A5A0 on an ABB Drive: What Safe Torque Off Is Telling You

If you walked up to an ABB drive and found A5A0 on the keypad, the drive is reporting that its Safe Torque Off circuit is open, and the first thing worth knowing is that this is usually the safety system working rather than the drive failing. A5A0 and fault code 5091 are the same underlying condition, which the manual describes as the safe torque off function being active because the safety circuit signals connected to the STO terminals were broken during start or run.

Put plainly, something in the safety chain opened, the drive removed torque from the motor the way it is designed to, and it is telling you so. The question is not usually how to make the message go away. The question is what opened, and whether it was supposed to.

What Safe Torque Off actually does

Safe Torque Off is a hardware function that removes the drive’s ability to produce torque in the motor, and it does it by cutting the gate signals to the output devices rather than by asking the firmware to stop nicely. That distinction is the whole point of the function, because a software stop depends on the software behaving, while STO removes the drive’s physical ability to turn the motor whether the firmware cooperates or not.

The circuit runs through two independent channels into the STO terminals, and both channels have to be satisfied before the drive will produce torque. Wire an e-stop, a gate switch, a light curtain or a safety relay into those terminals and you have a machine that cannot start while the guard is open, with no dependence on a PLC program or a parameter being correct.

So when A5A0 shows up, the drive is reporting the state of that circuit. The motor will not turn, and that is the design working.

Why it usually is not a problem

The most common reason a technician sees A5A0 is that the machine is sitting in a safe state and somebody expected the drive to run anyway. An e-stop is latched, a gate is open, a safety relay has not been reset since the last shutdown, or a maintenance switch is still in the position somebody left it in on the previous shift.

The second most common reason is that the drive is new or the panel was recently modified, and the STO jumpers that ship in place were removed during wiring and never replaced on a machine that does not actually use the safety function. If the safety circuit was never wired and the jumpers are gone, the drive sees an open circuit and reports it, because it has no way to know the difference between a missing safety circuit and a safety circuit that opened.

Neither of those is a drive problem, and neither gets fixed by clearing anything on the keypad.

The parameter that decides what you see

ABB drives let you configure how the STO condition gets reported, and parameter 31.22, STO indication run and stop, is where that choice lives. Depending on how it is set, the same open safety circuit can show up as an indication while the drive is stopped, as a fault when somebody tries to start, or both, which is why two machines in the same plant running the same drive can behave differently when the guard opens.

That setting explains most of the confusion around this code. A drive that trips and needs a reset every time an operator opens a gate has been configured to treat the STO condition as a fault, which is a legitimate choice on some machines and an operational nuisance on others. Changing 31.22 changes what the drive tells you and whether it latches, and it does not change the safety function itself, since STO is hardware and the parameter only governs annunciation.

Before anybody changes it, check the machine’s safety documentation, because how a machine annunciates and recovers from a safe stop is usually part of the assessment that was done when it was built.

5090 is the one that should get your attention

There is a second code sitting next to this one, and it matters more than 5091 does. The manual is direct about the difference: if only one STO channel is opened, fault 5090, STO hardware failure, generates instead of 5091.

Read that again in terms of the circuit. Both channels opening together is a safety circuit doing its job, because that is what happens when somebody hits an e-stop. One channel opening while the other stays closed is a discrepancy, and a discrepancy means one of your two independent channels is not behaving, which is the condition the redundancy exists to catch.

The manual sends you to the connection between the relay output of the safety module and the STO terminal, and that is the right place to start. A loose conductor on one channel, a failed contact in a safety relay, a broken wire in a cable that flexes, or a terminal that was never torqued will all produce this. A machine that throws 5090 has lost part of its redundancy, and it will keep running on the remaining channel until that one fails too, which is exactly the failure mode the two channel design was built to prevent.

If you take one thing from this page, take that one. 5091 is the system working. 5090 is the system telling you it has a problem inside itself.

Every major brand splits it the same way

This is worth knowing because it transfers, and because most people who learn it on one platform never realize it applies everywhere.

Allen-Bradley does the same thing on a PowerFlex 525, where F059 Safety Open means both of the safety inputs are not enabled and F111 Safety Hardware means one of the safety inputs is not enabled. Yaskawa does it too, where SToF, Safe Torque OFF Hardware, reports that only one of the two terminals H1-HC and H2-HC received the disable signal.

