Help Understanding “following errors”

[Cellar Dweller]

Hi all,

I am working through my Torq-Cut 22 DOM replacement.


I am getting a “following error” when the machine attempts to home. The drives engage, you can hear the mag switch pull in for the spindle drive.

There is no movement at all. The error is immediate.

BTW, do any of you very experienced guys know whether or not this error may occur on new software installs were some parameters are generic. For example, I see parms for a C axis, but I don’t have one. Maybe the software is looking for some feedback from a C axis card and alike, but since they don’t exist on this machine, the software throws an error????

I have gotten this error in the past on this same machine. It was during a program. I changed the gcode were it stalled and threw the error and all was well.

If someone could provide some insight as to what this error is actually looking for and so forth it would be a benefit.

Thanks,
CD

[jmullett]

Looking back at your past threads, I am having trouble following what servo amp control coniguration you are working on, but I have a torq-cut 30, which is just like the torq-cut 22 only longer in the X-axis, that was giving me sometimes sporadic, sometimes persistant problems with following errors and over current alarms on one of the axii.

Since I was lucky enough to have 3 identical amps on each of the X,Y and Z axis, I was able to swap the amp of offending axis with another and see the problem follow the bad amp. Once I determined that it was within the amp I bought a new one from BPT Parts. The funny thing is when I took the bad amp out I found a cold/burned solder joint on its circuit board where one of the screw terminals going out to the servo motor is connected. A little work with a soldering iron and it was working just fine. Now I have a spare amp on the shelf to get me out of my next jam.

You might want to check for a similar problem on your amp or at least rule out a physical/hardware issue.

[Cellar Dweller]

When you say “amp” as in amperage, or “servo amp control configuration” are you referring to the drive motor control boards? I am not sure what is the correct or common name for these boards. In my machine the boards that supply power to the (3) axis motors are almost identical. (see pic)

The connections on these boards are ratty, (see pic) so my next step is to go around stripping and tinning these wire ends for a proper connection to the boards. I will also keep a sharp eye out for any cold solder joints.

If anyone can weigh in on the real cause or indented cause of this error, it may be very helpful. One would assume that a “following error” is just that. The anticipated position and the believed actual position don’t match. But, how is the controller trying to perform this error handling? Encoder verse…what? What are the two, or more items or parms being compared, at least at a surface level?

Can you or anyone else with a torq-cut 22 or 30 check the input and output voltage from the power supply feeding the axis drive boards. Again, the voltage from T2 to the input of the power supply and then the voltage realized at the axis board.

Thanks again, for everyone’s input.
CD

[FanFan]

Hello, Not sure how to give a textbook explaintion /defintion of following error. But I give you my take on it. Following Error (F.E.) is a method to display in counts of how an actual axis movement is keeping up with intended ( commanded) actual movement. It is the difference between the two counts.

Say the axis is moving from point A to point B. The computer tells the axis to move. The axis responds and starts to move. As the axis moves there is lag (error) between where the axis actual is and where it is supposed to at any given point in to. An error ( difference) is ok but it is should be an predictable one.

It has been a long long time since I have played around with Following Error. We used it as a tool to tell us how the machine axes movements were doing. We basically knew/expected that a machine would act in a certain way under a certain set condition. If we had the feedrate set at 100% and jog ( in medium) an axis in one direction we would get an F.E. of 100 counts. ( It should not vary more than one count. If we jog it in the other direction then we could expect an F.E. of 100 counts also. Now if we went to the other axis then we should have gotten the same results. If we kept a feedrate of 100% but changed to a high jog then our counts changed. It has been over 10 years since I have done this but I think the anticipated F.E. was about 1550 counts. But in that high jog mode the F.E. count were expected to stay at 1550 counts when jogging in both directions and both axis. ( I think those FE numbers are correct on a high jog. I will try to verify then tommorrow). When I first started in machine repair it was interesting to me that on a VTL both horizontal and vertical axis had the same F.E. and in our shop that pretty much all machines with the same controls had the same F.E. under the same conditions.

Like I mention, above 100 % and at a medium jog......a F.E. of 100 counts was expected. On a well set-up axis we could say the F.E. was rocksteady. That is what we wanted. But, we would tolerated a varying count of 1-2 "- +." If the F.E. gets too large then the control would issue a fault and eventually put the machine in Estop.

If the machine got to point that it went in to estop we would look to see if we were losing our postion feedback or if the we weren't getting power out to the motor. If the the counts were varying too much...say over 10 or 20 counts we would probably look at ways, oiling system, belts. Things at being too tight or too loose. Motors and drives on their way out would also be suspect. Like JMullet asked. Is this during it on just one axis or more than one axis? If the problem is intermittant. ( Probably not a software setting issue) Ask yourself some questions. Does it happen on only one axis? Under one set of condition.? Which condtions? Has something been changed recently. a smah up?

