Thursday, February 28, 2013

Car Show

The car show that was scheduled last week on Wednesday was rained out, but luckily they pushed it to Wednesday this week.  I made it, and here's how I got there!

Two days before the show I decided I needed to fix the leaky heater core.  It's been leaking since I first put it in, but it's such a pain to get at I never bothered with it.  Well, since all my coworkers will give be crap about it and I'm tired of washing coolant out of the driveway it's time to get my hands dirty.  A few hours later I've got a paper gasket and some RTV in the flange and it's leak free!

I've driven the car around the neighborhood a bit, but not a whole lot.  The trip meter just rolled over to 0.1 miles.  It's still very sluggish and only seems to get up to about 5 mph (not enough to register on the speedometer).  There is a parameter in the controller that allows you to limit the maximum motor amps, and I'd set that parameter to 1 for initial testing (corresponds to about 104 amps) and later moved it up to 2 before I first drove it.  It seems worthwhile to bump it up to 3.  At 200 amps I figure I have 16 ft-lb of motor torque, and assuming friction, rolling resistance, hills, etc, I can see why that would be pretty slow.  Using the current squared rule, I should get 2.25x the torque by upping the amps to 300.  A friend is coming over with the car trailer soon so I don't have much time to try it out.  Acceleration is noticeably better, but top speed is hardly any better.  The trailer shows up the night before the show and we load the car.  To get up the ramps, my friend actually has to push a little.  Good thing there aren't any hills to climb at work!

The next day after a restless night, we drive in to work.  We unload the car and drive in (slowly).  Inside the campus there's a long stretch where I manage to accelerate to white knuckle speeds of 9 mph!  Good thing there's a 10 mph speed limit so I can pretend to have an excuse.  When I park the car I already have a crowd forming around me!  Too bad I had to shoo everyone away because I was already late for a meeting, but I get to breathe a sigh of relief that I'd actually made it!  0.5 miles on the odometer, my longest drive ever!

The show was a lot of fun.  60 different cars of all types: 3 Ferraris, a top fuel drag racer, a bunch of classic and modern muscle cars, a couple VW busses, and one electric car.  After 2 hours of answering questions I have to once again close up shop and go to a meeting.  

I'm disappointed in the driving performance.  With the way it's set up I should have plenty of torque to get around and with 80 volts I should be able to go at least 35 mph.  After work I pull out the laptop again and plan to take some real time data while we drive out.  Hopefully I can find something wrong that I can fix.  Of course it can't hurt to toggle the current setting up another notch to 4.  Acceleration is noticeably better, but this time I max out at 11-12 mph (the speedometer is barely above the first line!).  I timed my exit so I wouldn't have anyone waiting on me, but my dad (who also work here) zipps up in his TR4 and heckles me a bit for being a slow poke!  We loaded up the car and drive it safely home.  All in all it's a good day.  I'm fairly confident that there's something in the software that's severely limiting the performance.

On the drive back to the trailer, my friend took a couple screen shots of the data as it streams by.  The current commanded is about 60, which corresponds with about 100 amps, and the current feedback is the same.  So for some reason it doesn't seem to want to go above that. 
All these parameters have a max value of 512, so the value divided by 512 gives you the percent of max (except the thermistor).  The current readings are incorrect because we've tricked the controller by changing the current sensor and didn't change the software.
Green = Thermistor Reading (73 °F)
Red = Throttle position
Pink = PWM (about 17%)
Blue = Motor Current Feedback (controller thinks it's about 60 amps but it's really 100 Amps)
Light Blue = Battery Amps (controller thinks it's 10 amps but it's really 17 amps)

Luckily, after looking at this data, Joe figured out the problem the next morning!  When we tested the controller the first time back over the Thanksgiving weekend, we had it hooked up to some RC plane batteries and a treadmill motor.  In order to keep from ruining the batteries (which we probably did anyways), we changed one of the parameters to limit the current from the batteries to 10 amps.  We never removed this restriction, so I'm still limited to 10 amps coming out of the battery.  Since the current sensor is calibrated differently, it's actually 5/3 of 10 amps (or 17).  At 80 volts, that's only 1300 watts or 1.75 horsepower, so no wonder we could only go 12 mph!  Most hairdryers put out more power than that.

Stay tuned for an update on the performance after I monkey with the settings this weekend.  Including 0.8 miles at the car show I've now driven a total of 0.9 miles!

Sunday, February 17, 2013

It's alive!!!

