Yaskawa Servo programming

I’ve got 3 Yaskawa sigma 7 1.5kw drives and motors to program for the drilling machine. One has a 8:1 reducer driving a rack gear for the X axis. The other two drive ball screws with 5mm pitch for the two Z axis. There is no Y axis. Each Z axis has 7 - 2.2kw spindles and a single 7.5k spindle.

I am currently working only with the servos using SigmaWin 7+ to program and jog them. I used the setup wizard to set up the brakes, encoder, etc. I need to use the function to determine the correct moment of inertia ratio and then do the tuning process. This is my first servo installation but I’ve programmed all kinds of VFDs and PLCs. I can’t get the MOI ratio rountine to work. It seems to work, runs the servo back and forth multiple times then fails saying it can’t calculate the MOI ratio.

The basic problem is that never having done this before I can’t make good guesses for initial numbers to start with. I tried the X axis with only the motor and reducer and it fails there too.

Sugestions on what to do? Trying to decide if I need to bring in a technician and how long it will take to get one here at what cost.

Are you using the drive in tuning-less or auto tune mode? I built a test machine with 2 similarly sized sigma 7 drives and motors and had such smooth motion in tuningless mode, didn't go deeper than that. Was your the motion acceptable in tuning-less mode? What are you using to generate trajectories? Higher level motion controller or does each axis have a trajectory defined in the drive itself?

We used 60:1 on one axis with widely varying output inertia, though the motor inertia dominated that. The other axis was 5:1 driving rotary to linear timing belt setup , so pretty close to your x axis. I used the resonance detection feature which confirmed both axes inertia dominated by the motor.

I also defined the drivetrain one level up in an automation controller. The motor drive had very little insight into the drivetrain, so made very few changes to the standard axis settings. The motor data is stored (or referenced) in the motor encoder and is passed to the drive when you power it up, so it should already know all motor characteristics, including inertia.

Yaskawa tech support was very competent over the phone, but they seem to want to defer you to the reseller. It sounds like you've dug in enough to have a clear question for yaskawa.

Mike

Mike,
Thanks for replying. I haven’t tried driving the servo with our motion controller yet because this machine was built in China with all kinds of mechanical issues that forced us to tear it down and redesign it. So I have been trying to tune the servos so I can jog position for the purpose of doing precision alignment of all the parts as we put it back together.

Yaskawa has been very helpful but I unintentionally burned that bridge. Over 2 hours of tech support time over about 30 calls. In wiring up this machine we accidently hooked the braking resistor to the motor reactor terminals. That caused an under voltage error no one could figure out. I have 50 years experience with wiring industrial controls so it is embarassing to have this happen. I had a completely inexperienced young man helping me and he did a great job, I found no other wrong connections and there are hundreds in this machine. It may been me that connected then wrong too.

Not having done a servo before I thought it was important to properly tune the drive first. However your suggestion is likely a good one because virtually all the parameters in the drives are default indicating the machine builder didn’t tune the drives either.

We didn’t purchase their CNC. The first one we got from them was a Chinese PC, running a Chinese version of Windows XP! The motion controller was hobbyist grade and we got no supporting documentation, not even a wiring schematic.

Thanks for your help.

The default tuning-less mode and sigma win should let you jog the machine to do any alignment. Without the help of a person experienced in servo tuning in house it will carry you pretty far for motion too.

If things get very out of whack, save a copy of your parameters then reset everything to default. The motor and drive will sort out a safe starting point, and you'll probably need to reset the limit switch configuration.

Good luck
Mike

I’ve found that the Pn103 MOI setting in Sigma5 drives will only cause settling issues if it’s too low. Starting around 100 to 150 will give you a reasonable result. If it’s too high, it shouldn’t matter too much.

Mike,
I tried your suggestion to use tuning-less mode and use the PLC pulse and direction output to control the motion. I keep getting a 710 error, instantanious overload. I have the PLC trying to jog an unloaded motor with a pulse rate that ramps up slowly over several seconds from 0 to 4118 pulses per second. I verified the pulse rate using a multimeter with Hz input. The drive parameters say the pulses per revolution are 2048 so this should drive the motor at 120 rpm so the gear reducer shaft should turn 15 rpm.

Seems simple enough, what am I missing? I keep reading the drive manual over and over looking for the answer, checking and rechecking all wiring and I’ve got nothing.

Are you able to jog the motor successfully using sigma win, independent of your higher level control? I would get this sorted first as this would expose any issues with the drive current limits and with mechanical issues. This is the most fundamental way to prove out motion with your drive, motor and mechanism. If it doesn't run this way, there's no point in moving higher level until you sort it. The manual is pretty thorough walking through this procedure.

I'm assuming you have the SGD7S Analog drive for analog input or step and direction. is that correct? and is this the manual you are working from?
https://www.yaskawa.com/delegate/ge...ocuments&documentName=sieps80000126o_17_1.pdf

I remember the count per revolution feedback being confusing. The motors have a very high resolution encoder and the way the drive divides the counts down to provide feedback to the higher level device wasn't straightforward to me, but I was using an ethercat drive so a fairly different interface. If you can jog the motor with sigma win and the control mode is correct, I'd dig into steps per motor rev next. start out with single step count inputs and see how the motor responds. maybe the step size is big to the point where the motor reaches current limit trying to accelerate the load on the trajectory commanded.

That's about the limit of what I know about these. May also want to look if the firmware is the right version and try the factory default reset and even verify the hardware you have is genuine.

Mike