Don't Fall Into This Output Impedance Trap!

[@Elozt]

I fell into the same trap when I was starting out too, kinda learnt about this the hard way. Glad someone finally points this out to all the new players in the field :)

[@JimGriffOne]

Thanks for the info. A "trap for young players", as Dave from EEVBlog would say!

[@KeysightLabs]

Yep!

[@alienozi]

That scope is a Bobby Dazzler! As Dave would say. :)

[@SusiBiker]

I got caught by this effect on a Nicolet function generator - many many years ago. Wrecked some kit I was working on at the time. Hope others learn from this - the easy way! Nicely done.

[@bfx8185]

It's good to say rest of story ... cheap oscilloscope's doesn't have possibility to switch to 50ohm impedance. Expensive ones yes . In this case you have to use T-BNC with 50ohm terminator. Also because not all of function generator's have possibility to switch to high impedance mode, which is mostly also not so precise way for measurement. Therefore is suggested use of T-BNC and 50ohm terminator and do all measurement at 50ohm which is standard or switch osciloskope to 50ohm if there is this feature.

[@SCKentrol]

I fell into this trap a couple of months ago when I was working on a filter for signal capturing. For the first 20 minutes or so, when the scope was showing the signal hitting the rails of the amplifier output, i was worried it populated the wrong gain resistors. Panicking, rechecking under a magnifying glass, all the resistor look right... then my partner tap me on the shoulder and pointed to the signal generator. Good (heart racing) times.

[@AndreasSpiess]

Thank you for this quick tip. I thought my function generator is weird. Because I trusted my Keysight oscilloscope, of course ;-)

[@RoGeorgeRoGeorge]

Almost correct. Should have write the red 2V on the left side of the internal resistor, at the node between the internal voltage source and the internal resistor.

[@HerrHeisenheim]

Just to make sure I get this right : changing the 50Ohm/HiZ setting on the function generator only applies a x2 scale on the voltage amplitude, it doesn't change anything in the actual circuit? I've not used a signal generator yet but I assumed this setting electrically added or removed (shorted) a 50Ohm resistor in series with the output.

[@Caenem_]

At my job I've got a function generator controlled by a test program and there is a "output impedance" parameter set to 50 ohm. I tried to measure a basic signal from it with an oscilloscope but the amplitude measured on the scope was twice as much as the output amplitude of the function generator. Then I saw a parameter on my oscilloscope represented by an Omega letter and there were 2 choices : 50 or 1M. The parameter was set to 1M. I tried to switch to 50 ohm and boom, the amplitude matched the amplitude of the function generator.

[@user-gk1ls8wi8f]

Thanks for the video it helps a lot. One question: in the illustration at 3:26, is the 2V setup supposed to be at the left of the 50Ohm internal resistance(instead of on the right side)?

[@phuang3]

Yes. He just want to emphasize that the internal circuit is ouputing 2V peak-to-peak waves.

[@gusel9]

Before I watch this video, I just kind of new that if I got twice the voltage it was OK. Thanks to this video I finally got it.

[@cj691]

I put my Agilent MSO7035A in 50 ohm mode and nothing appears to have changed between it and an Aim/TTi TG5011A while also in 50 ohm mode.

[@tarunverma2887]

what output impedance we shall use. 50 ohm or High Z pls clarify. Or selection of output impedance depends upon? pls clarify

[@KeysightLabs]

Use 50ohm if you need to impedance match to 50ohms on your device under test, high impedance if you aren't. As you get to higher frequencies, designs will shift to a 50 ohm system.

[@tarunverma2887]

@@KeysightLabs thanks for your prompt response. So we can say that use high Z always excepts a load having an input impedance of 50Ohm.

[@douglasghiorso5978]

Can you match the rf wave to a 12 volt wave and try to power a electric motor with it.

[@BaldurNorddahl]

a very small motor sure. The maximum current will only be 12/50 = .24A and the max power .24*12 = 2.9 watt. Signal generators do not generate a lot of power. You will need to amplify the signal to do any real work.

[@dabdoube92]

You got me out of a trap I didn't know I was in. Nice ( :

[@AHMEDRAZA-yb8qm]

Sir, twice the output of the function generator is visible in the oscilloscope. Whereas the output impedance of the function generator and the impedance of CRO form a voltage divider circuit, so half of the input of the function generator should be visible in the oscilloscope. But it shows twice of FG input in CRO .Why so ? please explain

[@KeysightLabs]

Good question! The scope compensates for it's internal 1M resistance so it's not really a factor. It's eventually the opposite of the 50ohm phenomena I show in this video

[@hubercats]

A very nice presentation (clear and concise). Thank you!

[@athulyaskumar2995]

Thanks for the information.Well Explained sir! This was really a doubt.Can I know how this changes when we change our probe ratio?

[@KeysightLabs]

The probe ratio basically just divides down and provides a smaller level input to the scope (providing more isolation)

[@KeysightLabs]

What other weird test gear quirks have you found?

