So... I'm a bit embarrassed to admit this... but I have no idea how to use my scope. I managed to get through my electronics projects without one but several times when I had an issue, people would ask me what my "scope was showing". So I picked up a scope... now how do I use it?

Here's my sample issue. I have a video signal that being interfered with by a brushless motor running on the same battery. I want to build an LC type filter to clean up the line, but rather than a shot in the dark, I'd like to actually have some target frequency to look at, plus I'd like to know if the problem is with the camera, the transmitter, or both. Here's the setup... where exactly do I attach the probe and what settings do I choose?

Thanks.

- Ashamed

 

Begin by telling us what brand and model scope you have.
Do you have the User's Manual?
What probes do you have?
Are your probes 1x or 10x probes?
What is the level of your electronics knowledge and experience?
Sometimes stating your age is helpful but you don't have to if you prefer not.

 

Begin by telling us what brand and model scope you have.

QuantumAsylum QA100

Do you have the User's Manual?

Yes. Like many new and daunting toys... the instructions are just getting me more confused. Plus, I'm dyslexic so reading comprehension isn't the best for me.

What probes do you have?

100Mhz

Are your probes 1x or 10x probes?

Selectable by a switch on the probe

What is the level of your electronics knowledge and experience?

Very mixed. Tough question to answer. On the one hand, I have no formal training nor would I call myself an avid hobbiest. On the other hand, I've developed several commercially successful electronics gadgets ranging from r/c aircraft accessories to automotive electronics to an industrial washing machine controller as part of a government contract. While there is a ton I don't know, I am good at learning what I need to know to get a job done. I have gaps though. Its similar to how I am quite good with calculus but I still count on my fingers sometimes.

Sometimes stating your age is helpful but you don't have to if you prefer not.

30's. So not a child or a geezer if that's what you're concerned about lol.

Cheers.

 

Intro that's pretty general

Half the video is done on a whiteboard, so it's not specific to a scope, and good general information. The second half shows using those concepts on a Digital scope.

There are more specific tutorials, and more advanced, which should show up in "related videos".

When you have the basics down, read the User Manual. On page 51, it shows how to run an FFT of the displayed waveform, which is a plot of amplitude vs frequency (rather than the standard amplitude vs. time). This lets you see the peak frequencies of the noise.

 

That's a great start, thank you.

 

Attach the probes to the battery, since you believe the noise exists there. Clip the probe's ground wire to the battery negative terminal. Set the vertical coupling to AC, this will filter out the 12VDC and show only the noise. Set the vertical V/Div to mid-setting. Set the Horizinal timebase to around 30mS/div, whichever setting is close. That would give you one complete sweep for a 3600RMP motor. For a different RPM, use Setting=10/RPM. Set the trigger source to INT. Now start your motor. Turn down the V/DIV know until you see noise showing on the screen (the horizonal line gets all wiggly.) Now rotate your trigger level control until the display seems to suddenly "freeze", in the horizonal direction. The display will still show some changes, but you'll begin to see distinctive wave patterns that you didn't see before. Try turning the trigger polairty to "+" and "-" and see how it changes the display. If the display gets all wiggly again, you've lost lock. In that case, adjust your trigger control until the display freezed again in the horizonal. If you have any trouble with the above, let us know.

 

He could try hooking up the probes and hitting the "AUTO" button. If it's a decently designed scope, it should make those setting changes.

 

Oh that reminds me, to get an initial flat line display turn your trigger "Auto" "Normal" control to "Auto" After getting your scope set up, try turning it back to "Normal" If you're running in Normal mode and you lose your display, then adjust your trigger level. In Normal mode, if you lose lock, the display ceases scanning.

 

If you have the time, and really want to learn the basics of scopes (mainly analog scopes), take a look through the 14 videos in this playlist:
http://www.youtube.com/playlist?list=PL746BF38BC2E068E0

 

Oh that reminds me, to get an initial flat line display turn your trigger "Auto" "Normal" control to "Auto" After getting your scope set up, try turning it back to "Normal" If you're running in Normal mode and you lose your display, then adjust your trigger level. In Normal mode, if you lose lock, the display ceases scanning.

w2aew did an EXCELLENT job on This video on triggering.

