Throttles

[Pete Prodoehl]

I've been doing research on motor controllers and throttles and (correct me if I'm wrong) it seems that the more expensive controllers can use cheaper throttles (0-5v throttle) while cheaper controllers need to use more expensive throttles (0-5k throttle.)

I think I get how a 0-5v throttle would work. Similar to connecting a potentiometer to an Arduino and running 5 volts through it. (Though I guess some are now hall effect sensors rather than potentiometers.)

So what about the 0-5k throttles? Is there just a 5k potentiometer in there? What makes it different? How does it work?

I want to start hacking, but I'm hoping someone more electrically advanced than myself can provide some good info on this...

Thanks!

Pete

[Tom Gralewicz]

There are generally 2 types of throttles in use for electric vehicles:

1)  Pot box - the old standard, been in use for 50 years, and yep its just a 5K pot.  The bad news is these tend to have more moving parts, exposed contacts and resistor elements etc.  They tend to be industrial control type stuff and as such aren't made in big quantities and tend to garner a premium price - $100 and up for the commercial ones.

2)  Hall throttles - used on budget electric scooters and toys.  They move a magnet past a hall sensor, no wipers, no exposed contacts, fewer moving parts.  Mass produced and cheap - under $15.

In theory they both do the same thing:  Given power and ground wires, they provide a variable voltage between the two that the controller reads.

The problem is the Hall throttles generally take 5V in and output 0.8 to 3.6V.  Pots can take any voltage in and output 0 to full voltage.

As such, they tend not to be interchangeable.

It is possible to use an Arduino to read one type of throttle and output the correct voltage for the other, I'm currently using an arduino to generate Hall signals for the motor controllers in the easy chair I'm building.  It would be best to use a digital pot to keep the older controllers happy, just make sure the digital pot can handle the voltage.

The challenge comes if the controller wants to see more than 5V, some of the pot controllers can look for strange voltages.

I have one now that wants 6 to 12 V input from the throttle.

Easy solution:  Use the right throttle for the controller.

Cool solution:  An arduino that takes Hall input and outputs both Hall and Pot.

Bonus solution:  Have the Arduino monitor battery current and cut throttle accordingly

Double bonus:  Have the Arduiono also monitor battery voltage, warn when low, count AH used, etc.

Tom Gralewicz

Chronic Maker

(414) 380-1716

milwaukeemakerspace.org

--
You received this message because you are subscribed to the Google Groups "Power Racing Series" group.
To unsubscribe from this group and stop receiving emails from it, send an email to power-racing-series+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Ray Scheufler]

There are also controllers that use a 5K pot in a two wire configuration.  They have an internal fixed resistor and use the varying external resistor to change the voltage divider.  This ends up with some non linearity but does have the advantage that failure conditions are easier to detect.  With a two wire pot, if either wire comes disconnected it spikes to an infinite resistance which is easy to check for with a threshold.  If the wires short together, they have 0 resistance which can easily be designed to be 0% throttle.

A three wire configuration on a potentiometer can have some interesting failure conditions depending on which wire failed.  If the middle voltage input goes open then you have a floating input and that may cause uncontrolled acceleration.  If the low pin goes disconnected the wiper will be connected to the high pin which might be intrepreted as full throttle.  If the high pin goes disconnected the wiper will be at ground which will be safe (0% throttle).

The reason that a hall throttle has an output not full range is probably to give the software a chance to protect against broken wires with some range sanity checks.

If you are designing your own throttle input, consider what happens when things break or get shorted.  You probably don't want to go full throttle.

Ray Scheufler

To unsubscribe from this group and stop receiving emails from it, send an email to power-racing-series+unsubscribe...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[Jay Dellinger]

I use an arduino between my throttle (hall) and the motor controllers. That way I can just change code to do whatever I need it to do. It allows you to add nifty features. I use a potentiometer on my dash to scale my max throttle output, in case I want to limit that (e.g. if my kids are driving the vehicle around my neighborhood). Since I use one throttle and two motor controllers & motors, I can also effectively shut down the motors independently at run time by reading the state of switches on my dash.

