Proportional Throttle/Braking/Reverse with RC speed control

[UselessPickles]

My son received one of the Kawasaki ATV Power Wheels for his birthday and I was quite disappointed with the following:

• What would normally be a brake pedal on motorcycles is actually the primary throttle of this Power Wheels.
• The throttle on the handle bars is only used to activate "high speed" mode (only if the pedal is also pressed; does nothing otherwise).
• The throttle isn't proportional! Just on or off. It's very twitchy, and it actually scared my kid.

I decided to solve all of this by adapting it to use a speed controller for an RC car (a.k.a. "ESC", for Electronic Speed Controller). Some of them are very configurable for things such as "punch control" (smooths out twitchy throttle transitions) and torque limiting (to protect the gear boxes), among other things. Some even have regenerative braking and anti-lock braking. The torque limiting would be especially useful because it could allow me to use fast and powerful motors to get higher speeds without straining the gears with too much acceleration.

My plan is to modify the throttle tube to rotate more (about 1/8 of a full rotation) and link it up to a slide potentiometer. I'll also link up the pedal to a slide potentiometer. I'm a programmer and have a friend that knows electronics and how to program microcontrollers, so we will work together to create a custom circuit that converts the potentiometer positions into a single throttle servo signal, as expected by the ESC.

The throttle on the handle bars will be forward throttle and the pedal will be braking and reverse. ESCs treat reverse throttle as braking if the motor is currently spinning forward, and will not go into reverse until the motor has come to a complete stop. They can also often be configured to require that the throttle is returned to neutral for a second or two after stopping before re-applying reverse throttle to go in reverse.

Once I have the custom electronics and physical modifications complete, I will be able to remove the entire wiring system of the Power Wheels and replace it with any ESC of my choosing. I'll start by keeping the stock motors, but could easily use a high-voltage capable ESC, brushless motors, etc. I'll also be able to use lithium polymer batteries (a.k.a. "lipo" - light weight, small, can be charged faster, maintain voltage under amp draw better, etc). The ESC has built-in low voltage cut-off to protect lipo batteries from over-discharging. You can't use lipos without low voltage cut-off, because if you run a lipo down until you notice it's slowing down, the damage has already been done.

I'll post updates and pictures to this thread as I make progress :)

[UselessPickles]

The first step was to work on the physical modifications to the throttle and pedal so that they actuate slide potentiometers. After some searching and measuring, I settled on this potentiometer (picture of the website is not an exact match; the one in the pic is longer): http://www.mouser.com/Search/ProductDet ... 32015CPB50

Here's a pic of the main supplies I used for the physical modifications:


That's a potentiometer, various thin pieces of steel for making brackets, and countersunk M3 screws with nuts for mounting the custom brackets to the Power Wheels parts. The thin steel wire (19 gauge; galvanized) is what I decided to use to act as the linkage between the throttle/pedal and the potentiometer. It's soft enough to be easy to work with, yet stiff enough to move the slider without flexing.

Here's how the pedal turned out...






(sorry; didn't get a pic of the pedal in its original state)

• Removed the original heavy-duty spring-loaded pushbutton switch.
• Added a spring around the guide screw to return the pedal. I picked up a few random springs at a small hardware store that seemed promising, chose the one with the best fit and feel, and shortened it a bit to reduce the force and to allow full depressing of the pedal. That rough-looking black plastic washer is made out of the cap to a Powerade bottle. it keeps the spring from jamming up in the slit that the screw passes through. A metal washer binds up the motion of the pedal too much (and makes horrible noises).
• Used a combination of a drill, a Dremel with a cut-off wheel and grinder bit, a vice and a hammer to make the custom mounting bracket to hold the slide potentiometer in place.
• Made a small "L" bracket and mounted it to the bottom of the pedal.
• Drilled a small hole in the tab of the slider and in the small "L" bracket for the wire to pass through.

The throttle is a bit more complicated. Still working on it, but that will be the next update...

[UselessPickles]

Got the throttle done!

First, here's what it looked like before I got started:



Twist the throttle, and that little arm presses the red button.

Here's what it looks like when I'm done hacking it:




I had to clear away some plastic in the housing to make room for the potentiometer and to allow the throttle to twist more:


Mount a tiny custom "L" bracket to the little arm on the throttle tube:


Drill a small hole so I could hook a spring onto the arm:


And make a custom mounting bracket for the potentiometer (had to cut the lever shorter for clearance purposes):


The return spring isn't quite as strong as I'd like. The throttle does close by itself, but it doesn't "snap" closed. I'll probably take another look at the hardware store for a stronger spring the same size, but this will be sufficient if I can't find anything better.

[UselessPickles]

Quick update with a video...

