5i25 - Using Encoder on Axis

if that is the case , then you have either a wiring problem , or a mechanical binding issue , or the stepper motor or driver is just not strong enough for it
or your trying too hard to get fast movements , that are out of the reach of the steppers you have .
i know it's difficult from our perspective as we don't have the machine in front of us .

you'll need at least a 15 - 20A DC SUPPLY , at close to the highest voltage the gekos are spec'd too , will help , higher voltage = speed & less current required for the same result
z axis always loose steps in the downward direction due to weight and gravity , this shows it's self by either a missmatched number of steps , which you will know if you move say 2" and the axis moves less or more etc .
or due to velocity settings being far too high , and therefore loosing rapidly the holding torq , which you need for the z , or the most common using microstepping above say half step .
it's just not required on a mill or lathe , people wrongly assume microstepping to be for resolution , it's not , it's for clearing up any resonance issues , but with the advent of the new stepperdrives that can handle the current requirements
far better than their old counterparts , they get used for resolution . but people dont realise the holding torq , drops extremely rapidly over half step in a non linear fashion the higher you go .

hence the request to slow it down and use a lower microstep value , , but dont forget this changes the scale value also , so it's no good just flicking a few switches .

hope this helps you pin it down .

Unfortunately the problem is not dependent on load (as i completely removed the motor from the machine and had the same results).
Tried painfully slow acceleration with no success.
Check and re-checked wires, ran new wires just to be sure.
Check for voltage drop on power supply, its rock steady. And its a linear supply so i assume if i was pulling too much current the voltage should drop.
Tried switching the drivers around, no luck.
Seem like the only thing i didn't try was a different stepper motor but how likely is that to be the cause?

For now i have it working fine. I messed around with the hal enough to get the loop closed and also used the 10:1 box so i could use single step and have more encoder lines per step. So far its working silky smooth to my suprise, i assumed there would be some drawbacks but it just works. No herky jerky and ferror is staying under .001 no matter what. I know i just masked the problem but i'm ready to use this machine, and this setup is working so i'm gonna run with it.

If anyone has any other thoughts on what might be the culprit i'd like to hear them! Thanks!

jds,

Glad you got it working. Would you mind posting your HAL configuration for others who might want to do the same thing?

I am wondering if you had noise on your direction input to the drive. Is the direction high or low when the motor is lowering the z-axis? If it is low then noise on the direction signal could cause the axis to take a step upwards while it should have been going down. Just another thought.

We will probably never know now that you have it working.

Also using the 10:1 should work out well with the encoder. Now the encoder is evenly divided by your steps since you are single stepping.

Have fun with it.

Jim

NitroCH3NO2 wrote:

jds,

Glad you got it working. Would you mind posting your HAL configuration for others who might want to do the same thing?

I am wondering if you had noise on your direction input to the drive. Is the direction high or low when the motor is lowering the z-axis? If it is low then noise on the direction signal could cause the axis to take a step upwards while it should have been going down. Just another thought.

We will probably never know now that you have it working.

Also using the 10:1 should work out well with the encoder. Now the encoder is evenly divided by your steps since you are single stepping.

Have fun with it.

Jim

Here is my Hal config. The z section is a little messy but it works lol.