 

Attach the probes to the battery, since you believe the noise exists there. Clip the probe's ground wire to the battery negative terminal. Set the vertical coupling to AC, this will filter out the 12VDC and show only the noise. Set the vertical V/Div to mid-setting. Set the Horizinal timebase to around 30mS/div, whichever setting is close. That would give you one complete sweep for a 3600RMP motor. For a different RPM, use Setting=10/RPM. Set the trigger source to INT. Now start your motor. Turn down the V/DIV know until you see noise showing on the screen (the horizonal line gets all wiggly.) Now rotate your trigger level control until the display seems to suddenly "freeze", in the horizonal direction. The display will still show some changes, but you'll begin to see distinctive wave patterns that you didn't see before. Try turning the trigger polairty to "+" and "-" and see how it changes the display. If the display gets all wiggly again, you've lost lock. In that case, adjust your trigger control until the display freezed again in the horizonal. If you have any trouble with the above, let us know.

Alright, I found some time to give this a try. I'm really not sure what I'm looking at though. Does this mean anything to you? First screen is with motor off, 2nd is with motor on. Thanks.

 

Looks like your primary noise is at 150 Hz (3 per 20mS division), and your secondary is 550Hz or so (11 per 20mS division).

Rough guess. Compare with what is in the circuit that can cause noise.

 

Ok, I get you... so we're just looking at peaks in amplitude and counting them per div. Not NEARLY as preciese and/or informative as I would have expected from a tool that's supposed to be invaluable in this sort of thing, but whatever... its progress.

So I put in a simple L-C filter that's sold on the market for this particular application. It didn't help, and it makes using the scope even less useful in that I can no longer visibly discern between "motor on" and "motor off" with the filter in place. The noise appears much worse in both cases - here's an example - which is obviously counter intuitive. In application, the effect of the motor appears to be identical to what it was before.

 

It now looks like the L-C is oscillating, rather than filtering.

With the original waveform on the screen, switch to FFT mode and post that. It will give more accurate frequencies and levels from the full sample memory, rather than counting peaks per division, as I did.

That was how noise frequency was decided before scopes that did math. The FFT will tell you more detail.

What values are you using for the inductor and capacitor? You could probably leave the inductor out and have a better result, since the motor acts like an inductor.

 

Its a filter supposedly designed for the purpose. It has a 1000uF cap. The inductor (its a torroid wrapped in wire) is unmarked. Its kind of why I wanted to do the analysis so I could make one with specific values for my particular situation.

I'll do as you say. Thanks for the tips.

 

What amount of current does the motor draw?

Is it variable speed or fixed RPM?

Brushless motors are usually very quiet electrically, computer fans are entirely BLDC motors.

The cap across the supply for the motor should help, I'd also suggest adding caps to the input power of the other devices as well.

Shortening the length of the power cable can help as well, but that isn't always an option. If you could make a run from each device back to the power source, rather than tapping into the wire in several spots, that should improve issues.

The noise magnitude is small, is it physically close to the other circuits to the point EMI from the motor could be adding to the problem?

 

I don't think its EMI, as I can physically separate the components to some degree and I notice no difference. I would think if that were the issue, then moving the camera and/or transmitter away should improve the signal at least some, but it does not.

The motor draws a max of about 11A. In my tests its less than 3. It is variable and is controlled by a speed controller. The power cables are all quite short. Battery to controller is 3". Controller to motor is 3.5". Battery to 5v reg is about 3" and reg to camera is about 6". Battery to transmitter is the longest at around 10".

 

With the BLDC being variable speed, the noise frequency should vary with the motor speed.

Is that the case?

 

Yeah. There's a big spike when it first turns on, the it settles but it does increase in amplitude slightly over the range of speed but never seems to get as much as that initial spike.

 

Have you figured out how to get your scope into FFT mode?

You should see the "spikes" move left and right at different speeds.

With that, you'll know the full range of your noise issue.