I got all the wiring removed from the Power Wheels and re-wired the motors to plug into the ESC. I borrowed a radio system from one of my R/C cars to give it a quick test. If I was going for an R/C Power Wheels, all I'd have left to do is rig up a steering servo and plug it into the receiver.

I have lots of pictures of the process and will post details later, but it might be a couple days before I have time to do that.

[UselessPickles]

Now for some details on how I rewired the Power Wheels to use a speed controller from an R/C car. I'll break this into a couple posts because there's lots of pics (there was a fingerprint on my lens for many of these pics - sorry for the blur!).

The throttle grip and pedal were already removed for modification, so here was my starting point:


None of that wiring is necessary, so let's get rid of it.

The easy part is the portion of the wiring in the handlebar area. The handlebars had to come out first.






You can see what limits the steering in the above photo. The steering linkages for the wheels can travel much further, so you could tighten the cornering radius by simply trimming away some plastic in the area shown above. Just be careful to not allow the handlebars to turn so far that they could jab your kid's stomach.



There's the handlebars. Four screws in the bottom opens up that plastic part.



And now you can easily remove the wiring there. Just snip the zip-tie and cut the piece of tape.

To be continued...

[UselessPickles]

Next was to disconnect the forward/reverse switch. I'm still trying to think of ideas for what the switch could do now that I won't need it to toggle between forward and reverse.

Two screws hold the switches mounting plate, and it lifts out with a bit of finessing.



The white connector just pulls off, but it's pretty tight.




Now that the switch is removed, the main shell of the body can be removed to get it out of the way. There's just 2 screws under each fender (8 total).



Remove one screw and slighly pry the two halves of the chassis apart so you can get the wires out.




Now the wires are trapped in a narrow channel at the bottom of the chassis.



Remove the support rods for the footpegs.



And remove these five screws along the bottom of the chassis:



Now you can slightly pry the two halves of the chassis apart near each of the plastic tabs covering the channel just enough to get one wire out at a time.



To be continued...

[UselessPickles]

To finish removing the original wiring, the rear wheels need to come off. Start by popping the little caps off.



Then you use a wrench on each side at the same time to remove one of the nuts, and everything just slides out.



The gear boxes are not actually mounted to the chassis. They're just sitting there with a black piece of plastic holding them from moving too much. Remove the 3 screws and the gear box will slide right out.




Motors are held to the gear box with 2 screws.



You'll notice that the gear boxes are identical. There is no left vs. right.

The wiring is almost ready to come out, but there's a small circuit board mounted inside the chassis.



It's held by 2 screws (lower right corner and upper left corner). I was able to remove the 2 screws by contorting a bit and sticking my arm in through the hole for the gear box. It then takes some effort to pop the board off the two pegs (lower left corner and upper right corner).




Removing the wires from the motors was a bit tricky. there are female quick-disconnects crimped onto the wires, which slide onto tabs on the motors, but then they are also soldered. I had to put the motor in a vice so I could simultaneously melt the solder with a soldering iron and pull the connectors off the motor tabs.



The removal of the original wiring is complete!

[UselessPickles]

Now for something that many people might be interested in for creating a radio controlled Power Wheels: wiring up a speed controller that can controlled by the throttle channel of a radio system.

I'm using a Traxxas VXL-3s ESC because it's just an extra I have sitting around. It can handle up to a 3-cell lithium polymer battery (11.1v) and is rated for 200A continuous and 320A "burst". I think it should be able to handle the current draw involved in a Power Wheels. It also has a "training" mode that cuts speed and power in half. There's plenty of other ESCs out there that could work even better because they can handle more voltage and amperage and are more configurable (torque limiting to keep the gear boxes safe, for example). When my kid is ready for more speed, I plan on trying the Mamba Max Pro ESC by Castle Creations with 6-cell lipos (22.2v).

You're probably thinking, "11.1v? That's gonna be slower than the original setup with the 12v battery!". Lithium polymer batteries (lipos) can supply tons of amperage and don't drop in voltage much under load. I haven't measured the speed with a GPS yet, but I'm pretty sure that my 2-cell lipo (7.4v) is nearly the same speed as running the original 12v battery.. I have some actual measurements now, and it's slower than I thought, but still faster per volt than the stock battery. See here, here and here.

Other benefits are that they are much smaller, lighter weight and can be charged faster. The original battery has a capacity of 9.5Ah and weighs 10.5lbs. Two 5Ah 5.3Ah 3-cell lipos (11.1v) wired in parallel would give you 10Ah 10.6Ah capacity, would weigh about 3lbs 1.8lbs (more details here) and would fully charge in about 1.5 hours on a decent lipo charger (could charge faster if you have a charger and power supply that can charge at higher amps; as quick as 15 minutes if you have some real serious equipment).