[@UFObuilder]

Good information thumbs up

[@Jim_Bo]

I got this mutlimeter which sometimes has the nasty habbit of showing me double the voltage I'm measuring... it's kinda crazy when you're measuring for 240v and the meter indicates 480. I traced it back to a problem with the mode dial on the multimeter, a faulty/dusty contact and a little bit of wear on the dial, causing one of the contacts to not fully engage. Luckily for me, first time I noticed, it was when I was measuring a used 9v battery, which magicaly showed it still had 14v on it.

[@KeysightLabs]

Jim Bouche That's crazy!

[@JimGriffOne]

When I connect my oscilloscope probes to the mains supply, it tends to spark a lot inside the scope and not work any more. Probably not a quirk, though.

[@KeysightLabs]

Jim Griffiths haha, yes. Not recommended :)

[@muhamadsherko8801]

so we should use a hight output impedence to get exact result from device

[@LousyPainter]

Thanks! I learned something that I misunderstood. 👍

[@CLGilbert]

I'd love to know why I would ever use 50ohm mode. Why does this trap exist?

[@marekf5]

You could have also put 50 ohm resistor in parallel to the oscilloscope to show that peak to peak is 1V as expected.

[@alienozi]

Yeah that's the idea, impedance is basically resistance in this case

[@HiEveryoneChannel]

very clear and useful,thanks!

[@bblod4896]

Perfect explanation. Thanks.

[@Zhaymoor]

Can you please suggest me a low budget oscilloscope?

[@KeysightLabs]

From our lineup, the 1000 X-Series is a great scope! You can find more about it here: www.keysight.com/en/pcx-2759552/infiniivision-1000-x-series-oscilloscopes?cc=US&lc=eng

[@MissionFitnessCTC]

Good video!

[@rubenespada8117]

very very clear explanation...tnks for the info. very usefull

[@KeysightLabs]

Thanks!

[@Matthias051]

Thank you so much again for this helpful explanation...

[@ZsoltPinters]

Great video, as always, love your channel

[@KeysightLabs]

Thanks!

[@mikeoliver3254]

Great explanation.

[@javierpallalorden]

A very nice video, thanks!

[@silvestrenet]

Interesting info. Thanks.

[@raymondzhao9557]

good to know ! thanks!

[@nk-yj3ev]

*uses a 1,5 GHz oscilloscope*: "For the sake of time we avoid using 50 Ohm mode." Take care about reflections before switching to High-Z ;) ...from the first trap to the next one.

[@KeysightLabs]

It is definitely a trap, but worth a video all by itself!

[@alienozi]

The first time I used my scope to scope an analog circuit, the probe wasn't calibrated so the output just baffled me!

[@KeysightLabs]

I bet!

[@DoctorThe113]

Why the hell there is Korean auto-gen closed caption? Nevertheless the video is very good. Great explanation.

[@KeysightLabs]

It was translated by our team in Korea so they could use it over there, too :)

[@vaseemmehrancp9372]

Very informative

[@hafo821]

Thanks a lot!

[@NiamorH]

Well, you don't explain what HighZ mode is and why it behaves differently, so there is still half of the explanation missing.

[@uAslan53]

But why you would need such a mode

[@KeysightLabs]

50 ohm mode is very common with higher speed signals and systems. It allows for impedance matching schemes to get faster edges and cleaner signals.

[@beneemoon7431]

Useful for lots of RF applications. Transmission line and input impedance for lots of devices is 50Ω.

[@ElizatronicWarfare]

I have worked with university professors who do not (and refuse to) understand this principle, and just tell their students that the function generators are broken.

[@KeysightLabs]

It's a tricky one if you aren't aware of it!

[@glasslinger]

"I have my $5000 function generator and my $25,000 oscilloscope but the readings are all wrong!

[@uwezimmermann5427]

"for the sake of time"? - how difficult and time consuming is it to set your scope to 50 ohms? Give your audience another minute of your precious time and demonstrate this!

[@samtzam3774]

RTFM

[@cdrive5757]

If I were you I'd be more concerned about all that noise on your signal! What the hell is that?

[@blackturbine]

Thanks for the info only bad thing is I don't have the function generator I don't have oscilloscope nether But I will win one day damnit

[@bobbob-uv1oi]

Instructions unclear, connected oscilloscope to a un isolated test device. I no longer have a oscilloscope or test device. Ps dont do this this is a joke

[@KeysightLabs]

rip bob bob

[@ademch3580]

ok

[@fasulia67]

I actually understood that. The last 3 years at uni weren't a total waste I guess

[@RobeenaShepherd]

I know this isn't the best place to moan about this, but I just got a load of junk mail from RCRWireless News... did Keysight give them my email accidentally or on purpose? :(

[@Meiestrix]

Same here. They have apologized for the emails, but i am not sure how they got my email address.

[@KeysightLabs]

I don't know, I'll look into it! We sometimes do joint events (webcasts, etc.), so it's possibly related to that. I apologize for any annoyance!

[@BiecherBender]

I wondered about it too. I got 4 mails in the last few hours.

[@RobeenaShepherd]

Thank you nice people at Keysight!