I also use some toggle switches and push buttons to support a poor mans cruise control. The toggle enables/disables cruise control, while the push button locks in (or resets) the current throttle input value. Once enabled and set, the arduino continues to output the same voltage to the motor controller. It's useless in a race, but it's cool for demos or parades. I don't currently have an onboard speed sensor, so it's not a true cruise control...it's just an electronic throttle lock, basically.

I agree about making sure things fail safe, and avoid the full throttle runaway. I power my arduino off my main battery pack and just step down the voltage with a cheap buck converter.

[Pete Prodoehl]

Thanks, Tom!

I'm thinking through things like cheap versus safe and COTS versus DIY... And about what a team just starting out could handle/understand without getting too complicated.

So if I wanted to use a microcontroller (in place of a 5K throttle) would I just run the (desired) power through a MOSFET and control it with a PWM pin to give the controller the proper voltage? Then I'm just using a pot on an analog input. Is that right?


Pete

[Pete Prodoehl]

Any chance you have some wiring diagrams or even photos?


Pete

[Tom Gralewicz]

Ray, you've just touched the surface of the variations of pot throttles.  Thats why I recommend using a digital pot when interfacing to one.

Pete,

As Ray pointed out, there are many varieties of Pot throttles so just outputting a voltage to the controller may or may not work.

Ideally you want to use something like this:

http://www.digikey.com/product-detail/en/microchip-technology/MCP4162-502E-P/MCP4162-502E-P-ND/1874154

Controlled by SPI interface.

Tom Gralewicz

Chronic Maker

(414) 380-1716

milwaukeemakerspace.org

[Pete Prodoehl]

Ahhh, digital potentiometer? Cool!

I'll grab some, and try this: http://www.learningaboutelectronics.com/Articles/MCP4131-digital-potentiometer-circuit.php

Looks pretty straightforward, and within my domain.


Pete

[Pete Prodoehl]

After a good 6 hours of screwing around with digital potentiometers, I'm convinced that at least one of these things is true:

1. Digital potentiometers work, but not how I expect them to work
2. Motors and motor controllers work, but not how I expect them to work
3. I have no idea what I'm doing
4. I'm going to need Tom's help

Most likely, three of these things are true.

Pete

[Tom Gralewicz]

Let's sit down and play some time.  The theory says it should work but I haven't tried it myself.

[Chris Hemmerly]

Post schematics for review?

[Tom Gralewicz]

To unsubscribe from this group and stop receiving emails from it, send an email to power-racing-series+unsubscribe...@googlegroups.com.

[Pete Prodoehl]

You know how you can spend hours on a problem and just can't solve it, but when you come back a few days later you figure it out in 10 minutes? Yeah, that...

I should do a full write-up for this (and I will) but it turns out everything worked fine... perfect, in fact, except for me, who failed to noticed the value of the original potentiometer that came with the motor controller. It wasn't a 5K pot, it was a 100K pot! Which explains why the behavior of a 5K pot was what it was... which matches the 5K throttle that arrived (and was tested) recently.

So yeah, I think I've got it figured out, and I have some 100K pots on the way, which may not really matter when we get to real-world performance, or may dramatically increase the capabilities of the controller/throttle combo, and provide a programmable throttle at 1/3rd to 1/5th the price! (And yes, I'll open source it all, document it, and tell everyone who will listen.)


Pete

[Tom Gralewicz]

What controller are you using that has a 100K throttle?

They usually go with something under 5K for noise immunity. 

Tom Gralewicz

Chronic Maker

(414) 380-1716

milwaukeemakerspace.org

[Pete Prodoehl]

I'm experimenting with this one right now: http://www.ebay.com/itm/262136734386

Rizzo Racing uses it, along with a 5K throttle, and I think that will work fine, but I'm also 'experimenting' and trying to learn things along the way.

So yeah, I may not know what I'm doing, but I'm gonna keep trying! ;)


Pete

To unsubscribe from this group and stop receiving emails from it, send an email to power-racing-se...@googlegroups.com.