Anyway... on to the mods. The motors need to be wired up so they can be connected to the ESC. Bullet connectors are typically used. Some ESCs have female bullet connectors pre-installed, some don't. Just make sure you know exactly what kind of connectors you need. In my case, it was 4mm bullets. So you'll need wire (I suggest 12g wire from a hobby shop, made for high-powered RC cars), connectors, shrink tubing and soldering tools/supplies.



Also pictured above are 2 "Y" harnesses I made with one male connector and 2 female connectors each. These will be used to connect the two motors to the ESC in parallel.



Figure out how long you need your wires so that they're reach the ESC. Shorter wires are generally better, so try to find a place near the motors for the ESC to call home. Solder wires to the motor's tabs, and add male connectors to the wires. Notice that the red and black wires are connected to opposite tabs on the two motors. This is because the motors need to spin in opposite directions.



Now I'm not actually making an R/C Power Wheels, but you probably want to, so I'll show you how everything gets hooked up with a radio system.



Motors plug into the ESC (that's the blue thing with black cooling fins). This ESC can run brushless motors, so it has a third wire that will not be used with these motors. Make sure you understand how to hook up the motors to your ESC and configure the ESC to run the motors properly (RTFM). If you get everything setup and find that the vehicle goes the wrong direction, you can swap the wires, or your ESC might even have a "reverse motor direction" setting.

The battery (that's the shiny blue rectangle; it's a 5Ah 7.4v lipo battery) plugs into the ESC. Make sure you know how to setup the ESC. A special mode is necessary for lipo batteries (RTFM).

The ESC plugs into the receiver (that's the tiny circuit board; the plastic case was removed so it would fit better in the RC car that I borrowed it from) on the throttle channel (channel 2). The ESC supplies power to the receiver, so no additional batteries are necessary. Make sure you know how to calibrate the ESC to your radio system's throttle signal (RTFM).



There you have it... fully proportional and radio-controlled throttle, braking and reverse. None of the original wiring or switches are necessary. It's modular so that you can easily swap out the various components for upgrades. All that's left to do for a fully radio-controlled Power Wheels is to rig up a steering servo and plug it into the receiver. For my project, I'll be making my modified throttle grip and pedal generate the same signal that the receiver would provide to the ESC. I'm thinking about making it support some kind of radio-controlled override so that I could take over the throttle while my kid is driving it :)

NOTE: The ESC, receiver and battery are not secured in my picture. ESCs and receivers can usually be simply mounted with double-sided servo tape from your local hobby shop. I still need to work out a way to restrain the battery. It will probably involve some foam padding and a velcro strap.

And just for good measure, here's the video of the radio-controlled throttle in action again:

[haus]

have you tested this out under load? with your son riding it? in the grass? interested if the ESC can handle the current draw. im planning a full RC jeep build and this info would be very helpful. thanks

[UselessPickles]

Good point there haus... don't everyone go rushing out and doing this just yet assuming everything will be perfect just because I said I'm doing it.

I haven't done any extensive testing yet. I have done some quick indoor tests. It had no problem hauling me around (135lbs). I definitely will do some more extensive testing and post results later. I'll get temperature measurements of the ESC with an IR temp gun. Many ESCs have thermal protection (automatically shut down if they get too hot), so it's pretty safe to experiment if you think you're at least in the ballpark.

I don't expect any problems. The stock battery has a capacity of 9.5Ah and I've seen reports of "about 1 hour ride times under average riding conditions". That would be an average amp draw of about 9.5A. The car that this ESC came with (Traxxas Slash 4x4) gives me about 30 minutes of run-time at the race track under hard racing conditions with a 5Ah battery. That's an average amp draw of about 10A. The ESC is also rated for 200A continuous and 320A burst current. I'm pretty sure that'll be enough :)

The only problem I think might come up is that this particular ESC is known for getting hot pretty easily, especially on a 3-cell lipo (11.1v). There is a cooling fan available for it, and some mods to allow some airflow into the vehicle would definitely help. I suspect that I'll just upgrade to the Mamba Max Pro ESC (or maybe even the Mamba Monster) when I start using higher voltages. Those ESCs come with cooling fans pre-mounted and stay very cool. My kid is only 2 years old, so I'll be starting him off with the 2-cell lipos (7.4v) and the ESC in "training mode" for reduced power/speed. It'll be a while before I worry about upgrading.

[UselessPickles]

I did a bit of testing this evening. First, here's some simple speed results. This was done unloaded (kid was in bed already, and I'm too heavy; weight limit is 65lbs) on a flat, smooth asphalt driveway. Original motors, gearing and wheels.

