements showed that the heater can maintain a 60 to 70 degree increase in temperature at the highest fan setting.  But often I run it on lowest fan, and I notice it heat up and then cool down to just warm.  This is what it is designed to do.  Like Mini says, use at least the second fan setting with higher heat settings or the element will not reach maximum heat output.

By the way, 3000 watts is about 4 horse power.  The 120 volt AC heater that I use to preheat my Mini E with "shore power" is 900 watts, and I usually run it for a couple hours by timer before I leave for work.

04 December 2009

120 miles and freezing at the start

I have  almost 13,000 miles on #458 and it finally happened - I got to work (60 miles from home) and unexpectedly had to head back home without charging the Mini E.

No Problem!  I had even used the heater for half of my trip to work, something I rarely do but it was 29 degrees F on my way in and I did not have the hot drinks I usually have in the car to stay warm.  So I ran the heater some after the cabin cooled off.  (I had preheated the cabin to 88 that morning as usual, but after 20 miles it drops into the 60's.)

I have driven my commute several times very gently to see what I could do for range, so I was confident.  I had plugged in at work but left after only one hour, so I added only about 4% to the charge before I left again for home.

I took back roads and really took my time.  I have a column in my log to record how aggressively I drive, this is the only time I ever put down a "one" on the scale of one to five.  It was mid morning so traffic was light but I still annoyed a few people by going slowly.  On highway 12 in New Jersey, I drove maybe 30 mph and pulled onto the wide shoulder to let others pass at the 50 mph limit.  On the big hill I got in the truck lane and crawled up at 20 mph.

And I arrived with 1% charge left.  I even used the heater for the last 3 miles.  I know there are ten miles left after the gauge hits zero, but I didn't need it.  All in all I drove 124.6 miles, using the heater at the middle setting for about 30 miles.  Starting temperature outdoors was 29 F, ended at 49 degrees with 58 degrees in the cab.  In the middle I charged for one hour at 120 volts 12 amps adding only 4 miles to the range.

Despite the minor issues with this prototype car, we love it.  It has never let me down, even when pushed to the extreme in cold weather.  And I bet Mini will fix the minor charging problem next week when it goes in for the 13,000 mile checkup.  Actually, it is getting to be a bit worse than minor.