Are these normal behaviors?

Hi All

I have finally got it calibrated and am able to start printing stuffs. During the calibration process i have notice couple things that seem strange to me. I just wanted to point it out and see if this is normal behaviors or should I be start ordering spare parts because this indicate some components is about to go out.

  1. Left motor and right motor make different noise when its running. The right motor has a higher pitch noise and sounds significantly different then the left motor.

  2. When i adjust the Z motor, both of the Left and Right motor make noise. The both motors don’t move, maybe its just reading the sensors, but whats the point?

  3. After upgrading from firmware 0.98 to 0.99. I have notices pulsing when the left motor turns Clockwise. It only happens to the left motor and only clockwise. I downgraded back to 0.98 and the motors do not pules in any directions.

Thanks
Johnny

3 Likes

Very astute observation!

I believe that the different noises come from the fact that internally the way the two motors are controlled is different due to a pin which was shared with a timer library we are using. Basically the frequency of the PWM signals sent to the motors is different which gives them a different noise.

I don’t think having the left and right motors power up when moving the z-axis is a sign that anything is wrong, but maybe we should change that. You are right that there really isn’t any point to powering them on if only the z-axis is moving.

The pulsing of the left motor is more concerning. I don’t think you have a hardware issue, but it is possible that there is some strange behavior in the software in v0.99. We did make some significant changes to the way the motor speed is controlled which could be causing the effects you are seeing. Does it happen with the sled attached or without?

2 Likes

I have finally got it calibrated and am able to start printing stuffs. During the calibration process i have notice couple things that seem strange to me. I just wanted to point it out and see if this is normal behaviors or should I be start ordering spare parts because this indicate some components is about to go out.

  1. Left motor and right motor make different noise when its running. The right motor has a higher pitch noise and sounds significantly different then the left motor.

Yes, this is normal. They are driven by different PWM signals operating at
different frequencies.

  1. When i adjust the Z motor, both of the Left and Right motor make noise. The both motors don’t move, maybe its just reading the sensors, but whats the point?

the motors are moving, just not enough for you to notice. It’s possible to
change the software so that only some motors are enabled, but it’s far easier to
enable/disable them all at once.

  1. After upgrading from firmware 0.98 to 0.99. I have notices pulsing when the
    left motor turns Clockwise. It only happens to the left motor and only
    clockwise. I downgraded back to 0.98 and the motors do not pules in any
    directions.

This is not normal. There was a change in 0.99 that requires that you upgrade
both GC and the firmware at the same time, and if it’s sticking with settings in
your old config file, that could cause problems.

can you save off your old config, and try going to 0.99 and re-do the
calibration?

1 Like

@bar

Is this an updated Pid profile?

Thank you

It pulses with or without the sled attached. I can see the sled ripple when moving to the right direction.

I always move firmware and GC together.

The config meaning the ini file right? when you say re-do the calibration, do i just have to re-enter the motor distance, motor offset and rotation radius for triangular kinematics values? They should be a fixed value.

You shouldn’t have to re-do the calibration or mess with the configuration file. There is a change to the PID settings, but the way it was written the new values will be used automatically unless you go into settings and turn on the setting to enable custom values and enter something else.

Is anyone else seeing pulsing of the left motor with v0.99?

2 Likes

Yes, the changes required significant changes to the Pid profiles

I see smooth movement in both directions when running 0.99 both with the motors under no load and with the sled attached

@Johnny5

I have been in support for many things hardware and software for several years. I find the questions not asked are often more important, to finding the issues. I’m now chasing your workflow. Let me first acknowledge you might have uncovered a bug. Based on the thread it is assumed you are powering down the Arduino / Maslow Shield and the computer between cutting sessions? is this correct?

Are you on a desktop or a laptop?

Are you working in a fixed location?

Thank you

no, I Never power them down.

Desktop

that’s correct.

1 Like

@Johnny5

Thank you

@bar

I will test this as soon as I’m able and report back. I have 2 factors against me. I’m stuck watching the door for an important delivery. Second it’s too windy out so I need to wait for the wind to slow down during the daylight. So it might be tomorrow morning.

Thanks

2 Likes

It would be nice if the schematic (and also the PCB) had written on them what driver is linked to what timer.

This then could eventually give more insight into that as well.

In the back of my mind there is a ‘rule’ that says it is best to have the most important motors driven by the same timer. Preferably the most capable timer to make better use of the clockcycles…

Not sure if this is true, but for some reason this is in my mind…

When this is mentioned in the schematic and on the pcb then tehre is at least an easy way to lookup what goes where. To get that data in your mind to at least know it, or to know where to find it in a few seconds… Something tells me that i want to be able to look this up on the machine itself

2 Likes

That would require that every motor run at the same speed, the timers are what
are used to run the PWM speed control

This info is very vague in my mind so please forgive me for being vague.

I have to read up on the Atmel docs on this.

If one timer is faster with the math then the other then that affects the total speed of the process.
So for the motors that do the most work it should operate better if they are both on the same timer.

This was the case on the Arduino Nano. The ATmega is more powerfull so maybe it’s not an issue there. This is somewhere in the 300 page datasheet… In the timer section… I don’t even know why this still is stuck in my mind.

Unfortunately that’s not how the electronics was laid out. The TimerOne library is used to control the PWM frequency for the left motor because the internal timing register used by that gpio pin runs at a different frequency than the timers used for the other two motors.

1 Like

They have the same processor and clock rate, the Mega2560 has more internal memory and I/O. Depends on how you define powerful. I’m a MIPS moose so I’d call them equally powerful

I know why they are on two separate ports however would it be cleaner to have the Z axis for 3 motor systems on the “odd timer”?

@bar
What are your thoughts on that?

Thank you

1 Like

that would require a new controller board,possibleith the next set they have
manufactured, but I can think of a lot of other things to worry about before
that.

3 Likes

Maybe it’s just a matter of using a different plug and set the configfile accordingly

We have 4 H-bridges
What timer is on what H-bridge?

What is the current situation?
Left motor is on Timer …
Right motor is on Timer …
Z axos is on Timer …
Unused driver is on Timer…

Nothing has to be changed right away, first we need to know what is where so that we can have a look if it is possible and how to get there.