Traxxas VXL-3s ESC (training mode) + 2-cell lipo (7.4v): 1.5 mph
Traxxas VXL-3s ESC (non-training modes) + 2-cell lipo (7.4v): 3.4 mph
Castle Creations Mamba Max Pro ESC + 4-cell lipo (14.8v): 6.2mph

I don't have any 3-cell lipos, so I couldn't do a 3-cell and 6-cell test. It's slower than I first predicted, but it's still a bit faster per volt than the lead-acid battery. Based the ratio of mph increase vs. voltage increase in these results, I calculate about 4.7 mph for 3-cell (11.1v) and about 8.5 mph for 6-cell (22.2v).

The VXL-3s was barely warm (74°F in 58°F ambient temps) after driving around in the grass for about 10-15 minutes in "sport mode" (mostly unloaded; about 1-2 minutes with all 135 lbs. of me on it) with a variety of acceleration and braking. For reference, I have regularly seen 130-140°F when running the ESC in my RC car. I don't know exactly at what temperature the ESC goes into thermal protection shutdown, but seem to remember reading somewhere that it's up near the 200°F range. I know the Castle ESCs allow up to 200°F before going into thermal protection shutdown.

Preliminary results look promising. I still need to do some longer runs with my kid on the seat in warmer weather before I can declare any setup to be solid. I still predict that I'll have to go with the Mamba Max Pro ESC for anything higher than a 2-cell lipo if I want to never worry about ESC temps.

I do have a bit of a disappointment: the "torque limiting" feature of the Castle ESCs is only applicable to brushless motors. There are still several useful setting that work with standard brushed motors that will help protect the gearboxes: "punch control", braking strength (0-100%), and independent max forward/reverse speed (0-100%) settings.

[UselessPickles]

Progressing slowly...

I started getting wiring prepared to connect everything to the custom electronics (which I should get started on this weekend). I have 3 things that need to connect to it: throttle, pedal and the ESC. Since an ESC has a servo wire/connector for plugging into a receiver, and my custom electronics will be taking the place of the receiver, the obvious choice was servo wires and connectors to plug them together:


Those are Futaba "J" connectors by the way. I accidentally only ordered female connectors, but I'll eventually also need the male connectors.

With the female connector and the wire, I was able to create this:


The bare ends of those wires will eventually be soldered to a circuit board containing the custom electronics.

For the throttle and pedal, I have potentiometers to wire up. There's 3 connections to make: ground, voltage supply and a variable voltage "signal" connection (voltage varies based on the position of the slider potentiometer). I conveniently have a bunch of servo wire and connectors that usually provide a ground, voltage supply and pulse-width modulation "signal" connection between receivers and servos.

Here's one of the potentiometers with wires added:


Black is ground, red is voltage supply and yellow is the "signal" wire.

And the pedal and throttle with wires attached:



I have the potentiometers oriented and wired such that the "signal" will be a low voltage ready in the resting position of the throttle and pedal, and the "signal" voltage will increase as the throttle is twisted (or the pedal is pressed, in the case of the pedal).

I'll eventually put heat-shrink tube on much of the wiring for protection, and the throttle and pedal wires will eventually separately connect to the custom electronics through a male/femal pair of those Futaba "J" connectors.

I think I'll work on something to secure the batteries inside the chassis next...

[UselessPickles]

Battery hold-down is done :)

Here's the materials:


Velcro battery straps for RC cars and a sheet of foam. The foam will provide a soft surface for the battery to be strapped down against.

I'll strap the battery down where the original battery used to sit. All the way on the bottom between the two steel support poles for the foot rests.



Making a template out of card-stock paper always helps:



And the foam is cut:



That's for two layers of a foam pad for the battery. The bottom layer needs to be cut up because of 2 plastic ribs across the bottom of the chassis:



After some careful marking and measuring, I cut slits through the bottom of the chassis for the straps to loop through:




And the battery hold-down system is complete!




These straps in this setup seem to have enough range of adjustment to hold between 2 (7.4v) and 5 (18.5v) cell lipos. I'll have to get longer straps if/when I get around to using 6 cells (22.2v).

[haus]

anymore testing updates?

[UselessPickles]

No more testing. I've been busy working on programming the microcontroller. It's pretty much complete torture.

I got an initial version of the software working properly the other day (tested on a simulator), and we put everything together on a circuit board today. Everything tested out perfect on the actual hardware, so we finished routing wiring on the vehicle, hooked everything up, and it worked beautifully... for the first 5 minutes. Then it suddenly quit working.

Now I have to pull everything off the vehicle and wait for schedules to coincide so that I can meet up with my friend at his uncle's house where all the testing/development equipment is :(

[Woodson]

I bought the Mamba and will be working on it all day tomorrow. I have enjoyed reading your thread and appreciate all the assistance.

More questions tomorrow I'm sure. I forgot it was Easter Sunday so I wasn't able to spend the time I wanted on it.

Thanks again!

And GREAT job!

