Pulse width too long?

This forum is dedicated to those Hobbyists Interested in CNC machining & 3D Printing in their home shops. (Digital Read Outs are also topical, as is CAD/CAM as it relates to CNC)

Moderator: Harold_V

Post Reply
hobgobbln
Posts: 266
Joined: Tue Apr 13, 2010 7:31 pm
Location: Palmer, Ma

Pulse width too long?

Post by hobgobbln »

I had a problem setting up my lathe with the pulse width being to short to move the motors properly. Setting the width longer helped.

Can the pulse width be set to long? I have another machine that is experiencing the same problem of loosing steps. Figured I'd ask before I started messing with it. I'm using Mach3 and can't seem to find any info.

Griz
User avatar
Dave_C
Posts: 960
Joined: Mon Apr 21, 2008 10:34 am
Location: Springfield. MO.

Re: Pulse width to long?

Post by Dave_C »

Griz,

This is a tricky area for sure. I have no idea what you setup looks like or how big your steppers are as for as Oz In. Making the step pulse longer, or as one might say "wider", you may be overcoming a hidden problem with friction stick or too much mass being moved with too small a stepper. One way to know is see if it moves smoothly after it breaks loose from a stand still. Will it move smoothly at a low speed of say 1 IPM?

The wider pulse, or longer pulse, will come into play as there can only be so many long pulses per second sent. That's just seconds divided by the pulse in length for max pulses per second.

Mach 3 has some reading on this subject. I read it, don't remember exactly where it is but I think it is in the Mach 3 manual. Might have been on some forum.

Dave C.
I learn something new every day! Problem is I forget two.
hobgobbln
Posts: 266
Joined: Tue Apr 13, 2010 7:31 pm
Location: Palmer, Ma

Re: Pulse width to long?

Post by hobgobbln »

The other machine is a small benchtop mill (X2). 270 oz 48v steppers and a G540 driver. No problems with low speed friction.

I haven't had time to make and install the index pulse disk on my new lathe yet. I have been using the mill with a mandrel for turning threads for the last few months.

The issue is an accumulating error with the mill. After every 4 or 5 pieces I have to adjust the dro. Had the same problem with the lathe until I adjusted the pulse length (width). After I did that, rock solid. I'm mainly worried about hurting the controller or motors. The G540 specs a pulse with of 2 ms but I had to bring the lathe to 8 ms to stop the "drift".

Mainly wondering if to long of a pulse will burn out the motors or controller. I don't know what the internal circuitry of a stepper consists of or how it actually works.

Griz
User avatar
Rick
Posts: 481
Joined: Sat Jan 04, 2003 8:34 pm
Location: Stone Mountain, Ga.

Re: Pulse width to long?

Post by Rick »

Griz
I assume you are talking about the step and direction pulses from the controller to the drives? If so you need to look at the drive spec (which you did) and it should specify min step pulse duration the drives needs. If the pulse is to short the drive may not see it, therefore it will not respond with a step of the motor. Another thing that causes steppers to loose steps are harmonics of the system, these show up especially during acceleration and deceleration or coordinated motion moving on a curved path . To over come this you need to run the accel/decel as fast as you can with out loosing steps and then back off some. Getting this set is a tedious trial and error process. but this helps the system pass thru the systems harmonics area as fast as possible therefore limiting the effect or doing away with the effect all together. If your accel and decel is to fast you can loose steps also as the drive system can't respond as quick as the controller is trying to make it. if your accel and decel are to slow then you can have harmonic issues as the system is at this critical(harmonic) speed for a longer time. Are your motors and drives combination setup correctly(steps per revolution) / power settings etc.

To answer your question "can the pulse be set to long" the answer is no BUT you can limit the pulse rate /sec by being to long. The drives typically are looking the the rising or falling edge(depends on the drive, it will be one or the other but not both,voltage change 0v to 5v for rising, 5v to 0v for falling) of the pulse signal. The pulse duration should be just long enough for the drives processor to see it and react. So every time the drive sees a potential change in the signal it steps the motor, the duration of the pulse theoretically has no effect.

Hope this helps
Rick

“We make a living by what we get, but we make a life by what we give." Sir Winston Leonard Spencer Churchill (1874-1965)
"Peace is that brief glorious moment in history when everybody stands around reloading". Unknown
Murphy's Law: " If it can go wrong it will"
O-Tool's Corollary: "Murphy was entirely too optimistic"
hobgobbln
Posts: 266
Joined: Tue Apr 13, 2010 7:31 pm
Location: Palmer, Ma

Re: Pulse width to long?

Post by hobgobbln »

The change in logic state makes sense. I wasn't sure if the pulse width was how long the coils were being energized. That was why I was worried it could fry them. Like exceeding the "duty cycle".

Thanks for the info both of you,
Griz
Post Reply