Three manufacturers, three different sets of codes, one shared idea. Both channels open is normal. One channel open is a fault in the safety circuit itself. Once you see that pattern, the code number stops mattering and you know what to go look at on any drive in the plant.

What to check, in order

Start at the machine rather than at the drive. Walk the safety chain and find what is open, which usually takes less time than anything you could do at the keypad, since an e-stop button, an open gate, or an unreset safety relay will be obvious once you are looking for it.

If the chain is intact and the drive still reports STO, measure at the STO terminals themselves so you are testing the wire and not just the device. Check both channels, because this is the point where a one channel problem shows up as a 5090 and a genuine open shows up as both channels dead.

If the machine does not use the safety function at all, verify that the factory jumpers across the STO terminals are in place and tight. That is the legitimate configuration for a machine with no safety circuit wired to the drive, and it is the one thing on this page that is a wiring fix rather than a diagnosis.

Then, and only then, look at how 31.22 is set, because that is the answer to why the drive is reporting the condition the way it is rather than the answer to why the circuit is open.

The thing not to do

Do not jumper a safety circuit that is actually in use to make a machine run. It is worth saying plainly because it happens, usually under production pressure, usually with the intention of putting it back, and the jumper outlives the shift that installed it more often than anybody wants to admit.

A jumper across the STO terminals on a machine with a real guard defeats the one part of the system that does not depend on software being correct. If the machine needs to run with a guard open for a legitimate reason, that is a conversation about the machine’s safety design and a properly engineered maintenance mode, not a terminal block decision made at two in the morning.

Where this fits

Safety circuit codes are the ones technicians see most and understand least, largely because the drive is reporting somebody else’s circuit and the manual entry is three lines long. The pattern behind them is the same on every platform, and once you have it, the specific code number is just a lookup.

The full code list for these drives is in the ABB Drive Fault Field Guide, and the same ground on the Allen-Bradley side is in the PowerFlex 525 Fault Field Guide. The Yaskawa equivalent, including SToF, is in the Yaskawa Fault Field Guide. For how safety circuits and drive wiring get designed before any of this comes up, the VFD Installation Guide and the VFD Troubleshooting Guide are the parent guides.

If you want the competency behind the codes rather than the codes themselves, that is what the training is built to do, and Before the First Fault is the field record the whole approach came out of.

Codes and behavior on this page are taken from ABB’s published drive documentation. Confirm against the firmware and hardware manuals that match the drive and software version in front of you, particularly for anything involving a safety function. ABB, ACS380, ACS580 and ACS880 are trademarks of ABB, used here for identification.

Frequently Asked Questions

Is A5A0 a fault or a warning?

It depends on how the drive is configured. A5A0 and fault 5091 describe the same condition, which is that the Safe Torque Off circuit is open, and parameter 31.22 governs whether the drive reports it as an indication, as a fault on a start attempt, or both.

How do I clear A5A0 on an ABB drive?

You clear it by closing the safety circuit, not at the keypad. Find what opened, whether that is an e-stop, a gate switch, a safety relay that needs resetting, or a missing jumper on a machine with no safety circuit wired, and the condition clears when the circuit is restored.

What is the difference between 5091 and 5090?

5091 means both STO channels opened, which is a normal safe stop. 5090 means only one channel opened, which is a discrepancy between the two channels and a fault in the safety circuit itself, and it is the more serious of the two.

Does A5A0 mean my drive is damaged?

Almost never. The code reports the state of an input circuit rather than a failure inside the drive, so the usual causes are out in the machine. A repeated 5090 is the case where hardware, whether a safety relay contact or a cable, is genuinely failing.

Can I jumper the STO terminals to get the machine running?

Only on a machine that has no safety circuit wired to the drive, where the factory jumpers are the correct configuration. On a machine with a real guard or e-stop in that circuit, a jumper defeats the safety function and should never be used to work around a fault.

Why does it show up on some machines and not others with the same drive?

Because 31.22 is set differently, and because how a machine annunciates and recovers from a safe stop is a design decision made when the machine was built rather than a drive default.