[UselessPickles]

Much thanks to my friend, I have my custom electronics working again. There was much soldering, starting over, and probably tons of cursing before a config setting for the microcontroller program was found to be wrong and causing the new microcontroller to not work. The new circuit board is now safe and sound inside a slightly modified project box from Radio Shack so that it won't die from static electricity again.

I have lots of pictures to post later, but I'll whet your appetite with the final results now.

Here's everything completely installed (with velcro) and wired up:


It's not entirely clear in the picture, but there's two sets of 3 wires each running from the throttle grip and the brake pedal back to that little black box (you can see them in the narrow channel at the bottom of the chassis). The magic black box converts the throttle and brake position readings into a signal that is sent to the ESC. The ESC provides power to the black box and receives the throttle position signal through a set of 3 wires connecting the two.

And here's a short video of the proportional throttle, braking and reverse in action:

More details to come later...

[UselessPickles]

Now for some details on the final pieces of the puzzle...

I spent many hours doing stuff that doesn't make for exciting photos: writing code for the microcontroller so that it will read the positions of the throttle and pedal, then output the appropriate corresponding signal to the ESC. It's some very tedious work. Much more tedious than general application development, mostly because it's so difficult to test and isolate bugs.

Here's some code with the spec sheet for the microcontoller in the background (I spent much quality time reading through that document):


And one of the microcontroller simulators I tried out to test my code and get it working before we even hoped for it to work on the actual hardware:


After many late nights, I finally had some bare minimum code working in the simulator. I hard-coded the end-point calibrations of the two potentiometers, and the output signal to the ESC is directly related to the current measured positions of the brake and throttle. It's enough to make it quite usable, but I have plenty of ideas for improvement to the code in the future.

Once the code was working in the simulator and I knew which pins I would use on the microcontroller, my friend put together this circuit board (this is actually the second complete board, after the first one fell victim to static electricity):



The two connectors toward the top of the picture are the throttle and pedal inputs. the connector toward the bottom connects to the ESC, and the ribbon of wires to the right is used for reprogramming the microcontroller. That's a much bigger microcontroller than we really needed, but it's what was readily available. I'm only using 3 of the 36 I/O pins :). Capacitors are used to filter out noise in the power supply to the chip, and a couple diodes bring the voltage down into the operating range of the chip (max 5.5v; the ESC provides 6v to the board). I never would have known that stuff would be necessary, because I'm just a software guy.

And after the static electricity incident, I promptly modified a small project box from Radio Shack to protect the board:


Before putting it all in the vehicle, we tested the output signal with an oscilloscope (these pics contain the original board that was sadly taken from this world before its time):


The output is a pulse-width modulated (PWM) signal with a period of 20ms (50Hz frequency):


A duty cycle of about 1ms at full reverse throttle (pedal pressed all the way; throttle released):


A duty cycle of about 2ms at full forward throttle (throttle twisted all the way; pedal released):


And a duty cycle of about 1.5ms at neutral throttle (throttle and pedal released):


IT'S WORKING!

Next up: final assembly

[UselessPickles]

Time for final assembly. I finished up the throttle assembly and pedal assembly by adding connectors to the wires and adding some abrasion protection to key areas of the wiring:




Here's an overall view of the entire electrical system so you can see how everything connects:


Added velcro to the magic black box and the ESC for easy mounting and removal:


You may have noticed that the wire coming from the throttle was way too short. That was on purpose. An extension cable meets up with it near the front of the vehicle:


All wires are neatly routed through the vehicle:


And now the magic black box and ESC are velcroed in place and all connected:


IT'S DONE!!!!*

*not really - I'll continue to play around with the microcontroller code and I might even try making my own motor controller that is integrated with the throttle/pedal position processing to skip the need for an expensive ESC in addition to all the custom work :)

Here's the test video again for good measure:

[CJB]

:shock:

I'm not much into RC, but this is amazing!

I need to learn programming code...

[Woodson]

I'm just thankful you pointed me to the mamba max pro. It was worth the money basically plug and play once I understood the software.

What you're doing here is outside my realm of understanding. :shock:

[UselessPickles]

Here's my kid's first ride on the newly modified Power Wheels.

He's only 2 years old (recommended minimum age is 3), so he doesn't quite have the attention span and coordination required for some serious stress testing, but the ESC was barely warm after about 15 minutes of this (60°F ambient temps). I might have to borrow someone else's kid to use up a full battery pack when it's warmer for some real test results.

[UselessPickles]

I got a new battery. Here it is compared to a Power Wheels battery:




It's an 11.1v (3-cell) lipo with 5.3Ah capacity for $48. It doesn't quite match up to the 12v 9.5Ah original battery, but I measured a top speed of 4.9mph with a GPS, so it's close enough on voltage. The battery has 62% of the capacity of the original battery, but I can fully charge it in under an hour.