[Tom Gralewicz]

Hmmm, I would expect it to have a micro processor on board so it should be looking for 0-5V from the pot.

With the original pot connected what voltages do you see from the wiper?

It should also have a 5v supply on board that can power your arduino - saves you adding an external regulator to handle 36V

If its 5V, I would just use the analog output from an arduino (RC filter) and connect it directly.

Tom Gralewicz

Chronic Maker

(414) 380-1716

milwaukeemakerspace.org

To unsubscribe from this group and stop receiving emails from it, send an email to power-racing-series+unsubscribe...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "Power Racing Series" group.
To unsubscribe from this group and stop receiving emails from it, send an email to power-racing-series+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Pete Prodoehl]

Tom, when the semester is over I'll start coming back to the space, and we can investigate more... but in the meantime, I did get it working. I now have an Arduino and digital potentiometer taking input from a throttle (or analog potentiometer) that can control the motor via the controller. And I can see how it would be easy to add "cruise control" and a speed limiter, etc. Nice! (I suppose a next step might be to add monitoring/sensing to keep an eye on the motors/fuse/etc and respond appropriately.)


Pete

To unsubscribe from this group and stop receiving emails from it, send an email to power-racing-series+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "Power Racing Series" group.
To unsubscribe from this group and stop receiving emails from it, send an email to power-racing-series+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Sean Denny]

Wow! a $13 motor controller (w/free shipping and overhead)? I'm cautious because of my experience with $25 controllers, but I wish you success. Looking forward to your write-up.

Sean

[Pete Prodoehl]

Yeah, Rizzo Racing used the same controller on both cars this year and they worked well. One did blow up (capacitor blew) when the cooling fan died. (And it was like 95 degrees in KC this year.) These controllers do need to be actively cooled, they run hot! I'm also designing a few pieces to hold a fan in place and provide a quick mount/release method just in case one does need to be swapped.

Pete

On 11/1/16 9:14 AM, Sean Denny wrote:

Wow! a $13 motor controller (w/free shipping and overhead)? I'm cautious because of my experience with $25 controllers, but I wish you success. Looking forward to your write-up.

Sean

--

[Pete Prodoehl]

I finally finished my write-up of what I've been doing the last six months...

    http://rasterweb.net/raster/2017/05/29/controlling-the-controller-cheaply/

Comments and feedback welcome!


Pete

[rynait]

with throttle, there are alot of choices. however i got the misfortune of head scratching in san mateo on why my minecraft atv car had weak throttle resolutions (hall sensor) on throttle: made from china.

later, disassembled throttle at home. i discovered there are several different hall sensors. falling into 2 categories; one is reasonably digital (to arduino digitalRead), or very analog (use with ardiuno analogRead).

anything that I is noted compatible to either s49 or ss49 branded sensor is suitable for digital reading.

our sensor was branded sx49 and search on web discovered it a) discontinued product, b) ineffective for digital reading, i switched connections to analog read on arduino (any of analog ports) got my throttle "back".

so what you buy, make sure it works to your design or plan.

roy

[Pete Prodoehl]

Wow, I didn't realize there were throttles that used a digital input...
got any links with more info? Are they using SPI or something?
Otherwise, I don't see how that would work at all.


Pete

[J.Ashinghurst]

Hall effect sensors can be analog output or digital output. Some of the
cheaper throttle use digital hall sensors, so the throttle is either on,
or off. sometimes two digital hall sensors can be used as a two-stage
throttle. (Brice's old pocketbike used one like this I think).

While I'm sure you can get fancy hall sensors with SPI, he's referring
to digital latching hall sensors that are either on or off.

[Pete Prodoehl]

Ahhh, okay, that makes more sense if it's just on/off. I had what
appeared to be an on/off twist throttle with two speeds on an electric
scooter I tore apart.

I'm guessing you wouldn't want something like that for PPPRS as it's not
much better than the sparkplate method, right?

Pete