"Count" vs Frequency?

[bitman]

I have a Rigol 1054 - and I'm practicing to get familiar with the scope and some basic principles.  I have a function generator creating a sine wave at 5KHz and my meter shows it spot on - however initially when I used the scope the count on the top was WAY off (see picture). I then picked "Frequency" on the measurement side and I get the exact number I was looking for.

Question: What's the difference? Why is one number way off? I see a square wave on the count, so I wonder if that has anything to do with anything?

I'm definitely a newbie struggling with some fundamentals here. My quest is to learn, so I'll be happy to be brought to materials that can teach this and other fundamental stuff.

[ataradov]

Count comes from the actual trigger counter. Measurements below are performed on a captured signal.

Try to enable bandwidth limiting in the channel settings. And also try to move the trigger level to 0.

[Bud]

It is a very bad idea to use chinese made garbage to learn the fundamentals.

[ataradov]

It is a very bad idea to use chinese made garbage to learn the fundamentals.

I have a surprise for you. American made Teks behave exactly the same.

[bitman]

Count comes from the actual trigger counter. Measurements below are performed on a captured signal.

Try to enable bandwidth limiting in the channel settings. And also try to move the trigger level to 0.

Worked like a charm. Make quite a bit more sense now.

[Fungus]

It is a very bad idea to use chinese made garbage to learn the fundamentals.

Welcome to the Internet, bitman.

[Fungus]

Question: What's the difference? Why is one number way off?

They're both calculated differently.

The number at the bottom is calculated from the data on screen. The value will depend on your zoom, the number of visible waves, the noise on those waves, etc. This is usually the preferred number, although you always need to bear in mind what's visible on screen (eg. it won't work at all if there's only half a wave visible).

The number at the top comes from the trigger circuit. It counts the number of times the trigger was fired per second. If there's noise on the signal you might get extra counts as it wobbles past the trigger voltage level. This is why turning on the bandwidth limiter can help - it's a low pass filter on the input and can help a lot with the accuracy of measurements when you're looking at low frequency signals.

Note: Neither of these methods is guaranteed to be 100% accurate in all circumstances. That's oscilloscopes for you.

(Note to Bud: this is true of all oscilloscope, not just this one).

You can also take measurements manually. Capture a single wave on screen with as much zoom as possible. Turn on cursor mode. Move the cursors to the start and end of the wave you just captured.

I see a square wave on the count, so I wonder if that has anything to do with anything?

That's just a graphic, nothing to do with anything.

[bd139]

Just a point with the DS1054Z counter. It is pretty accurate. I've used mine to align radios quite happily when I couldn't be bothered to dig a cable for the counter out.

If you're going to measure frequency of your function generator however, connect the sync output (if it has one, if not separate square wave output) to channel 4, tell the scope to bind the counter to channel 4 and use that instead. Also set your trigger to trigger on the sync output as well. Sometimes it's easier to measure and trigger on a reference signal on the generator than the actual signal you're looking at. If for example you turn your sine amplitude down below the trigger threshold or you want to move the probes to different bits of the circuit then it'll still trigger fine and you can make realistic measurements. Also you can use phase measurements against a reference point then as well so if you run your signal through a filter you can measure what happens.

It is a very bad idea to use chinese made garbage to learn the fundamentals.

They work fine and are affordable.

[jpb]

Question: What's the difference? Why is one number way off?

They're both calculated differently.

The number at the bottom is calculated from the data on screen. The value will depend on your zoom, the number of visible waves, the noise on those waves, etc. This is usually the preferred number, although you always need to bear in mind what's visible on screen (eg. it won't work at all if there's only half a wave visible).

The number at the top comes from the trigger circuit. It counts the number of times the trigger was fired per second. If there's noise on the signal you might get extra counts as it wobbles past the trigger voltage level. This is why turning on the bandwidth limiter can help - it's a low pass filter on the input and can help a lot with the accuracy of measurements when you're looking at low frequency signals.

Note: Neither of these methods is guaranteed to be 100% accurate in all circumstances. That's oscilloscopes for you.

On my scope (WaveJet) the hardware trigger counter is much more accurate than the values from the displayed waves - it reads to 6 sf and is only about 2 counts down (when compared to my standalone counters and gps OCXOs). The screen values are much more approximate (4 sf at most) and vary up and down by relatively large amounts.
And, as you say, the screen ones are very dependent on what is being displayed.

[Fungus]

On my scope (WaveJet) the hardware trigger counter is much more accurate than the values from the displayed waves - it reads to 6 sf and is only about 2 counts down (when compared to my standalone counters and gps OCXOs).

Yes, they can be. It depends on the amount of noise on the input.

They certainly have more digits.

[bitman]

It is a very bad idea to use chinese made garbage to learn the fundamentals.

Welcome to the Internet, bitman.

Thanks but I cannot help chuckling. I haven't heard that since the mid 1980s ...  (and since then, it's been what I've told others). While I'm new in electronics I'm not new in IT or by any means "young". Just a hobby I wasn't able to pursue when younger. I've been around the block many many times when it comes to stupid stuff on the net - this is mild and frankly I barely noticed it had it not been for "you guys" repeating it.  And yes, you read the time right. I was online before most knew what the internet was, before browsers - heck I didn't even use PCs   Anyway if you can imagine it, I've seen it online.

I've lurked for quite a while here, asked a few dumb questions in "beginners" and by far this has been enormous helpful in getting me on the right track when I screw up. I see very little noise and amazingly a very high percentage of on-topic replies/posts. Kudos to everyone here and the admins for keeping things flowing and useful. And thanks for having me

[effectivebits]

You want to measure frequency where the slew rate is the fastest.  Near the top of a sinusoid is the wrong place to do it.  Notice that your measurement is made at the 50% point, but your "frequency counter" appears to be near the top.

I'd recommend you move your Trigger Level to the middle of the waveform and try again.

Joel Avrunin