It also only weighs 14oz (~0.9lbs), compared to 10.5lbs for the original battery :)

[UselessPickles]

I just hooked up a 2-cell and 3-cell lipo in series for a total of 18.5v and measured a top speed of 7.7mph. It was also very fun to do controlled power slides, thanks to the proportional throttle control :) I really need to get one of the Power Wheels made for 2 kids with a higher weight capacity so that I can ride it without worrying about breaking it.

[Woodson]

I just hooked up a 2-cell and 3-cell lipo in series for a total of 18.5v and measured a top speed of 7.7mph. It was also very fun to do controlled power slides, thanks to the proportional throttle control :) I really need to get one of the Power Wheels made for 2 kids with a higher weight capacity so that I can ride it without worrying about breaking it.

Good to know. I'm currently running 12V w/ 12V backup and LEDs / cooler mods through a 6V.

One of the reasons I went with the dual F-150 was weight capacity. I just returned from Destin, FL today and the sugar beach sands would destroy the motors no doubt, plus steering would be difficult.

Phase two ability to maneuver through sugary beach sand? Plexiglass skim plate on servo and plexi glass cube over motors with heat sink fan?


Atlantic side I have it covered but will need to modify for return trip in July... Any ideas?

Again thanks.

[UselessPickles]

Phase two ability to maneuver through sugary beach sand? Plexiglass skim plate on servo and plexi glass cube over motors with heat sink fan?

Some type of skid plate (probably aluminum) would be good for generally protecting the servo from solid objects. For protection against sand in the steering area, you could probably enclose the whole area as much as possible with something solid (again, aluminum would be more durable than plexiglass, and easier to cut/shape as desired). You won't be able to completely block it off, because you'll need to leave slits for the steering linkage. You could cut the openings there slightly over-sized to avoid binding, then mount some brushes somehow to block of the hole while creating negligible resistance against the steering linkage.

For the motors, a plexi-glass cube will keep sand out, but it will also keep hot air in. A heat sink and fan can only be effective if there is a supply of cooler air. Something that might would would be a box-like frame around the motor, then cover the frame with a nice fabric that will keep sand out but let air flow through easily. The fabric would tend to keep the warm air trapped a bit, so a heat sink and fan (inside the fabric barrier; the fan needs to be protected also) will help encourage air movement.

Depending on the exact arrangement of things and the possible entry points for sand, it might be easier to just block off a large area with a simpler (flat) fabric shield rather than trying to build boxes around the motors.

Hot glue also works well for sealing loosely fitted joints (seams of chassis/body parts, holes for wiring, maybe any places around the wheel where sand could get through to the gearbox). It can usually be removed with a bit of effort, but without damaging the parts, if disassembly is needed later. Maybe test the ease of removal on a hidden spot first.

[4runnerhouse]

RC4WD.COM sells a 1:2 gear reduction set for the 540/550 motor that you can bolt on to double the speed

[UselessPickles]

Doubling the gearing for (theoretically) double the speed would also double the load on the motors, which would increase amp draw (exactly double? I don't know for sure). I predict doing so would lead to a much shortened life of either the motors or the ESC.

[kickingrass]

Thats awesome!! all projects should be so well documented. I dont comprehend the coding part, so are the controllers available pre-programmed or plug-n-play for the guys like me who did not pay that much attention in school?? (or paid more attention to other things) :mrgreen:

[UselessPickles]

Thats awesome!! all projects should be so well documented.

Thanks :)

I dont comprehend the coding part, so are the controllers available pre-programmed or plug-n-play for the guys like me who did not pay that much attention in school?? (or paid more attention to other things) :mrgreen:

The circuit board and programming was completely custom. I'm not aware of anything similar that is available for purchase anywhere.

I have some ideas for taking this idea another step to get rid of the need for an ESC. I would create my own ESC, integrated on the same circuit board. This would allow me to have greater control over the motors, add some cool features, and possibly end up with something that I could sell for a reasonable price if there's enough demand. If I ever follow through on this, it will be a long journey. I really don't know much about electronics, so creating my own ESC will likely be a disaster :)

[sirironduke]

first off, i must say, job very well done. wonderfull documentation as well. interesting use of a esc.
my question is, since you are manualy controlling the motors ,instead of making it rc, why not use a dimmer switch set up? they are rated for 120volts and im sure enough amps. i dont have one infront of me to make sure. i could be wrong.
then you wouldnt have to make your own esc.

[UselessPickles]

why not use a dimmer switch set up?