I hope this give you some info. Like I said I haven't played around with F.E. for years if I have made an error or left something out I hope someone let's us know. FanFan

[underthetire]

Fan fan gave a good example, simply put its the difference from actual position to commanded position during movement.

[Al_The_Man]

Most controls have a In-Position Parameter, to set the allowable difference between the commanded and the actual position, this is the allowable error that can occur before a following error is displayed.
Al.

[Cellar Dweller]

Thanks a lot for everyone’s input, very cool, and generous.

Ok, that being understood, how does it do this comparison on a Torq-cut22 of very similar bread of machine?
This, “actual position to commanded position”? I assume (haha) it gets one signal from the encoder (located on the drive motor armature) but where is the second or compared value coming from? I get “command position” that means the “programed” or “jogged” position. But, were is the data coming from? Is it just the “programmed” or “jogged” position in the software compared to the data from the encoder?

Is it comparing the amount it thinks it should move to the encoder? Basically, there is only one device (encoder) being looked at. It does not say, “Compare the encoder to a glass scale” or other electromechanical device?

To answer the question, it happens on all three axis. The machine will not home. The F.E. occurs during the first attempt to home upon start up.
There has been no crash.
The HD went bad, so I installed a new DOM, etc. I am still leaning towards software at this point. The machine was basically fine until the hard drive failed.

Could one increase the following error (in parms) to a ridicules number to see if the axis actually move? As trouble shooting tool or method. If the X and Z start to move and Y does not, you know where to look. What are the risks?? Many I assume!

Thanks,
CD

[Al_The_Man]

If the error is immediate with absolutely no movement, and all three axis, I would think there is something else in software going, especially after a s/w reload.
When you hear the drives power up, do they resist movement?
What I do to detect encoder or feedback problem, if the control displays anticipated position rather than posn. via feedback, some controls will switch between these.
If it is the former, I disable one axis by removing the power into the motor, power up the unit and by H.W. on the lowest setting, move the handle and watch the display, at some point the following error will pop up.
Remember the value and reset the system, this time move the axis H.W. the same way and watch for a distance LESS than the error distance, now turn the motor, B.S. etc an amount that would be equal to the distance on the screen, you can keep on advancing the display and turning the motor continuously without error, if the encoder is good.
If your display shows feedback position instead, then you know the encoder is good if the display changes.
This is mainly for systems that show the anticipated screen position.
In your case I am not sure if this will help?
Al.

[FanFan]

CD,
What do you mean with a DOM replacement? Is that directly related to the hard drive replacement? Do/did you keep a written hard copy of all your parameters, including the changes that you made to them?

When you turn the control and take it out of estop do any of the axis "jump" at all or do you just get following error messages?

The command signal that I am talking about is coming from the computer section. Perhaps it should be called an error signal. It is part of a closed loop system. The computer control tells (commands) the drive ( servo ampflier) to move the motor. The motor moves which in turns should make the encoder ( because it is coupled to the ball screw or to the motor) detect a change in postion. The encoder then send info ( postion) back to the Computer control. The computer control then compares how much the motor was able to move the axis along. It then computes a new signal error signal.... This happens probably hundred times a second. Constant updates are made hopefully getting it smoothly and accurately from point A to point B.

FanFan

[Cellar Dweller]

FanFan,
I do have a BMDCPRMS.SYS from 2003 as a backup. I never made one. And I bought machine in 2009. So I don’t know if the information in that file is accurate. I can view it note pad.

Yes, I installed a DOM or in my case to be more specific, an IDE to CF adapter. Loaded DOS 6.22, etc.

Order of events:
Turn power on. OK
F1 to continue around my check sum and bios battery errors
Type “run” at c: prompt.
Machine software loads.
Get “low air pressure” error. Turn on air pressure and/or ESC
No error, flashing “Drives off”.
Press 7, “engage drives”. Hear door lock engage. Hear spindle power switch engage.
Spindle controller is powered up. RPM = 0 in red display. (inside the control box)
No errors at this time. Nothing has moved. Not that is obvious anyway maybe in the .000 or .00.
Press 8 “Home Mach”, immediately get F.E. Again, no movement was even attempted at any axis.
The F.E. error caused the spindle power to drop out, like an estop command.

Thanks for the continued explanation of what happens during the F.E. That is basically how I understood it work. As of right now I am trusting the error at face value. Meaning, I am looking at why the command to move is not causing the drives to turn.

Can you see more specifically which drive is related to the error, like in maintenance page, etc?

Thanks,
CD

[Al_The_Man]

Did you try to see if the motors show any resistance before trying to home or move as I suggested?
Al.

[Cellar Dweller]

No, Not at shop, tomorrow. sorry. What would you expect to see or I should say feel?