Over Heating Stepper Drivers

If you find that you can cook bacon on your stepper drivers or the motors they control there is a way to stop that happening. I have been doing a lot of 3D printing lately mainly because I have been testing a new build plate adhesive that I had developed (more on that in the coming days)

During my testing, I saw that my beloved Hictop i3 printer was having problems with heating the hot end. I tested the board with a multimeter but realised that something had gone wrong with the part of the board that heats up the hot end. I tested it with the multimeter and the main power was showing that part was working. Then I tested the bed as it heated up and that was showing it had power. Still nothing from the hot end connections though. The message I kept getting was “reset the printer because the heating had failed” every time I tried to bring the hot end up to temperature. This particular board had seen a hard life one way or another. I had it in an enclosure, printing ABS for hours and I didn’t realise that the stepper drivers only had heat sinks or passive cooling. They didn’t have fans directly blowing air on to the tops of the stepper drivers. I’m surprised the board didn’t just get up and walk out!

I had a spare board laying about so I swapped it out and everything worked which was a good thing because it at least proved that I knew a little bit about using a multimeter. “Every cloud” as they say.

I reassembled the printer’s mainboard and I then loaded the Hictop marlin firmware. Once this was complete I decided to quickly check the calibration. I only wanted to double check it was still working after installing a different part. I usually check the calibration if I even look at my printers differently. It’s not like they are highly strung or anything is it.

After I checked the calibration I tried a calibration shape as a final tick in the list of things that the upgrade had to pass to be acceptable. This is where things went wrong, but the fix was really easy and so I decided to check my other two printers and found these to be out as far as this particular setting was concerned. That’s why I decided to write about it and share what I found and detail the steps to fix it, so it might help others and guarantee my place in heaven.

About 3 minutes into the print I saw a major layer shift and I mean about two inches! So I stopped the print and sat there wondering what was going on. I checked the usual things then thought I would slice it again. Just in case it was bad slicing by Cura that was causing the problem.
This time it went for about five minutes before it slid off about an inch at the top. I stopped the print again and checked the mainboard. I happen to touch on of the stepper drivers and to say it was as hot as Mount Doom from Lord of the Rings was an understatement. I would have badly burnt my ring had I leaned against it.

So there was my clue I touched the stepper motors and they were overly hot for the amount of work they had done too. I decided that the stepper drivers were getting too much power and they needed to be calibrated. I have to be honest I have never done this before. I’ll usually throw another fan at the problem and hope that fixes it. I was thinking of doing that again, but I realised that even if that did fix that particular problem the stepper drivers would still be getting too much power and thus causing too much heat in the motors themselves. Which would eventually kill them? So I should man up and deal with the problem head-on.

Tools you will need to tackle this issue with Amazon links if you need them.

  1. Cheap and cheerful multimeter you can get one from Amazon from this link http://amzn.eu/d/3qP0OlW.
  2. Some wire with alligator clips attached (although not compulsory) http://amzn.eu/d/f7PNHKQ.
  3. Small insulated flat or small cross head screwdriver.

The job itself is quite simple and will only take 5 minutes to do all of your stepper drivers. First, you need to find out what VREF setting your stepper driver should be set at. If you want a proper equation to how this is done it’s “Vref = 8 * current * Rsense” The drivers I was checking had a Rsense of 0.1 ohms, so 0.8V will give a 1A current.
For A4988 I set mine to 0.72V (slightly under) but before I started they were set at 1.56V so quite a saving in terms of power consumption. You will find that most of the common run of the mill Chinese stepper drivers are well overpowered when fitted to their 3D printer kits.

DRV8825 stepper drivers should be set at around 0.85V to 0.95V, but check your particular driver requirements first.
I have set my DRV8825 to 0.85V and they seem to be fine. Run the machine and try some fast moves of known distances. Make sure it moves the distance you asked it to move. If you hear a high pitched whine, but little or no movement that’s classed as a stall. You’ve probably reached the limit for that motor at that current level. If you need it to go faster turn it up a little and try again. As I said it’s not difficult to look yours up so do that. Then you will have peace of mind going forward.

Once you know what the correct VREF is depending on your drivers and stepper motors you are ready to begin.

You need to have the power switched on that goes to your RAMPS board otherwise you won’t get a proper measurement. Of course, if you are doing this yourself please be careful, it might only be 12V but that can still make you jump if you don’t show it some respect.
Reminds me of a rhyme, my friends, Dad used to say to us years ago. “Little Billy pair of pliers. Little Billy electric wires. Little Billy big blue flashes. Little Billy pile of ashes!!

So I want you to turn on your power supply to the RAMPS board. Then I want you to get the crocodile clip and attach one end to the metal part of the screwdriver. And the other end to the end of the red multimeter metal end. This allows you to turn the pot (potentiometer) of the stepper driver and get an up to date reading. See the pictures below for more information and a proper walk through. If you don’t have “Some wire with alligator clips attached” it’s ok you will just have to check what they started at. Make a note of that settings then turn the top of the potentiometer and retest it again. Note: the screw on top of the pot requires very little turning before it changes. Test and change, test and change until you get it where you need it to be.

The Multimeter setting you need to go to for doing this is the 20V setting

Once you have attached all the wires you will now place the black (ground) probe on the ground connector of you RAMPS board.

The red wire which should be attached to the insulated screwdriver goes to the pot (potentiometer) adjustment screw. It should be mentioned here that you need to be a bit careful anddon’t go randomly stabbing things with the screwdriver. If you do you may short the board and do some damage. Be careful and just take your time.

Making sure you can see the screen of the multimeter and before doing anything else, read the value that it is already set to. That way you are making a mental note if you need to back it off to what it was in the event that things don’t work. But that’s not going to be needed because you know what you’re doing! You do know what you’re doing right? Here was mine.

Now you have that information gently turn the screwdriver one way or the other till you start to see the number on the multimeter go down. Keep doing that nice and slowly till the multimeter reads the VREF that you want your stepper driver to be set too. In my case it was 0.72V. When you have done the first one move on to the rest set them up and in no time at all you’re done.

When you have finished it’s worth doing a little printed object test like a calibration cube Just to make sure it still has its tolerances and calibration in tact. If you have another printer it’s worth testing its stepper drivers with your new found skill. If you’ve worked through this till the end I applaud your tenacity (clue for an up coming post) and thanks for you time.


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.