• If it worked, a dimmer switch position would directly translate to power to the motors. To prevent stripping the gears and giving kids whiplash, I need to be able dampen the throttle position to place a limit on how quickly the throttle position changes. I currently do this with code on the microcontroller.
• Dimmer switch wouldn't provide braking or reverse.
• I'd be surprised if a dimmer switch actually worked. They're designed for AC lighting, so I don't know if they'd work on a DC motor.
• A quick search shows that a 600 watt rating is common. At 18v, 600 watts would only allow for up to 33 amps. That seems a bit low for comfort for me, considering that my 6 lb. RC car can draw around 100 amps under hard acceleration.
• How would a hook up a dimmer switch to be actuated by the throttle?
• ...

Processing everything through a microcontroller to decide what the motors should be doing allows for a lot more flexibility/customization, and also safety. For example, code in the microcontroller could refuse to switch into reverse until the vehicle has come to a complete stop. I can also make the vehicle coast more realistically when releasing the throttle, among other cool possibilities (perhaps traction control and ABS?). It would also make lighting (brake and reverse lights) very easy to implement. The microcontroller would already know when the brake is applied and when the vehicle is in reverse, so it just needs to set output pins to switch lights on/off. Most importantly, it's cool to be able to say that I've programmed a Power Wheels :)

[sirironduke]

well i guess a relay and your foot switch could give you reverse, but no dampening or braking. so in that yes the microprossesor is better. the ability of addons like lights is always a plus. abs would be interesting to see.
and yes its quite awsome to be able to say you've programmed a power wheels.

[dukedlf]

thank you for the well documented project here. I am picking up a project has been on ice forever, and a quick scan of this thread seams to answer most of the questions i had unanswered in my head with regard to R/C wiring and power.

[UselessPickles]

New feature added!

Radio-controlled throttle override, so I can remotely hit the brakes for my kids to avoid collisions.



The radio receiver is just to the right of the big black box containing my custom electronics.

More details here: viewtopic.php?f=26&t=6375

[wired]

Great build.

[UselessPickles]

Still tweaking things a bit...

It turns out that setting up the pedal as a brake/reverse wasn't such a good idea. That pedal is positioned in a place that makes it nearly impossible for a kid to not step on it when mounting, dismounting, or even just sitting on the vehicle. I've seen enough accidental full-speed reverses.

My first idea was to wire up the original reverse switch to function again. It would sit between the ESC and the motors so that the switch would reverse the polarity of the motors. I could then setup the ESC to have braking, but no reverse. The pedal will then always act as a brake, but never cause the vehicle to accelerate. The throttle will cause the vehicle to move in the direction determined by the reverse switch.



Mission accomplished! The ESC output wires plug into one pair of wires, and the motors plug into the other pair. The reverse switch reverses polarity of the wires going to the motors. Everything worked as planned, but I still wasn't happy with it for the following reasons:

• The full amp draw of the motors is now flowing through a stock switch. This would be fine if I wasn't planning on eventually running higher voltage (which causes higher amp draw). I'd rather not do this.
• There's nothing stopping a kid from switching directions while going full speed. If this happens, the motors, gear boxes, and possibly the ESC, will not be happy. The kid may or may not be happy depending on what result they were expecting.
• My radio controlled throttle override is now limited to moving the vehicle in the direction determined by the reverse switch.

My next idea is to wire up the reverse switch as an input to my microcontroller so it can take care of deciding how to handle transitions between forward and reverse. To make this possible, I have to remove the brake pedal from the equation completely. Only the throttle will be used now, and I'll have to rely on the ESC's configurable "drag brake" to apply braking force while coasting with 0 throttle input. The state of the reverse switch will control whether the microcontroller produces a forward or reverse throttle signal proportional to the position of the throttle, and the ESC will be returned to its full functionality mode (reverse enabled).

The major benefit with this approach is that I can write the code in a way that prevents switching between forward/reverse while throttle is being applied. If the reverse mode changes while the throttle is applied, I simply output a neutral signal to the ESC (vehicle will coast down to a stop fairly quickly), and the throttle will not actually cause the vehicle to move in the opposite direction until I detect that the throttle has been fully released, then re-applied.

My radio controlled throttle override is again fully functional too. The reverse switch only affects whether the vehicle's throttle generates a forward or reverse signal. When using the signal from the radio system, the radio's throttle position is used directly.

Here's my 2nd version of the rewired stock reverse switch:


I'm using servo wire again because it fits the pattern of ground, voltage supply and a signal wire. The switch causes the signal wire to either be connected to ground or the voltage supply. Since I'm no longer using the brake pedal, I don't even need to modify my circuit board. I can just plug this switch into where the pedal used to plug in and update the microcontroller code to configure that input pin as a digital input rather than an analog input. If the pin is connected to ground, it's "off". If the pin is connected to the voltage supply, it's "on".

Everything works beautifully now, and I'm sure my kids will appreciate the simplified controls and absence of reverse runaways when mounting the vehicle.

