Author Topic: measuring frequencey with an Oscilliscope  (Read 2790 times)

0 Members and 1 Guest are viewing this topic.

Offline kennTopic starter

  • Contributor
  • Posts: 10
  • Country: us
measuring frequencey with an Oscilliscope
« on: August 04, 2016, 04:04:40 pm »
I am trying to use a OWON DS7102 scope to measure the frequency of a toy remote control. I would have thought that I could hook the probe up to ground and then the antenna and see the frequency of 49mhz which is what the crystal says however that isn't the case, depending on what I set the seconds per division to I get various different ranges but it's always changing a little bit no matter what. For example, at 50ms when it's just sitting there it sits a fairly stable 60hz however when I move a control is jumps to like 105hz and constantly changes going a little above and a little below, changing the seconds per division changes the hz, for instance at 20ns I get nothing for the frequency when sitting there and when I move a controller I get closer to 49mhz, it's like right at it +/- .5mhz.

I guess what I am trying to figure out is, what am I doing here? I am increasing the sampling rate correct? If so - how do I know what is right when I am trying to measure something that didn't have a sticker on it saying 49mhz? I have watched various oscilloscope howto's but either they didn't explain this or I didn't understand them.

Thanks for any advice.
Kenn
 

Offline rstofer

  • Super Contributor
  • ***
  • Posts: 9886
  • Country: us
Re: measuring frequencey with an Oscilliscope
« Reply #1 on: August 04, 2016, 04:15:23 pm »
Is it possible the controller only transmits when controls are changed?  You said you got the right answer when <something> was moved.  I didn't understand that bit.

You probably want to be sure you are using AC coupling, a very low Volts/div and you might do better to hang a long wire (antenna) on the probe and not make a direct connection.  And make sure you are using the x10 setting on the probe.

Most scopes will include some kind of frequency counter but to get them to work you have to get a stable waveform.
 

Offline mathsquid

  • Regular Contributor
  • *
  • Posts: 247
  • Country: us
  • I like math.
Re: measuring frequencey with an Oscilliscope
« Reply #2 on: August 04, 2016, 04:21:21 pm »
For example, at 50ms when it's just sitting there it sits a fairly stable 60hz however when I move a control is jumps to like 105hz and constantly changes going a little above and a little below,

The steady 60Hz is probably the scope picking up the mains frequency.
 

Offline klunkerbus

  • Supporter
  • ****
  • Posts: 162
  • Country: us
  • Electrical Engineer (retired early)
Re: measuring frequencey with an Oscilliscope
« Reply #3 on: August 04, 2016, 04:27:52 pm »
Does the remote control have an on/off switch?  If not, the remote will predictably only transmit when some control movement is being made.

When the frequency of a test signal is unknown, you can always start with a high sweep rate/high sample rate and work your way to a lower setting.  Undersampling errors can mislead you if you start at low sweep rates and work upwards. 
« Last Edit: August 04, 2016, 06:18:51 pm by klunkerbus »
 

Offline kennTopic starter

  • Contributor
  • Posts: 10
  • Country: us
Re: measuring frequencey with an Oscilliscope
« Reply #4 on: August 04, 2016, 05:40:22 pm »
Thank you all,

First off - Yeah, you are probably right about the 60hz being mains, but check this out, I increased the sweep rate to 10ns with 100mv per division and also took the antenna from another toy remote and connected that to the scope instead of direct coupling it. The last thing I did was to switch the probe to x10. With this configuration I get a nice steady 49Mhz (and yes the controller needs to have a stick moved in either up/down or left right to work but holding it in place keeps it on).

The thing I wanted to find out was how could I take something I know nothing about and find out what frequency it's transmitting at. It seems that when you get the sweep rate set about right (it's represented as seconds per divisions on my scope) changing it a little bit won't impact the reading, in this instance, at 2ns, 5ns, 10ns, and 20ns I get a pretty steady 49Mhz single with vary little deviation from that reading.

Thanks all - just learned a lot about my scope and how to measure frequency.
 

Offline rstofer

  • Super Contributor
  • ***
  • Posts: 9886
  • Country: us
Re: measuring frequencey with an Oscilliscope
« Reply #5 on: August 04, 2016, 05:57:25 pm »
A 50 Mhz signal has one complete cycle in 20 ns.  If you set your scope to 20 ns/div, you will get one complete cycle every division.  If your scope has 10 horizontal divisions, you will see 10 complete cycles.  That is, it takes 200 ns to scan across the screen.

If you had 60 MHz, each cycle takes about 16 ns and in that 200 ns scan, you would get about 12.5 cycles.

If you pick up some arbitrary transmitter, just get it to display on some scale and try to get as few cycles on the screen as possible.  Measure the division (and fractions of a division) and multiply by the time/div.  Then take 1 over the time/cycle to get frequency.  Or, look at the frequency counter on the screen (if it has one).

It just takes a little math.
 

Offline kennTopic starter

  • Contributor
  • Posts: 10
  • Country: us
Re: measuring frequencey with an Oscilliscope
« Reply #6 on: August 04, 2016, 07:47:06 pm »
Sadly That never dawned on me till you explained it - YES! hertz is a measure of time, so of course, when you want to measure the frequency of something you need at least set the sweep of you meter to the lowest time it takes to get a single complete cycle. That is why I can set it to 2ns, 5ns, 10ns or 20ns and it's fine, however once you increase past that the measurements of frequency are wrong - note the scope is displaying frequency for me.

If I was a bit more clever that would have dawned on me but now I know...

Thanks!
 

Offline alsetalokin4017

  • Super Contributor
  • ***
  • Posts: 2055
  • Country: us
Re: measuring frequencey with an Oscilliscope
« Reply #7 on: August 06, 2016, 01:04:41 pm »
Actually Hertz (Hz) is a measure of "inverse time", that is, events PER second, in our case waveform periods or cycles per second. So for example 60 Hz means 60 events, periods or cycles PER second.
 
Now you might try practicing calculating frequency of a periodic signal using the scope's graticule markers and the horizontal timebase setting alone, without relying on the scope's built-in frequency counter. This will serve as a "sanity check" when you start working with more complex signals so that you can see if the scope is being honest with you.

It's a mistake to rely totally on the "numbers in boxes" that a DSO can spit out; one of the great advantages of having a waveform display is that you can check the automatic measurements yourself using the graticule and the H and V settings of the scope
The easiest person to fool is yourself. -- Richard Feynman
 

Offline w2aew

  • Super Contributor
  • ***
  • Posts: 1780
  • Country: us
  • I usTa cuDnt speL enjinere, noW I aR wuN
    • My YouTube Channel
Re: measuring frequencey with an Oscilliscope
« Reply #8 on: August 08, 2016, 08:24:02 pm »
This video might help you to better grasp the relationship of the scope's sample rate, waveform sample points, signal processing, etc.


YouTube channel: https://www.youtube.com/w2aew
FAE for Tektronix
Technical Coordinator for the ARRL Northern NJ Section
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf