AC Wiring diagram? Questions for you EV guys

[brucey]

Working on my electric today and realized the diagram I have on hand isn't the most useful.



First Question: I get where the throttle goes and since I have a 2 wire (PB-6) throttle, I won't need the black and white wire. However, my throttle has 2 wires and 3 contacts on it. (Visible in this picture at the bottom left of the throttle.) Do these connect to anything?

Second question: The Orange/White, Blue/White, and White wires all go to the solenoid. My solenoid has 2 larger posts that look like they can handle some amps, and 2 tiny ones. I'm guessing the big posts go inline with the actual battery cables. But I still don't get how I'm supposed to wire it looking at this diagram. I've attached a picture of the solenoid at the bottom. Adding to the confusion, on my wiring harness the Orange/White and the Blue/white wires are relatively short. While the White wire is 4 times as long. So I guess they shouldn't be going to the same point.

Third Question: The fuse itself. Where should I put it? I was thinking of putting it at the 'end' of the batteries. Like say on the positive terminal. Or should I put it in between them? Does it really matter?

4th Question: The White/Red and the Blue ones. It is connected to Menu push button then just joined up at a switch to turn the system on and off? Then the red goes (with fuse) to the battery+ on the other side of the switch? That doesn't seem right, looking at my wiring harness.

Well that's all the confusing questions I have today, any help would be awesome, I'm going to keep googling until then.

[TomEV]

The three contacts on the bottom of the PB6 are probably the throttle switch. One of the contacts will be the common (probably center), one will be normally closed, and the other will be normally open.

In an AC motor system, the throttle pot can actuate regen as well. Before you use a two-wire setup, (instead of the full yellow/black/purple on the diagram) you may want to research if your system has regen, and how it is actuated.

The wiring diagram you have shows the 'control' portion of the controller.

The portion not shown is the HV wiring. The AC motor usually has three large cables that go to the controller, and the controller has two cables that go to + and - on your HV pack. Some controllers have a 3-cable system from the main pack. In this case, the B- goes to pack -, the B+ goes to pack +, and the high-amperage fuse goes between F+ and B+.

Your main contactor can be used to turn on power to the + or - cable from the pack, but sometimes the main contactor is omitted.

[brucey]

Haaa this big AC stuff is new. I remember on my small DC set ups I'd just have to have 3 phase wires and a battery connection.

Thanks for the help Tom! I'll ask more questions when they come up.

[DJBecker]

That's the idle position switch.

It can inform the controller that the pedal has fully returned. This is important because the physical structure around the TPS (throttle position sensor or "pot box") should set the physical travel limit, not the potentiometer itself. This makes it appear that the pedal is always slightly pressed. When the switch closes, the controller knows that the motor should be at idle/off.

A software-based controller can use the resistance/voltage/current value read as the baseline for "go pedal" actuation, while an old analog controller can use it for an easy "high pedal lockout" implementation.

Good controller software should allow hooking this switch up, and occasionally update the idle value to handle any change in the TPS response due to heat. Without a 'closed throttle' switch the controller has to be a little smarter: keeping a "deadband" where it doesn't activate the motor when it appears there is a slight pedal application, and updating its idea of the full-off position if this condition persists.

[DJBecker]

I am curious why people use the Curtis PB-6.

I'm used to automotive TPSes. Bosch has a good selection with available datahsheets, they are small, inexpensive, and have proven to be very reliable in harsh conditions. The four wire ones even have dual variable resistors, with the second one only reporting fine pedal movement in the first 25% or so of travel.

[HPEV]

I think people use the Curtis units simply because they've been around and so many others use them.

Personally, I prefer a Hall effect type unit myself.

How inexpensive are the Bosch units you speak of? That sounds intriguing.

[DJBecker]

The Bosch TPSes are very common. Just like oxygen sensors, most TPSes out there are Bosch or Bosch-licensed copies thanks to Bosch's lead in computerized engine control when emissions regulations started.

They can be found for $20-$100 online, but they are readily available from the boneyard for less. When you get them from the boneyards around here they don't charge extra for the mounting hardware or connectors/wiring (if you don't go overboard -- 1-2 feet of wire) which can be a big time savers vs. buying new.

Here is a datasheet that covers most of the Bosch units:

Bosch - Throttle Position Sensors

[sawickm]

Originally Posted by DJBecker

I'm used to automotive TPSes. The four wire ones even have dual variable resistors, with the second one only reporting fine pedal movement in the first 25% or so of travel.

Have you interfaced a TPS to a ReVolt Controller ? How did you do it ? Schematic ?

[DJBecker]

We have been working on our own controller design.

The two output Bosch TPS is a direct connect, and allows us to use two 8 bit ADC channels and still get good resolution. It also provides the potential for a software cross-check, which we should probably add at some point. The Bosch TPS has multiple wipers over the surface to avoid "skips" from dirt or vibration-induced wear-through, but it could still happen.

[DJBecker]

Just an add-on: I'm thinking about ordering Paul's control board, so I may be trying out a Bosch TPS on it.

That would also let me contribute code and suggestions with a hope of having them be directly useful.

I have lots of experience in writing device code from specs. I once even wrote a network device driver for hardware that was halfway around the world and got it working in a single weekend. But it's way easier when you have the board right in front you.

As a semi-introduction, I've written a bunch of open source software. I've written many (60-70) device drivers, many of those for Linux, as well as many other pieces of software.

Our motor controller is still design in the early stage. We are trying to keep the "smart-side" requirements as generic as possible so that we can eventually put it in the display unit.

The mid board sits with the power board. It contains the isolation interface, which takes three mutually isolated PWM channels and an I2C bus from the smart side and uses it to talk to the gate drivers and an I2C ADC for measuring the traction and gate voltages.

Our test "mule" is a Advance MTC4001RM coupled to a Miata transmission. (Our target car is a classic Miata, not yet purchased.) The coupler is a standard SH taper bushing and the spring hub of a Miata clutch plate (a stock clutch with the friction disk removed) joined with an aluminum piece I turned on a Very Big lathe during a Very Fun machine shop day. I also turned a smaller hub that might work with a flywheel if clutch-less operation doesn't work out.

There is no load hooked up to the mule, only the transmission. We were originally planning to use a 10KW generator head on the transmission output, but the generator purchase fell through. We bought a used PennTex Industries PX-4-G5 alternator (200 amps continuous at 6000 RPM) that uses external regulation, but building the mount and figuring out a regulator circuit (plan: 42V output to simultaneously charge the three lead-acid batteries that power the mule) is a future project.