I’m bringing up a couple of DRV8825 stepper driver modules and attempting to set the current limit via Vref.
I’m measuring Vref (GND → trimmer) as per standard guidance (e.g. Pololu docs / common tutorials, e.g. https://www.youtube.com/watch?v=89BHS9hfSUk&t=192s), with the board powered and RST/SLP held high. No motor is connected at this stage.
Observed behaviour:
Vref typically reads around 20–30 mV
While adjusting the trim pot, it can suddenly jump to around 3.2–3.7 V
With even a very small additional adjustment, it drops straight back to ~20–30 mV
I cannot achieve a smooth or continuous adjustment across the range
The behaviour is repeatable but highly non-linear / discontinuous
Additionally:
If I apply downward pressure on the trimmer while adjusting, I can momentarily get a stable and controllable Vref reading
As soon as I release pressure, Vref collapses back to ~20–30 mV
The trimmer rotates very freely (can be turned with finger/thumb), which seems unusual
What I’ve ruled out:
Tried two separate DRV8825 boards → same behaviour
Removed breadboard and wired directly → same behaviour
Common ground is present
VMOT is powered from a bench supply (~12 V, current limited for initial testing)
Questions:
Is Vref expected to be measurable and stable without a motor connected? (my understanding is yes)
Does this behaviour point to:
a faulty trimmer (wiper contact issue), or
poor soldering/mechanical connection on the module?
Has anyone seen clone boards with this kind of trimmer behaviour?
At this point it feels like a mechanical/electrical issue with the trimmer rather than a configuration issue, but I’d appreciate a sanity check.
Hi Adrian
Welcome.
Remove the “trimmer” and temporarily connect a “proper” potentiometer or multi turn trimmer in its place and try that.
If these 2 devices are cheap clones then anything is possible. I think this type of trimmer can be a bit dodgy at the best of times and sometimes seem a bit flakey.
Cheers Bob
@Adrian263969 yes — Vref should be measurable. The clue’s in the name. VRef is ‘V Reference’.
Yeah, that sounds like a connection error. If you can physically improve the voltage reference just by pressing down on the trimmer, that’s usually a pretty good sign there’s some dodgy solder on it or the component itself is playing up.
It might be worthwhile closely examining the trimmer’s joints for anything that could indicate a poor join. @Robert93820’s suggestion is the one I’d try next.
I wouldn’t trust those modules for setting motor current until that’s sorted — a jump from a few tens of mV up into the volts is the sort of thing that can make current limiting unpredictable. If you can post a couple of close-up photos of the trimmer area, that should help confirm whether it looks like the part itself or the soldering around it.
This of concern as the “top” of this pot should be connected to 3.3V hence the reading at the slider should not go beyond this.
As you are getting EXACTLY the same results with 2 devices I would also check measurement techniques. or possibly your DMM accuracy. In fact DMMs are NOT the holy grail they are sometimes (read mostly) made out to be.
Another idea would be to establish what voltage at this point to get the current limit you need. Then calculate a divider using resistors and replace the pot with these components. Provided the chip input resistance is high enough this should get you close to what you want. The overall resistance should be about 10kΩ to emulate the pot you are removing. This method should be OK as this setting should be a “set and forget” sort of thing. It could actually be more reliable in the long run.
Cheers Bob
Hi Adrian
I just had a look at that link about setting and measuring current in the stepper coils. Don’t agree with all of it. Her conversion to power to estimate what is used is valid but her measurement of current is not. She uses the mA range on the DMM. Now ALL meters have a thing called “Voltage Burden” and most quality instruments will document this. They measure current by measuring the voltage drop across a known resistor value. For the mA range this could be quite high and this resistance is effectively in series with the load, in this instance the motor coil. That is where the apparent missing power went. The voltage she measured across the coil is lower than it would be in practice with no meter in circuit.
Where you have lots of volts (hundreds maybe) and only a few mA this voltage burden (the voltage drop across the DMM) does not matter in the practical sense but as the voltage lowers this becomes a greater percentage of what is available so when you get down to 5V or less this effect becomes intolerable and measurements are meaningless.
The fix here is to use the 10A (or 20A) range where the resistance is very small (probably about 0.1Ω) so will have minimal effect. Most meters of any quality will still have a 1mA resolution on these ranges.
So the bottom line would be take note of most of this video but filter out the bits that are not quite kosher.
Cheers Bob
PS: I will humbly apologise to the lady concerned. The video is pretty out of focus and when I checked and enlarged the screen it would appear that she did in fact use the 10A range. I mistook that blob in front of the “A” as being “m” so I thought it was mA. Sorry.
I will not delete the above however as if you were not aware of this particular pitfall when measuring current it will not hurt for you to know about it anyway.
