RC Box - combining PowerFunctions with RC equipment

[MisterCreators Custom MOCs]

Hello technic community,

I want to share some of my thoughts and my progress designing my RC Box.

I'm recently working on something very similar to @NoEXIST 

The concept is an RC box with (up to) 16 channels. The receiver, associated telemetry and the power functions outputs are integrated in the box, while a battery must be connected externally in the form of a 2-3S lipo. As a proof of concept, I am currently working on a 10 channel version, which has 2 full size ports for servos and motors and 8 dual channel ports for motors only (Size: 4x9x5 Studs). The Dual Channel configuration allows to control two motors separately over one PowerFunctions Connector. Goal of this project is being used for complex models with many motors and functions (therefore up to 16 channels) while using an RC-grade remote (long range, fast response time) and a strong lipo to get longer running times. 

My current high-end model (Pistenbully 600W Polar) has 14 Motors and requires 3 Buwizz 2.0 Units to work. Even with them, the runtime in snow is only about 10 minutes. When using an 3S Lipo I would be able to get even better performance out of the motors as well as a longer runtime (if the lipo capacity is bigger).

This is the first Dual Channel test i've done:

 

After two months of work, I now can proudly present my first handmade prototype of my 10 channel RC Box. There is very much work to be done, I have to change my connectors for better contact and for tighter fit. Also, I have to look for another way to address lego servos for proportional control since they don't use PWM. 

But have a look by your self:

 

Feedback and thoughts are appreciated :)

Best regards
Adrian

[oracid]

Ok, now we have two achievements pursuing the same objective, using an RC radio control. Emulation is a good thing.
Obviously, the more motor and servo connections the box has, the more efficient it will seem.
The problem is that the more actuators there are, the more the need to coordinate all these actuators with a program will be felt.

[NoEXIST]

Heavy duty gear! That's a nice idea to operate 2 channels through 1 port. I'm preparing to show the inside of mine box soon, looking forward to see yours too if it's possible:)

 

I've seen the prices for remote... How much would it be for the box?

[BrickTronic]

18 hours ago, MisterCreators Custom MOCs said:

... The Dual Channel configuration allows to control two motors separately over one PowerFunctions Connector. .....

Hello,

I assume, your PF-Adapter wires GND from Bottom to C1 on Top and 9V (BAT+) from Bottom to C2 omn Top, right ?

I've borrowed from Philo's Page :

So your RC-Box put PWM (lets call them C3 & C4) to the PF-Connector instead GND & 9V.
Does you conform ?

I would be interested, why your Box is so high ?
What is inside ?

Jo

[MisterCreators Custom MOCs]

1 hour ago, NoEXIST said:

I've seen the prices for remote...

Yes the remote is quite expensive, but there will be ways to use other remotes. For now you also could get an opentx remote and an AFHDS3 module and it should work fine together-

 

1 hour ago, NoEXIST said:

How much would it be for the box?

Not quite sure about that but probably somewhere around 200-300€ depending on the number of channels (no guarantee for that at this time XD)

 

27 minutes ago, BrickTronic said:

So your RC-Box put PWM (lets call them C3 & C4) to the PF-Connector instead GND & 9V.
Does you conform ?

Can confirm

29 minutes ago, BrickTronic said:

I would be interested, why your Box is so high ?
What is inside ?

it's so high because i needed space for all the components. At first there is the receiver, then there are 10 motor controllers that output the different PWM Signals, then there is telemetry to transmit the battery voltage to the remote and there are wires in there. And they are not small. A peak current of 50A must be able to flow and need to be turned of and on. And 50A cannot be switched via the small switch, so a corresponding transistor is installed which is controlled via the switch.

I still hope that i can reduce the height 1 Stud, but I'm not sure about that at the moment. Also, I'm not ready to present the internals, since there are wires everywhere and it does not look good XD

[gyenesvi]

Although I am more into the simpler RC car use case for now, it is interesting to see this other quite different use case being tackled by someone!

The 10 motor controllers probably take up much of the space inside. Are you using ESCs or simpler motor controllers? If so, are they dual ones, like in the Lego IR receiver, or individual ones? That could save some space.

Interesting trick for for doubling the capacity of PF connectors. For the ones that are in the middle and not doubled, are you planning to wire it just as regular PF connectors are on a Lego PF receiver (as it is up in the quoted image from Philo)? Then it will be able to control servos as well.

[MisterCreators Custom MOCs]

1 hour ago, gyenesvi said:

Are you using ESCs or simpler motor controllers? If so, are they dual ones, like in the Lego IR receiver, or individual ones? That could save some space.

These are ESCs with PWM output, and they are dual ones.

1 hour ago, gyenesvi said:

Interesting trick for for doubling the capacity of PF connectors.

Thank you very much :)

1 hour ago, gyenesvi said:

For the ones that are in the middle and not doubled, are you planning to wire it just as regular PF connectors are on a Lego PF receiver (as it is up in the quoted image from Philo)? Then it will be able to control servos as well.