That's right folks, I've brought life to my creation.  Today I actually drove my car!  Not very far, but it certainly moved on its own.  Here's how I got there.

I started Saturday morning and worked 14 hours until I didn't have any physical or brain power left.  By the end of the day Joe made the comment "yeah man, your mind is wasted."  I started the day by setting up more batteries to charge.  I got them all up to 3.41 volts, then strung them up in parallel so they could all top off exactly the same.  Here's my brother setting them all up.  Don't cross those wires, there's potential for 10's of thousands of amps here!

 It's also time to finish off the wiring.  Here I'm getting ready to make all the high voltage cables.  I've hooked up all the cable terminals where they need to go, and sheesh there's a lot of them!

And they're done!  Too bad there's a crucial error in here that I didn't realize until taking a last minute look at my schematic at 9 pm.  Luckily it's not a showstopper so I'll correct it when I pull the batteries again.

Here's the manual switch for the high voltage system.  When I flip this I'll get +230v to the controller through a precharge resistor.  When I turn the key, the B- contactor closes and completes the circuit to the controller and precharging starts (which takes about 1.5 seconds).  The B+ contactor closes when you first start to move the accelerator pedal, initially it does nothing.  In order to protect the controller and make sure it has a chance to precharge, you have to push a little red button on the dash that activates the switch hooked to the pedal.  Got it?  Okay, let's move on.

And here's the point where my mind started falling apart.  Joe showed up with his laptop and computer so we can test it out at high current.  Back over the Thanksgiving weekend we tested the controller at 80v and about 20 amps, but that's were we stopped.  There's basically no easy way to get higher current or voltage without hooking it up to a car.  It'd been so long I forgot how I we still had more to test on the controller.  At this point, Joe took out his phone to take a video of me starting it up for the first time.  I think his words were something along the lines of "If it works, we've got a great video of it.  If it's a spectacular failure, at least we'll have a cool video for all the effort!"  Luckily, "all" we heard when I hooked up power was the clack of the contactor closing.  After getting the motor to spin out of gear, we lifted the rear end, popped it in 4th, and spun up the tires.  Nice...

So that was Saturday.  Today I had a little tidying up to do before I was ready for a test drive.  All the cables were hanging out the bottom, a few things weren't screwed on very well, and I wanted to check over a few things I did in my frantic state the night before.

With everything in order I clear out all the tools, drop the hood, and hop in the car.  I start backing down the driveway (yeah!!!) and start turning so I can avoid running into my Jeep at the end of the driveway, and what??? why isn't the steering wheel turning???  I get to the end of the driveway and it feels like the steering column is locked.  I drive back up the driveway and shut everything off.  Ok, whew, it worked and nothing blew up.  But why the heck can't I steer?  Long story short, 7 months ago I made a major error that somehow didn't surface until this moment.  The motor mount bracket that cleverly snakes around the steering shaft isn't so clever after all.  If you turn if far enough, one of the clamping bolts on the shaft runs into the bracket.  Oh well, nothing that can't be overcome in and hour with a dremel, a die grinder, and a reciprocating saw.  I'll add that to the list of things to fix properly between now and taking it on a major road.

So back in the car, down the driveway, do a couple laps around the cul de sac, and it works!  I still have a bit to go to get it ready for the show, but overall I'm pretty happy.

Thursday, February 14, 2013

Time is running out...

Dang, I had a nice long update written out here and somehow it didn't get saved.  Oh well, you'll just have to deal with this.

I've got 5 days left before the car show, so it's definitely go time.  A week ago I even took a day of vacation to try and catch up because I was falling behind!  A few days ago I admitted to myself it was going to be too risky to try and drive the car all the way to work.  To get all the way in I'd have to have all the batteries in place.  The problem with that is my charger is weeks away from being ready, so I'm stuck with charging 10 at a time with a small charger.  And charging 10 at a time means I'd have to remove the rear batteries.  It's not really feasible to test the car at full voltage, then remove, charge, and replace them.  A friend at work said he could trailer me in to the show, so I think I'll do that and drive it in from the parking lot on 80v (the front battery pack).  That took a lot of pressure off finishing the rear racks and allowed me to switch to wiring.  So that's the plan, I won't drive it from home but should be able to drive in under my own power.

Speaking of batteries, I started charging them this evening.  The charger runs off 12 volts and needs 300 watts to run at max power, so that means I need to supply around 25 amps.  A regular car battery would be drained pretty fast at that rate, but I've got something way better.  The DC-DC converter I got for the car runs great on 120 Vac, and it puts out over 600 watts.  Just had to sacrifice an old PC cable and here you have it!

