Tuesday, May 24, 2011

Some Over Due Updates

I mentioned a couple weeks back that I had some changes planned for the EV-Z3, and the time has come.  I'm not going to spell them all out now.  Rather I'll write about each as I tackle them, and go into detail as to why I decided the change was necessary.

Air Conditioning:

It doesn't take someone on a full ride scholarship to MIT to see why I'd want air conditioning in a car in Arizona.  Last summer, I thought I'd tough it out and put up with what I remembered to be "minor discomfort" when driving cars with no AC in my youth.  But I must be getting old, soft, or both because after a couple trips I'd had enough and the car started to see action only in the evenings when the temperature dropped to around 100 °F, or on rare occasions into the 90's.

You may remember that I had managed to install the original compressor into the car and planned to run it off a pulley mounted to the tail shaft of the motor.  I didn't have it hooked up because the compressor was in a different place in the engine bay, and the hoses didn't reach.  Well I'd just thought I'd have them remade once I was up and running.  With space being as tight as it is, there was no way to do that work without taking out the batteries.  It became clear I was going to have to remove the batteries in order to get AC.  So, into the garage the car went and up on jack-stands. 

There's an inherent problem running the AC off the tail shaft.  When the motor isn't spinning, you can't compress the refrigerant in the system.  I could sit with the clutch in and idle the motor, but I don't want to do that. I want a car that works normally, and not one where I hand the keys to someone and have to give them a talk about how to make the AC work properly.  That meant I need to run the compressor off a secondary motor.  I had bought a little 2.5 HP DC motor, originally used on a tread mill, back when I started the project, specifically for this purpose.  The problem is there simply isn't enough space to put that motor, the compressor, and the metal assembly needed to hold them together into the car.

Most people into EVs have heard of or seen the small DC powered compressors made by Masterflux.  Great product, and a perfect solution for me.  They are small enough they will fit in the space I have.  The only problem is the compressor and the controller needed to power it come in at over $1200.  In the scheme of things, that's not too much, but I'd rather avoid that expense if I can.  As it happens, someone is selling a number of older, new stock of these compressors on eBay at the moment.  The good news is that they are asking $200 for the compressor and controller.  The bad news is they require 48 volts input and can draw up to 18 amps.  So, I bought one.

Now, I have no way of delivering 48 volts to the system.  Especially at that level of current.  A friend of mine, who is quite savvy with electronics has offered to help me build a power supply for the compressor.  I know what you're thinking, and I can't really say your wrong.  I'm an idiot.  But I figured, the worse case scenario is that I can't build the power supply, and I simply re-list it on eBay.  But if I can get it to work, I've got a great little AC unit, at 1/5 th the cost after I figure in the cost of parts for the power supply.  I haven't convinced you have I?  I don't blame you, I'm not convinced yet myself.

Controller Cooling:

A second big issue I need to solve is better cooling for the Zilla.  The radiator I'd originally installed to cool the Zilla is 4x8 inches, or 32 square inches of area.  I installed two 120mm DC fans that continuously pull air through the radiator in an attempt to keep things cool.  It did just great 9 months out of the year!  I had enough room for a bigger radiator, and I found one that uses just about all that space.  Here it is right next to the old one.


It's 8x14 inches for a total of 112 square inches.  That's 3.5 times more surface area than the smaller one.  And the fan that came with it moves 331 cubic feet of air per minute, while drawing less than 5 amps.  The old set up had the fans running all the time.  During most of the year, and certainly when the cars moving through the air, this is a waste of energy.  What I intend to do with this set up is run the fan off a thermostatic switch.  I have a switch that will turn on at 122 °F, and turns off at 104.  The only thing is that it's a surface mount switch.  That means I'd have to find a way to physically press it and hold it to the Zilla, and preferably the area on the Zilla that gets the warmest.  That happens to be dead center on the top.  Hmmm...  Not sure how I'm going to accomplish that without venturing into the steam-punk genre of EVs.

What I'd really like is an inline thermostat I can splice somewhere in a hose.  The only problem is that all the ones I've found are for ICE cars so they come on at over 200 °F.  Not suitable for my needs.

Removing some junk from the trunk:

You may recall that to monitor the batteries, I'd originally installed some test lines which I ran to terminals in the trunk and in the front of the car.  The idea was that it would make it easier to test the batteries to find if they were different voltages to each other.  Well, I've found that's unnecessary.  Since bottom balancing the pack, I've learned which cells hit the top first when charging; and by top I mean 3.45 volts.  Of course the real top end of the batteries is 3.6 volts, but the difference in the actual energy stored in a battery that's 3.45 vs one that is 3.6 is so small that it's simply crazy to try to push that extra little into the cell.  Especially considering the crazy fast rate the voltage will rise above 3.6 at the end of the charge and the damage that can be done if it goes higher.

To safeguard those few batteries, and in turn all the rest of them, I have the charger shut itself off when those few cells get to 3.45 volts.  If I want readings off of other cells I just take them.  Having those extra wires in the car was not only ugly, but a fire hazard.  So out the came.

I had installed some relays in the trunk in an ill fated attempt to turn off the DC to DC converters when the car is not in use.  The problem is that those Iota converters have a large bank of capacitors in them that store energy quite nicely.  So when I would turn the car back on, there was a fair amount of arching inside those poor relays, and they welded themselves shut in 3 or 4 cycles.  The only reason I wanted the controllers off was that in the summer, the ambient air temperature is warm enough that their fans run constantly.  Slowly, but constantly.  I may address that at some point, but for now those relays and all that extra wiring was not needed.  So out it came.

I also decided that I would eliminated the spaces in between all the cells in the battery box in the trunk.  You are supposed to clamp all LiFePo4 batteries together between something that can keep them from swelling.  Apparently they have a tendency to bulge in the middle a bit after multiple charges.  Well, I'm not so sure about that.  I took off all the connecting straps that I'd made so that I could leave an 1/8" between each battery, removed all the hardware anchoring them down and shoved them together.  They nested up snugly, one to another just like when they were new.  Keep in mind that this is after one year's worth of driving and 287 charging cycles!

Giving some thought to this, I've come to believe one of two things is going on.  Either I simply haven't cycled the batteries enough to see this expansion they speak of, or my practice of not charging the cells to 100% full capacity has removed this danger.  While I have no conclusive proof, I'm leaning toward the second option.  Knowing what kinds of nasty things these cells can do above 3.6 V and how it affects them, it stands to reason that charging them to 3.6 V every time you charge them has got to put strain on them.  Regardless, I'm removing the gaps and re-securing them in the space.

Anyway, look how much cleaner the trunk looks without all those extra wires and components.


Many more enhancements to talk about, and of course I have to document the conclusion of each of these items.