The ones in the middle are wired up normally for Servo support, but while testing I discovered that LEGO servos don't take a PWM signal, so there is no promotional control yet. You can only steer fully or not, but I'm already trying to find a solution for that.

[gyenesvi]

2 hours ago, MisterCreators Custom MOCs said:

These are ESCs with PWM output, and they are dual ones.

I guess those are relatively large, simpler (dual) motor controllers could save you space, at the cost of not being able to take more than a few amps, but that's enough for PF motors. Have you considered those?

2 hours ago, MisterCreators Custom MOCs said:

The ones in the middle are wired up normally for Servo support, but while testing I discovered that LEGO servos don't take a PWM signal, so there is no promotional control yet. You can only steer fully or not, but I'm already trying to find a solution for that.

As far as I understand, the Lego servo needs the same signal from the motor controller as other PF motors (on wires C1 and C2), plus it needs GND and 9V (or whatever voltage) directly from the battery on the other two wires. Did you wire it up that way?

 

[MisterCreators Custom MOCs]

5 hours ago, gyenesvi said:

 Did you wire it up that way?

Yes I did, but the servos don't understand the PWM signal the ESCs put out. So for now servos can not be controlled proportional, but they can be controlled left or right (-90°, 0°, 90° only these 3 positions and nothing in between)

[NoEXIST]

12 hours ago, MisterCreators Custom MOCs said:

At first there is the receiver, then there are 10 motor controllers that output the different PWM Signals, then there is telemetry to transmit the battery voltage to the remote and there are wires in there. And they are not small. A peak current of 50A must be able to flow and need to be turned of and on.

Yeah, know that, which cables are you using for power circuit? I went the most practical way and used same wires as ESC were produced with

[MisterCreators Custom MOCs]

1 minute ago, NoEXIST said:

Yeah, know that, which cables are you using for power circuit? I went the most practical way and used same wires as ESC were produced with

I went with the same ribbon cable that is used in the motors and extensions.

[NoEXIST]

6 hours ago, MisterCreators Custom MOCs said:

I went with the same ribbon cable that is used in the motors and extensions.

Just PF ones?

[BrickTronic]

11 hours ago, MisterCreators Custom MOCs said:

Yes I did, but the servos don't understand the PWM signal the ESCs put out. So for now servos can not be controlled proportional, but they can be controlled left or right (-90°, 0°, 90° only these 3 positions and nothing in between)

Hello,

Your Statement is valid for many cheap chineese Servo-Clones

Beside "Center" the Lego Servo support 7 Positions to the Left and 7 Positions to the Right.
In total 15 Positions.

Unfortunately Lego uses not the "usual" Servo-Control (20ms Periode and 1,0 to 2,0ms High.
Center is 1,5ms and +/-0,5ms for Left/Right (depend on Servo if this result in +/- 270° angle or only +/- 90°)

Lego drive C1 = 0 and C2 0 to 100% PWM
or C2 = 0 and C1 0 to 100% PWM

In an Youtube-Video (time-Stamp 0:52) you can see an opened Lego-Servo and the 15 Positions for Position-Control/Regulation

Jo
 

[MisterCreators Custom MOCs]

5 hours ago, NoEXIST said:

Just PF ones?

Yes, same cable PF is using, 

47 minutes ago, BrickTronic said:

Unfortunately Lego uses not the "usual" Servo-Control (20ms Periode and 1,0 to 2,0ms High.
Center is 1,5ms and +/-0,5ms for Left/Right (depend on Servo if this result in +/- 270° angle or only +/- 90°)

Well thank you very much for that, then I have to take a closer look into my servo control 

[gyenesvi]

5 hours ago, MisterCreators Custom MOCs said:

Well thank you very much for that, then I have to take a closer look into my servo control 

Have you been testing with an original Lego PF servo, or a clone? Maybe that's the problem?

I think I have seen some video that simply used an ESC to control the servo as well (with the two extra wires connected properly).

In some way the signal on C1 and C2 has to be the same as for controlling a regular motor because both has to work from the same PF plug.

[NoEXIST]

14 hours ago, gyenesvi said:

In some way the signal on C1 and C2 has to be the same as for controlling a regular motor because both has to work from the same PF plug.

Not sure, but C1 and C2 are just to power electronics used in servos and PF receivers. As I know they don't work exactly the same as regular motor contacts do

[BrickTronic]

2 hours ago, NoEXIST said:

Not sure, but C1 and C2 are just to power electronics used in servos and PF receivers. As I know they don't work exactly the same as regular motor contacts do

Hello,

PF-Receiver and Lego-Servo are powered by GND/0V and 9V Pins of the Power-Function connector.

