Opinions about motors with integrated encoders

There’s a certain amount of mis-information in this thread.

Here are some facts.

  1. You do need to be careful when installing the IMEs. Make sure the PCB does not have any long pins or solder tails that may scratch the encoder wheel, snip these off if they are really bad. Do not over tighten the screws that hold the IME to the motor.

  2. The IMEs should not affect motor operation in any way, that is, they should not cause overheating if installed correctly.

  3. Static can affect the IMEs, prepare for this to happen in the code. Any relative encoder (ie. the red quad encoder as well as the IME) can cause problems on an arm system if used without limit switches or other methods to stop the arm traveling too far.

  4. ROBOTC and ConVEX handle the IMEs much better than EasyC (IMHO), faster update rates and reconnection if communication is lost under certain circumstances.

  5. IMEs can, under extreme conditions, cause a cortex crash, however, the watchdog timer was enabled in ROBOTC to handle this situation.

  6. I find IMEs on a drive suffer more from static if mecanum wheels are used.

I own 12 IMEs and have never had any significant problems that could not be handled by software. We also used IMEs on the three 8888 team robots this year without problems.

The cons of quad encoders also include the usage of digital ports. Each single quad encoder needs two digital ports while IMEs require none. Well, it depends on how you plan your sensors.

Well, on the other hand, all the IME’s you use are chained together, so if the one that is directly plugged into the Cortex gets messed up, you got a problem.

-Nick

You don’t need to use both wires. You can just plug in one and config it as an encoder 1 wire. Going by the technical description it shouldn’t matter which one you use.

However, from an initial quick test, the encoder value this way is half what it is with both wires, so it seems to have less accuracy.

However, before you go and decide to do this, I would recommend testing more and verifying that it works. This is only based on the technical description of them, and about 30 seconds worth of testing.

I’m fairly sure that without both wires, you don’t get the spin direction of the axle, just ticks without direction. It’s fine as long as you code in a variable that remembers which way the motor was last told to spin, and use that as the direction.

Yes… Static does kill

Thanks for the tip. I have worked with one wire shaft encoders and you do loose spin direction which is not a big deal because you can always make up for it in code.

HAHA yeah after this year I am just going to switch back to quad encoders simply because the IMEs are not as reliable as the older encoders.

Same exact experience here.

During the season, especially in the California dry winter months, it was pretty bad. That included CA state championships.

At Worlds…no issues with the encoders.

How do the red optical encoders work ? Do they just have a hole you put a shaft into ?

Yes. They have a square hole. For the Vex Cortex, they provide 360 pulses per revolution. The old Pic documentation indicates 90 pulses/rev.

http://www.vexrobotics.com/wiki/index.php/Optical_Shaft_Encoder

http://content.vexrobotics.com/docs/instructions/276-2156-instr-0312.pdf

That was for the original encoder (which was not two channel quadrature). The pic could only interrupt on one edge, hence 90 pulse/rev. The cortex can get interrupts on both edges (so 180 pulse/rev for a single channel, 360 for quadrature operation).

Are the Pics even legal in competitions anymore? I often get confused when looking at the sensor documentation linked to by the Vex sensor product pages because much of it describes the Pic processor. I’m surprised Vex has not yet updated those documents.

I think the reason people had less trouble at World’s is the attention they paid to static. I would get into the arena early and they were walking around with spray and an ESD meter testing all of the arenas.

We never had the problem much in Florida, but we are humid year round.

Was wondering if we would be allowed to drag a chain from the frame to knock down the differential?

The fields at worlds are treated with an anti static spray to reduce static discharge. We have found that the robots can build up quite a bit of static from the foam tiles, especially when the air is dry.

<R9> Robots must use ONLY one (1) VEX EDR Microcontroller.
a. Examples of VEX EDR Microcontrollers are the VEX v.5 PIC Microcontroller and the VEX
Cortex Microcontroller.
b. Microcontrollers that are part of other VEX product lines such as VEXpro, VEX RCR, VEX IQ, or
VEX Robotics by Hexbug are not allowed.

You can still buy the PIC.

Yes, that is correct that the quad encoders with one wire cannot tell direction. However, I don’t think that’s such a big problem, at least on the drive, as in autonomous you (hopefully) know in which direction you are going. The only use I see is for some sort of an adaptive brake/trim system.