The acceleration control potentiometer.
Available as a part of drive system only
The most used control in the vehicle - throttle control. In fact it's in *continuous* use, thus very high reliability requirements are imposed to this system. Likely, most of the parts used in ICE operation are reusable and since design bugs are worked out a long ago, it is encouraged to use as much of existing system configuration as possible.
The most power controllers and inverters require a potentiometer at the input to control the output. Some have option for opto- or inductive sensor, but can work with regular resistive element as well. Particular inverter I use requires a pot with any value between 1 kOhm and about 10 kOhm. Since it acts as the voltage divider, absolute resistance and its stability is not relevant. What relevant is durability and convenience of integrating into existing throttle mechanism.
The throttle potentiometer (better term is acceleration potentiometer) I use is made for OEM electric vehicles. It is highly reliable, enclosed, dust and water proof, already contains return spring and two contacts inside, and equipped with the lever on its shaft. The mating surface of the lever and the shaft has micro grooves preventing slipping but allowing fine adjustment of the initial lever position anywhere within 360 degrees. Made by Bosch, it conforms to all automotive environment standards, such as low/high temp, vibration, etc.
The body of the pot has two flanges for mounting. Initially I wanted to wrap existing throttle cord around the pulley installed on the shaft, and even made this stainless steel pulley, but ended up with using stock lever. I only needed to make it shorter since full travel distance of the cord linked to the pedal is only 30 mm, and I wanted to utilize at least 90 degrees of the pot rotation. Actual resistance in Ohms at both dead ends is not important since whatever it happens to be, it can be assigned by the inverter software to be 0% and 100% motor current demand respectively.
Two tasks were ahead - to pick the most convenient place for smooth operation and service, and fabricate suitable bracket to mount it. Even though the pot is waterproof, I didn't want to subject it to the water in case of rain or if I wash the car. I had limited length of existing throttle cord and somewhat limited choices, especially because other components are positioned pretty tight. I chose to, mount it to the top of the fire wall. Once good spot was found, I had to make rigid enough bracket to mount the pot. It has to have the place for the threaded end of the throttle cord sleeve allowing minor length adjustment. The work began from making a paper template for the bracket, it's just the trial and error process. I made a hole in the template for the pot, folded the paper to resemble future bracket and tried to "install" it. By trimming the corners and changing the bending angles to accommodate the shape of the place it's going to be bolted to, I approached the final version. About 3 iterations were enough. Next step is to make a mock up from a thin steel, the metal can for tomato juice was cannibalized, cut flat and shaped to the future bracket. The material is easy to solder, so if I cut out too much, repair is quick. After I was satisfied with what the bracket look like, finally real one was made. I wanted the bracket to be one solid piece, so I cut out the corner of an aluminum box which already had nice weld in the corner. Eventually when all was done and the lever parts were assembled, the effort of making two templates and many iterations paid off - the bracket fits as it was made by Honda, sleeve adjustment is perfect and operation is as smooth as it can be. Perhaps this is the most professionally looking part made by hand!
Acceleration potentiometer as it comes from
The work starts with the measurements. The pot lever is taken off, connector cut off and the wiring strapped to the pot body.
The lever is shortened. Other small parts to attach the lever and the throttle cord.
The first iteration of making the pot bracket.
Let see, it's going to be here. Does it fit well? Yes. Does the lever touch the firewall? No, but still better move the pot a bit away.
Final version of the bracket - painted and ready for the pot.
The pot mounted on the bracket, lever installed and whole assembly ready to be put in place.
The pot is installed and the throttle cord position adjusted. I'm proud of this OEM looking assembly.
Regen control potentiometer.
It took me several weeks to think what is the best place for the regen control. The first obvious choice - activation by the brake pedal - turn out to have serious shortcomings:
- I have to modify existing braking system
providing slack for the pot control before braking pads are engaged in action;
- In case of emergency braking the time before the pads engaged is increased;
- The braking action is always applied to the front wheels first, there is no way to bypass it and have all four wheels start braking simultaneously as before. This means that on the wet or icy road the car will be prone to skidding as the rear wheels still roll free.
So, finally I've decided to use manual hand control, allowing to bypass regen all together if I have to. Existing braking system is not touched and still independently functions as before. For the minimal braking effort I picked sliding linear pot and installed it on the shift stick. Driving my other ICE car I was trying to envision how would I activate regen and the best position for it from ergonomic and safety stand point. The solution was to place it vertically in front of the stick with default position wiper down. The height was picked such that I can easily reach the slide wiper with my fingers while the palm is resting on the top cushion, and pull (lift) spring loaded wiper up to activate regen. The top position is max regen current; when the wiper released, it slides down and regen stops. This proved to be slick and extremely convenient design - I don't even have to move my foot away from the throttle pedal to slow down the car. The top portion of the pot is placed in the slot in plastic palm rest, and is held in place with RTV sealing compound. To get nice grip whole unit was dipped into "Plasti Dip" compound used to mold the handles on the hand tools.
I did not want to have any springs hanging on the pot and pulling the wiper down, so decided to mount return spring in remote location, linking it to the wiper with the flexible cable in the jacket (similar to the hood release cable). This also allowed to mount a little addition to the assembly - brake light switch. Since the brake pedal is not going to be touched (other than for complete stop at the red light), the brake lights somehow should get activated when regen is on. First I thought of the clamp type current sensor on the battery cable - when current is flowing into the battery, the brake lights are on. Actually, this solution turn out not to be so good: if the current is small, the turn on point becomes not very reliable. So I mounted an amplified IR optical sensor near the cord and attached a small shutter to the cord, so when the wiper is lifted, the shutter gets out of the sensor gap and relay turns on, turning on the brake lights. The sensor (Omron EE-SX672) has an amplifier built in and designed to directly drive a relay. Very convenient - I only need two components then.
You may ask why not just use a micro switch? Well, two reasons. Firstly, ask any hardware designer what is the least reliable component in any electronic equipment, and the answer will be - connectors and switches. I didn't want to compromise this important function - reliability of the brake light switch is crucial safety concern. Besides I have only two components instead of one micro switch - very simple circuit and mechanical arrangement. Secondly - I wanted to have return spring to be soft, so I don't have to pull the pot lever with much effort. If I'd use a micro switch, the pot return spring would suppose to be stiffer to overcome the tension of the switch return spring. With the optical sensor the movement is smooth without any mechanical obstacles. And it works like a charm.
So now the magic happens - I can block my front wheels by *literally* lifting one finger, no other motion required. That's the power control! To start accelerating again I just press the throttle, and leap forward, even before regen pot is released and reached down position - inverter software takes care of prioritizing the throttle pot function over regen pot. What a pleasure to drive the car now! When I brake (especially going long downhill), watching the battery pack being charged and the counter of spent battery Ah running backwards, my EV grin now seem to be proportional to the regen current, can't help it :-D. Sweet!
The brake (regen control) pot and the shift
stick palm rest cushion with the slot for the pot.
The wires are connected, the pot is molded in the cushion with GE rubber sealant and is ready to be dipped into "Plasti Dip".
Coating whole thing will provide better grip and uniform surface. The pot slot is protected from the compound getting in.
Close up of the optical sensor.
Rested position - the shutter (single pointing arrow) blocks the gap. Relay is off and so the brake lights.
Regen activated - relay (and the brake lights) on.
Assembled control unit.
The end of the cable where the link to the slider attached, and the link made of the stiff wire.
The pot is installed on the shift stick, the wires are connected to dedicated inverter's input.
The cable is linked and its jacket attached to the stick with a clamp. Works really well.
Finished assembly (slightly rotated toward camera for better view).