Brushless motors have a lot of advantages over traditional brushed motors. However, testing them can be a bit of a pain. Because the resistance of the motor’s coils is usually very low, a standard resistance check isn’t likely to be useful. Some people use LC meters, but those aren’t as common as a multimeter or oscilloscope. [Nils Rohwer] put out two videos — one two years ago and one recently — showing how to test a brushless motor with a multimeter or scope. Oh, you do need one other thing: a drill.
You don’t have to drill into the motor, instead you use the drill to spin the motor’s shaft. Since a motor and a generator are about the same thing, you can read the voltages produced by the spinning motor and determine if it is good or not. The first video shows the technique and the second, more recent video shows a scope reading a bad motor. You can see both videos, below.
In the second video, you’ll see that a pair of coils have shorted — probably due to a failure of the insulation, perhaps due to heat — and are putting out about 1/10 of the voltage a good coil produces. In the end, he opened the faulty motor and looks at the suspect coils, even though there wasn’t much to see visually.
We’ve seen working brushless motors used as encoders. You can even build your own brushless motor demonstration, or opt for a more practical design.
17 thoughts on “Testing Brushless Motors With A Scope (or A Meter)”
You can actually do much more using this method, I use it to measure the Kv rating of brushless motor I did rewind.
By coincidence, I wrote a blog post on that last night:
Wow! Added that link to my bookmarks!
I’m a bit surprised the oscilloscope demonstration worked. Nils measured the voltages across the coils by connecting each scope probe and its ground across a coil. I expected this to require differential voltage probes for the scope, as I thought that all probe grounds are common. However, this is apparently not the case, as he is indeed measuring voltages.
Can anybody shed some light on why this is working? Thanks in advance.
Just hooked an ohm meter across the ground clips of the 10x analog probes on a Tek MSO2014B scope. Measured a constant approx. 0.4 ohm between the ground clips, regardless of the coupling mode (dc, ac, ground) of each channel.
Assuming the presenter’s scope is operating the same way, he may have been passing current through the scope’s ground plane, which is a no-no.
I haven’t watched the demo yet, so I don’t know what scope is being used. But I’m pretty sure my old tek ths720 portable scope has isolated channels.
I’ll check when I get a chance.
Thanks. Did some digging and it looks like your Tek THS-* scope belongs to a special class of oscilloscopes that has isolated channels. See the blue box on page 4 of below Tek doc for details.
@All – This Tek doc gives a good overview of this sort of measurement problem and compares a number of approaches.
Fundamentals of Floating Measurements and Isolated Input Oscilloscopes Application Note
Thanks for sharing that application note. I did confirm that there is no measurable DC resistance between the ground clips on my 720, but you probably already know that from this Tek doc.
I’m no scope expert, but I had been under the assumption that scopes that could run off of batteries wouldn’t/couldn’t have truly grounded commons, and thus were more likely to have completely isolated commons.
When the scope runs off batteries, there is still a good chance that all the channels share a ground signal. When that’s the case the circuit is so much simpler…
Dave Jones discusses this topic on his EEVBlog.
EEVblog #279 – How NOT To Blow Up Your Oscilloscope!
I didn’t see a view of the connections in the oscilloscope video, but something doesn’t seem right. There’s only three leads to the motor, so if you ground one, it would only leave two traces on the oscilloscope. He may have left the grounds floating. (or connected them to a chassis piece that is floating) The probes do have a high-impedance connection to ground, so even without a connection between the probes, it is still possible to detect the AC differential between the coils.
Scope connections shown briefly at 6:40 in first video.
Here’s a link to that part:
Seems like each probe’s ground is connected to next signal/probe’s tip:
– Yellow probe’s ground is connected to red probe’s tip
– Red probe’s ground is connected to green probe’s tip
– Green probe’s ground is connected to yellow probe’s tip
You mean this guy shorted all windings together and then measures some low mv signals from his motor while using his scope probe cables as heating elements.?
At least the capability of delivering current from a bad motor also diminishes if one of the coils is broken.
Gerald, Paul I think you guys are right.
But then again, perhaps he should check every new motor he buys for defects just to be sure :)
Now if it just weren’t such a pain to rewind fried motors…
If something sucks, automate it…. then it can suck with an arduino.
The RPM calculation actually has to do with the number of magnetic poles (number of magnets in the rotor) in the motor not the number of windings. The windings are proportional to the amplitude of the voltage not the frequency. To calculate the RPM of the drill press => RPM = 360 * frequency / number_magnetic_poles
Please be kind and respectful to help make the comments section excellent. (Comment Policy)