See here again Picture from Philo :

Signals on C1 and C2 at Servo define Direction and Angle
A simple H-Bridge-Driver (that is inside the Power-Function IR Remote Receiver) that usually control an DC-Motor can generate this Sinals for the Servo

Jo

[gyenesvi]

7 hours ago, NoEXIST said:

Not sure, but C1 and C2 are just to power electronics used in servos and PF receivers. As I know they don't work exactly the same as regular motor contacts do

3 hours ago, BrickTronic said:

PF-Receiver and Lego-Servo are powered by GND/0V and 9V Pins of the Power-Function connector.

This. As far as I understand, it's exactly the C1 and C2 signals that are shared by regular DC motors and the servo. When you drive a regular DC motor using an ESC, you need to connect the ESC output to C1 and C2, and can leave 0V/9V disconnected. That's what @MisterCreators Custom MOCs did here, that's how he can multiplex two signals into one wire. For the servo, you need to connect 0V/9V as well. The difference is I think, that the DC motors use C1 and C2 for driving the motor directly, and the speed/power will become proportional to the PWM fill rate. But for the servo, C1 and C2 are not enough, it also needs 0V/9V which is used for driving the motor, as it needs full power to move, and C1 and C2 only encode for left/right angles separately (only one of them can be nonzero). 0% PWM signal on both means center position, and 100% PWM signal on C1 or C2 is max left/right. It cannot use the C1 and C2 for driving the motor itself, because then for example when it is turned completely to one side, and then you give it a signal of 0 on both C1 and C2, there would be no power to move back to the center.

4 hours ago, BrickTronic said:

A simple H-Bridge-Driver (that is inside the Power-Function IR Remote Receiver) that usually control an DC-Motor can generate this Sinals for the Servo

And as far as I understand, the output of an H-Bridge driver has roughly the same encoding as that of an ESC, it's just that ESCs are more sophisticated in terms of precise speed control (use back EMF and more sophisticated algorithms), and can also handle more amps / more powerful motors.

What I don't know though is whether the H-Bridge can be driven by the same input signal from the receiver as an ESC. I wonder if it would be possible to modify a Lego receiver by removing the IR receiver and putting a small 4CH RC receiver into it instead, and wiring it up to the H-Bridge controller. Anyone?

[NoEXIST]

14 hours ago, gyenesvi said:

This. As far as I understand, it's exactly the C1 and C2 signals that are shared by regular DC motors and the servo. When you drive a regular DC motor using an ESC, you need to connect the ESC output to C1 and C2, and can leave 0V/9V disconnected

Oh, now I see, I swapped contacts while writing post. I meant gnd and 9v

[MisterCreators Custom MOCs]

19 hours ago, gyenesvi said:

, it also needs 0V/9V which is used for driving the motor, as it needs full power to move, and C1 and C2 only encode for left/right angles separately (only one of them can be nonzero). 0% PWM signal on both means center position, and 100% PWM signal on C1 or C2 is max left/right. It cannot use the C1 and C2 for driving the motor itself, because then for example when it is turned completely to one side, and then you give it a signal of 0 on both C1 and C2, there would be no power to move back to the center.

My servo outpouts are fully wired up, so they do have 0V, 9V, C1 and C2 thats why they are not labeled as a dual channel. But even with that, I had no luck getting the servo to turn proportionally, no matter how less PWM I applied it always went 100% left or right, that's why I thought that they might need proportional voltage input and not PWM input.

So I was able to turn the servo and the connections were ok, but the servo was not able to decode the signal correct for proportional turning.

I will try some more things this weekend

Edited April 5 by MisterCreators Custom MOCs

[gyenesvi]

5 minutes ago, MisterCreators Custom MOCs said:

So I was able to turn the servo and the connections were ok, but the servo was not able to decode the signal correct for proportional turning.

But is it an original Lego servo or is it a 3rd party one? Only original Lego ones are capable of proportional turning, none of the 3rd parties.

[MisterCreators Custom MOCs]

4 minutes ago, gyenesvi said:

But is it an original Lego servo or is it a 3rd party one? Only original Lego ones are capable of proportional turning, none of the 3rd parties.

It was a original LEGO Servo which was tested on Buwizz to work properly. 

[gyenesvi]

1 minute ago, MisterCreators Custom MOCs said:

It was a original LEGO Servo which was tested on Buwizz to work properly. 

Okay, I see. That's weird then..

[MisterCreators Custom MOCs]

I now tested everything again and built a small testing rig, to make sure that it's nothing wrong with my RC Box. But same as before, the servo still only turns all the way to the left or to the right, no proportional steering possible. I used an 2S lipo to make sure that I'm not overvolting my servo.

 

[MisterCreators Custom MOCs]

I've now hooked up my multimeter to the Buwizz ports and as I already expected, the Buwizz does not output an PWM Signal, it outputs proportional voltage.

That means: to get the servo turning just a little bit you need 1,6V from C1 to C2. The more voltage there is between C1 and C2 the more the servo will turn. As the PWM Signal always provides full voltage, the servo will always turn 100%.

I hope this is understandable for everyone and clarifies any open questions.