Only problem is the fan makes a fair amount of noise.  And that's actually an understatement because it sounds like a small vacuum cleaner is running in my kitchen!  Just 5 hours or so of this and I'll be ready for the next 14 batteries...

So to make up for last week, I can finally say it.  Nice rack!

I finally bolted this baby down and she's ready for batteries.  Though I've got a little more to do before that.  Here are a couple pics of the rear racks in process.  This one's all done except for paint.

And here I've got my liquid cooling pump.  I was sure glad to get this all hooked up and not see water leaking out of the controller!

I don't have the pictures yet, but I've got most of the 12v stuff wired up.  That includes the fans, pumps, etc.  All I have left are the contactors and power to the controller.  Once that's done I have the high power cables to build.  Next is putting in batteries, and after that...testing the controller on the real batteries!  Duh duh DUH!  Keep your fingers crossed for me, I'm hoping to do a test drive this weekend.

Saturday, February 2, 2013


A long awaited day has finally come.  I have my batteries!  I planned all along to wait until I'd finished everything that was feasible to finish before ordering the batteries in order to avoid losing out on warranty time.  The car show has pushed it somewhat faster, so I still have a little work to go, but with any luck I'll be ready to start fitting them in the battery racks after this weekend.

They actually arrived a lot faster than I thought.  I ordered them on Tuesday figuring they'd come in next week, but they were actually at the Phoenix UPS freight center by 10:30 Wednesday morning waiting to be picked up!

Ok, enough small talk, it's time to get technical!  A few weeks back, the 787 was grounded worldwide due to lithium battery fires on a couple planes.  Ever since then I've been fielding questions about when my car is going to "inevitably" burn itself into a smoldering hole in the ground.  Well the answer to that is hopefully never, but I have a few reasons to say that.  To start, there are a couple different varieties of Lithium-ion batteries out there.  For years I've been flying RC planes on what we always called Li-po (lithium polymer) packs.  I'm not sure if there really is a difference between the polymer or ion names, but I think they're pretty much the same thing.  These have a lithium cobalt oxide (LiCoO2) cathode and have the highest energy to weight capacity out there.  Each battery has a single cell nominal voltage of 3.7 volts.  Well, all the cell phones, laptops, and now planes that have caught fire also have this style of battery.  In the last 4 years or so, a new type of lithium ion battery has become economically viable, and that one has a lithium iron phosphate battery (LiFePO4) for the cathode.  If you look up LiFEPO4 on wikipedia you'll find some comparisons between the two.  The LiFePO4 battery is 14% heavier for a unit of energy storage, but is inherently much more stable then the LiCoO2 cells.  They have a nominal cell voltage of 3.2 volts, and are more environmentally friendly to their cousins without the cobalt.  I'll let you read more if you want, but it's supposed to be a superior technology.  So why didn't they use that on the 787?  Well, they probably picked out the battery 10 years ago before this type was really available.  Unfortunately, changing battery types at this stage would probably lead to a year's delay in getting the planes back in the air due to the lengthy process required to certify stuff on planes.  At any rate, enough about the aircraft industry, let's get back to my car!

In the last post I listed all the stuff I need to get done in order to be ready for the car show on the 20th.  Since then I've finished 60% of the battery racks, mounted the heater core (for cooling the controller), removed the motor and modified the flywheel adapter (it now has a runout of a mere 0.0004 inches), and reinstalled the motor.  This weekend I need to finish the battery racks, get the liquid cooling system all set up, and mount the vacuum pump.  I also out to tie up some loose ends with the controller.  If I can get that all done, I'll be on a good path to do a first test drive in 2 weeks!  Just in the nick of time to make the show.  Let's cross our fingers that I don't run into any insurmountable snags.

Alright, here's the part you've been waiting of the progress.  Here's the interior almost all put together.  Still quite a few things to touch up here, but way better than before.

Here's a single cell for my battery pack.  I also got a bag of stainless fasteners and bus bars to hook them all together.  

Here's the heater core turned into a motor controller cooler.  Took me forever to figure out where to put this, (I've been thinking about it for literally months) and finally figured it out.  It's hung in the perfect place right in front of the AC condenser.

I've got a lot of work done on the racks but apparently I didn't take any pictures yet.  Too bad, I was hoping to work in the line "nice rack" somewhere.