A stepper motor that loses position on a long job is almost never a motor problem. It is usually a driver set to the wrong current, and the machine only makes it visible once the job is long enough for the error to pile up.
What makes it expensive is the timing. A badly matched driver runs perfectly for an hour. Then the axis drifts a few tenths, the part is scrap, and the operator spends the afternoon checking the mechanics instead of the driver - because mechanically everything looks fine, and it is.
Here is the order we work through when matching a stepper motor driver to a motor, and the settings that actually change the outcome.
The current setting on a stepper motor driver is the whole game
Every stepper driver has a current setting, and the factory default is usually higher than the motor wants. That matters because a stepper run above its rated current does not fail on the spot - it gets hot, heat changes the winding resistance, and the torque curve you sized the axis against quietly stops being the curve you have.
Set the driver to the motor's rated current first, then decide whether to go above it. Running at exactly rated current is the safe default. Some applications deliberately push a little higher for more low-speed torque, but that is a decision to make with a thermometer in hand, not by default.
Two ratings get confused constantly here. Peak current is what the driver can deliver in short bursts. RMS current is what the motor can take continuously. Setting a driver by peak instead of RMS is a reliable way to cook a motor slowly.
Voltage: why 48 V pulls harder than 24 V on the same motor
Supply voltage does not raise holding torque - that is set by current. What it does is let the driver push current into the coils faster, which is why the same motor on 48 V feels dramatically stronger at speed than it does on 24 V.
This is the cheapest upgrade available on a machine that runs out of steam at higher feed rates. If the axis holds fine at 500 mm/min and stalls at 1,500, and the driver is sitting on 24 V, the answer is usually the power supply rather than a bigger motor.
| Frame size | Typical rated current | Supply we normally pair |
|---|---|---|
| NEMA 17 | around 1.5 A | 24 V |
| NEMA 23 | around 3 A | 36 V to 48 V |
| NEMA 34 | 4.2 A and up | 48 V |
Treat those as starting points rather than rules. The nameplate current on the motor in front of you is the authority, and it is worth reading it instead of assuming from the frame size.
Open loop or closed loop, honestly
An open loop driver sends pulses and assumes the motor followed them. A closed loop driver reads an encoder and corrects when it did not. The price gap is real but smaller than most people expect, and which one is right depends entirely on what a lost step costs you.
If a lost step means a scrapped part, closed loop pays for itself the first time it happens. If a lost step means the operator re-homes the machine and carries on, open loop is a reasonable place to save money. What closed loop does not do is give you more torque - it gives you the ability to know when you have run out, which is a different and often more useful thing.
Our HBS57 closed loop driver covers the NEMA 17 and small NEMA 23 range and takes the encoder straight off the motor. For the larger frames, the NEMA 23 closed loop kits ship as a matched motor and driver pair with the current already set for each other.
The two wiring mistakes that cost an afternoon each.
Enable polarity set backwards, so the driver is disabled whenever the controller believes it is enabled - the motor holds position and refuses to move, which looks exactly like a dead driver. And PUL/DIR swapped at the controller end, so the axis turns but counts in the wrong direction. Both take two minutes to rule out and hours to find cold.
Microstepping is not a precision upgrade
Setting a driver to 8 or 16 microsteps sounds like it should improve accuracy. It does not, in any way you can measure on the finished part. Microstepping smooths motion and cuts resonance, which is genuinely worth having. Positional accuracy still comes from the motor and the mechanics.
What microstepping does change is the step rate the controller has to produce. Push it high on a slow controller and you can end up losing steps for the opposite reason to the one you were trying to fix.
Telling a lost step from backlash
These two produce nearly identical symptoms on the finished part, and they need opposite fixes, so it is worth ten minutes to separate them. Backlash shows up as a consistent error that changes direction with the axis: approach a position from the left and from the right and you get two different numbers, repeatably. A lost step does not repeat - the error grows over the job and resets when you re-home.
Run the same move twenty times in one direction without re-homing. If the error is climbing, it is the driver and its current setting. If it stays put and only appears when you reverse, the mechanics want attention and no driver setting will help.
What we need to match a driver to your motor
Three numbers off the motor label and one off the machine: rated current, rated voltage, step angle, and the supply voltage the control cabinet already runs. With those we can tell you whether you need a driver on its own or a driver and power supply together, and set the current before it leaves.
We stock stepper motors and matching drivers from NEMA 17 up to NEMA 52, open loop and closed loop, and the pair always ships set for each other. If you are chasing a step loss you cannot explain, send us the motor label and the axis details - there is a fair chance we have seen the same combination before.
If you are buying a driver to go with a motor you already have, the label is all we need. Send a photo of it with the supply voltage and we will come back with the driver and the current setting to use.
We supply stepper motors and drivers from NEMA 17 to NEMA 52, open loop and closed loop, with the current set for the motor each one ships with - send the motor details through here.
