The yellow tube is a small fuse in series with a resistor (Value 25W, 750 Ohms, the gold item bolted to the plate), across the SW-200 main terminals.

This allows a very small amount of current flow into the capacitors in the controller and allows them to be 'precharged' (stops current rush on turn on which can harm them). This means that even with the relay off, there is actually 132v at the controller.

Full turn off/isolation can only be achieved by pulling the emergency stop switch.

The Crash Sensor cut out switch (the item with the rubber top) mounted on the throttle cable flange.

The Crash Sensor switch is similar to an airbag trigger. It consists of a weight, held by a magnet, in between two spring terminals, in a crash, it's momentum pops the weight out and breaks the circuit. Press the top to push it back into the springs and 'reset' it.

The Crash Sensor spec sheet. Page 1.

The Crash Sensor spec sheet. Page 2.

In the set position contact is made between the two outside terminals. This breaks in a crash.

The "Ooops! The charging cord is still plugged in" circuit is just a microswitch connected to the gas flap door. It only closes (allows current to the start relay) if the door is closed fully. (ie no cord plugged in). I had planned to use a standard car door open/closed interior light push switch, hence the hole I drilled in the door closed flange. But the spring was too strong to allow the flap to 'push' it.

NB. CEV supplied the male socket and marine charging lead, and, for some reason supplied a 30A 125V only configuration, even though the charger (a Manzanita) can accept 220V. I plan to swap this out for one rated at 30A 125/250V, which is Leviton part number 3334-C. The matching cable mount female plug is Leviton part number 3333-C, matching cable mount male plug will be Leviton part number 3331-C, finally the wall mount receptacle is Leviton part number 3330. (Had to order these from the US, my Canadian electrical wholesale supplier didn't stock them.)

With the batteries in the rear end sags 'till I can't get a hand between the wheel and wheel arch. I'm fitting Firestone Coil-Rite Air Assist bags to cure this.

They fit inside the rear springs, and need a way for the inflation hoze to get to the end of the bag. So, after removing a coil spring I drilled a hole in the center of the bottom spring mount.

In theory the knuckle should be easily removable by undoing and withdrawing the bolt at the bottom. It wouldn't undo, despite liberal pentetrating oil over several days. Can't use heat due to the rubber bush, and trying it on a scrap car I sheared the head off, so I'm dropping the strut down instead to get the spring out.

Here is the sheared off bolt, both ends sheared(!). This knuckle is actually from a Swift GTI which has disk brakes on the rear. I was collecting a pair from a scrap yard with the intention of doing a drum/disk swap at some point in the future.

I need to see if the rubber sleeve assby can be pressed out of the knuckle and replaced. Part # anyone?

Geting the bag inside the spring is a pain. I deflated it and held it squashed with cable ties, made it a lot easier.

The airbag has an inflation hose situated in the center of the end. I drilled a hole in the center of the spring seat in the lower arm to allow the hose to pass through.

Bag sitting in spring, bottom view.

Bag sitting in spring, bottom view. With air hose attached. Also a half inch thick spacer dropped over the end.

Bag and spring back in the vehicle. The air hose passes through the hole in the bottom arm, and will be routed to an inflation tire valve in the rear trunk. Both sides will be independantly adjustable to allow separate leveling of each side. (Note the spring has yet to be rotated to locate the end in the notch in the baseplate.)

Firestone 'Coil Rite' instruction sheet.

5 to 35 psi is the recomended inflation range.

Prototyping the motor mount in eighth inch ply and hot glue..

I had to make the chassis rail part that high to allow the rubber mounting (which is wider than the rail) to fit into the space where the wheel arch goes away.

Prototyping the motor mount in eighth inch ply and hot glue..

Motor dropped down a bit to allow access to the top right bolt hole.

Hot glue and ply wood again, a great way to take templates. (I use this a lot to scribe kitchen counters to walls, etc..)

Prototyping the motor mount in eighth inch ply and hot glue..

Notice the washers also hot glued to the template. I literally put the bolts through the washers into larger holes in the ply and then into the motor's tapped holes. Then squirted some hot glue behind the washer and did the bolt up up.

Hot glue sets in about 30 secs, just by cooling.

A perfect template ensures... ;-)

Prototyping the motor mount in eighth inch ply and hot glue..

Above: Welded article, which will be welded to the chassis rail. (Rats! Shame I'd painted the chassis after the front end repairs...Oh well..).. Note the red engine mount on top.

Below: Ply mock up.

Motor mount drop arm. Finished, welded article, 1/4" plate.

I hope to read motor RPM and use it to drive a tacho. Here's the specsheet for the Hall effect sensor I found. A matching collar will go on the motor's secondary shaft. It will put out a pulse each time a magnet on the collar passes it.

The firefly's tach used to be for a 3 cylinder engine, which gave 1.5 pulses per rev. One and a half magnets is not possible ;-) I hope to recalibrate it for 4 cylinder engine, then 2 pulses (magnets) per rev should do nicely!