[Jornum1]

love the override Useless, wish I had the programming knowledge to add that into my unit.
I figured out a way to use my mambo manually just using a servo tester, but would love to be able to switch back and forth via the controller. I know servo city sells a remote relay, I wonder if I could somehow wire that to my gear selector on my TQ4.
Sorry to jack your thread, you just got me thinking! Great job!

[taz11]

Congrats....... your thread has been added to "Moderator Picks" ;)

[VIDALBABBOON]

Ohhh no.

Im having more ideas....

[Houdeani01]

Thanks for this thread it gave me many good ideas for my project. My pedals are just about done.

viewtopic.php?f=19&t=6503

[18VoltsOfury]

Awesome work. I have been into RC since the early 80's. Was once considering doing a similar mod, but went the conventional route because when I looked up info here on the forum (old forum) about using RC esc's and lipos everyone had negative comments on the idea.

Anyway now I see you have the Castle esc in it. More amps? Now you need to go brushless and drop 1/8th motor in it lol.

Hey for us less skilled that cant or don't want to program and wire up our own PCB's something like this can be modified to simulate the Tx signal.....

[redline]

I bought a similar servo tester recently for exactly that purpose :) ... thinking ill swap the 'round' pot for a ' linear' one to start with :)

[summbear48]

The first step was to work on the physical modifications to the throttle and pedal so that they actuate slide potentiometers. After some searching and measuring, I settled on this potentiometer (picture of the website is not an exact match; the one in the pic is longer): http://www.mouser.com/Search/ProductDet ... 32015CPB50

Here's a pic of the main supplies I used for the physical modifications:


That's a potentiometer, various thin pieces of steel for making brackets, and countersunk M3 screws with nuts for mounting the custom brackets to the Power Wheels parts. The thin steel wire (19 gauge; galvanized) is what I decided to use to act as the linkage between the throttle/pedal and the potentiometer. It's soft enough to be easy to work with, yet stiff enough to move the slider without flexing.

Here's how the pedal turned out...






(sorry; didn't get a pic of the pedal in its original state)

• Removed the original heavy-duty spring-loaded pushbutton switch.
• Added a spring around the guide screw to return the pedal. I picked up a few random springs at a small hardware store that seemed promising, chose the one with the best fit and feel, and shortened it a bit to reduce the force and to allow full depressing of the pedal. That rough-looking black plastic washer is made out of the cap to a Powerade bottle. it keeps the spring from jamming up in the slit that the screw passes through. A metal washer binds up the motion of the pedal too much (and makes horrible noises).
• Used a combination of a drill, a Dremel with a cut-off wheel and grinder bit, a vice and a hammer to make the custom mounting bracket to hold the slide potentiometer in place.
• Made a small "L" bracket and mounted it to the bottom of the pedal.
• Drilled a small hole in the tab of the slider and in the small "L" bracket for the wire to pass through.

The throttle is a bit more complicated. Still working on it, but that will be the next update...

ok...so not even sure if anyone will see this since its so long ago! But If I hook up a scooter controller, could I use this slide potentiometer set up to use original pedal on power wheels to control it?

as I understand, you cant just use this for a regular 12v power wheels, but if it is hooked up to a scooter controller (24 or 36v) you could substitute this for a hall effect twist/thumb/or the plastic foot pedal one right? I could hook up a 3 pin connector to the potentiometer? If anyone knows which positions I would hook up 3 three wires to this slide, would love the input!

[Sedzy]

I know this is an old post, but it's still awesome! Well done UselessPickles! The custom throttle pot is beautiful. This is exactly the level of what I have in the works, but I'm still in the block diagram phase. I can see from the 1 pic, you/or your friend codes C well. (ie proper headers/commenting) I have a question for the PWM.

I guess your PIC is using the ADC to sample the Pot, and converts that to the PWM OUT to feed the ESC in RC-PWM Format.
Q1: What is the resolution you used to get a nice smooth transition?
Q2: Did you need to filter the ADC input from the POT to keep noise out of the signal? (either analog filtering, or digital averaging)

[sall]

Why not just use a standard ESC setup with hall effect twist throttle and use the OEM pedal for the brake? That is what I did on our quad.

[chiflyer]

I know this is really old, but did the code written to the mircocontroller ever get released?

[PacificPilot]

Old topic I know. But I was wondering if anyone has tried to replicated that circuit board with an Arduino. Price of brushless motors have gone down, I would like to try them in my kid's Mustang PW. I have the steel first gears from MLToys. I was thinking of using the shifter to switch the pedal input, so I won't have to use a second pedal for reverse. Brake will be handled by the ESC. I wonder if it'd work. Hmmm... :| :| :|

[Ammonite]

I'm not much into RC, but this is amazing!