Troubleshooting Guidelines - Start Here!
Getting started on troubleshooting? Need a refresher? Here are guidelines for poking at a system that isn’t doing what you need it to do.
Troubleshooting a motion control system requires a methodical approach to isolate and identify problems. Check each component, understand what the system is attempting to ‘tell you’, try one thing at a time, and attempt to understand repeatable aspects of the problem.
The questions asked in each section are the same exact phrases that Application Engineers ask daily to root out problems in all sorts of systems. From textile rollers to metrology systems, the root ‘motion control’ can be deduced to simpler parts via lines of pertinent questioning.
If the machine was doing what it was supposed to, and now it’s not, something changed. If that were untrue, the system would be behaving exactly the same. It may not be operator error, it could very well be component issues or lightning strikes, but something changed.
Table of Contents
Expand the links below or click through the Table of Contents above to jump to specific sections…
Initial Checks & Questions
Key Point: Can you isolate the issue & make it repeatable?
Connections
Do not handle live voltages. Safely check all electrical connections involved for loose wires, corrosion, or broken connectors. Ensure these connections are seated properly. Power, communications cables, feedback cables, sensor cables, etc.
Example: This Parker PSD was shutting off unexpectedly on a testbench despite a measured, stable, 24v control power being applied. A very light tug-test determined the return on the 24V path was not seated properly, and the power was dropping when the table was bumped. Issue dissolved when the wire was seated properly.
Narrow the Scope
Key Note: Context is incredibly important.
Begin framing your scenario. Lookup errors, question where communication breaks may be occurring, etc
Find LEDs to begin framing the scenario controller or drives that indicate operational status and/or faults.
What are the LEDs telling you based on the manual’s definitions?
Isolate when faults occur. Here’s some hypotheticals to frame your questions. In the following sections are generic checks, notes on software/programming, and how to go about deducing the problem.
Hypothetical 1: Does an Overcurrent occur at time of enable, or disappear when a spare motor is implemented, or only occur when ramping up the system?
Hypothetical 2: Does an “intermittent” Undervoltage actually appear when an adjacent machine turns on?
Hypothetical 3: An Overvoltage occurring at time of Enable without a motor connected may be a bad product… However, an Overvoltage during deceleration could be too much Back-EMF; if it just began occurring in a working system (without any additional load added!) it could be the braking resistor has burnt out instead!
Quick Fixes & Checks
Key Note: Try one thing at a time. Revert changes if required. Be methodical.
More on some of these checks later in this guide, though take note of what you have and what you can do initially. Pay attention to part numbers.
Do you have any exact spares present? Any backups of programs from the controls if needed?
For motors and cabling when swapping exact part numbers, this can be a simple check if the issue is suspected to be a bad motor/cable.
For drives and controls, a backup of the project will be required if swapping. More on this later.
Can you swap exact components if something is suspected to be ‘bad’? From feedback cards, to motors, to cables, sensors?
Does a Bridge Fault disappear if the motor is disconnected?
Does the motor stop running away if exact cabling is changed out?
Do you have a multimeter if pertinent? Are there any components ‘between’ connections to check? Line reactors, breakers, Estops, fuses, etc.
Are any of those components shorted based on multimeter readings, is the line reactor values far too low for the manufacturer’s spec?
Manual Operation
Attempt to isolate the problem outside of normal operation. This could be different ‘modes’ of the machine, or physically moving the system [safely] by hand with the drive disabled depending on nature of the issue.
You can often utilize the software to isolate components by running specific combinations. Such as jogging a motor and drive together without the controller to deduce proper motor functionality, and/or proper controller communications. More on that in the following section.
Present the Information
In troubleshooting endeavors, half the battle is presenting what you see and what you’ve methodically tried, to others. From colleagues, to technical support, to higher ups. Ensuring everyone is on the same page is critical to avoid doubling up on work done, and preventing others from undoing potential fixes if they were not kept in the loop (especially if the problem is multi-part!).
Key Point: Write down the steps you’ve done somewhere so that others are aware. It helps you, and everyone else involved, keep track! Write that down, write that down!
Bullet Points
These are your friend. More complicated issues may need additional details or phone calls, but an initial presentation of information goes a very long way in bringing others up to speed and relaying information.
Here’s an example of information presented by a technician, that was sent to their colleague, that was then transferred to Parker Application Engineers:
The part numbers in the system are Motor: BExyz, Drive: PD-04P, Controller ACR74C-xyz
I’ve attached a couple pictures of the nameplates.
I’m getting an Overvoltage fault AL-41 on the Pdrive. I see it happen when it decelerates during specific moves in operation I think. It runs fine beyond a few points depending on what process to run.
This was a working system before, for a couple years actually and this just began happening.
We’ve replaced the motor because we had a spare (the motor was the exact same part number). We don’t think this would have helped all that much as it seems to run fine during other parts of the cycle, but the night technician tried it.
The original motor ohm’ed out fine, no shorts or anything.
We’ve tried to go online with the drive software [DST]. We see that we can edit a few parameters to adjust, but this was working before?
The manual says potentially to much energy, and the machine has a speed setting, so we turned the knob down, and it looks like it happens less frequently. I doubt we’d need to actually change anything in the program…? What does this mean?
I don’t think the drive is bad it decelerates in other processes?
Any ideas??
Signed,
[Customer]
Follow Up
Even if the above is as far as the technician got, or if the events listed were not wholly beneficial in hindsight, the presentation helps understand what was done. In this case someone completely unfamiliar with the specific system but familiar with generalities of what constitutes ‘Overvoltage’ & product knowledge would be able to understand what was done; and potential next steps.
Followup questions an Apps Engineer asked:
Does the system have an external braking resistor? The Pdrives have one built in, but is one external?
Was additional load added to the system at any point recently, requiring more energy to slow it down? Did anyone speed the machine up or tweak the parameters under the hood for quicker decelerations?
You’re correct that we should avoid editing anything in the code/parameters under the hood if the machine was working prior. Though we may get there in the ACR controller for testing purposes only, just not currently.
The fact you turn the speed down through the machine’s own user interface and it seems less frequent helps point to the drive getting too much energy, sure. Do you have any details to how less frequent? The type of motion/moves being done? Only if the machine runs a certain number of cycles now before it occurs, or now only occurs with the heavier materials (if the machine runs different materials, or other processes run faster / slower)?
The technician came back:
We were monitoring the incoming mains voltage with the Oscope that came by, looked okay on the AC mains through most of operation. No additional load was added, no one sped it up either (shouldn’t have at least…). It does have a braking resistor, we missed that as it’s tucked up in the corner… turns out that’s degraded and basically almost shorted. Far smaller R value read with a multimeter, than what it’s spec’d to be. Braking resistor replaced, issue dissolved. Thanks!