#137 

Maybe you should share this video with your TA so that they'll know the right answer too!

On my Rigol, it's also Utility

This is a good question! Something I'll have to do a video on. Basically, to determine output impedance, you would observe the amplitude change as you change the load, and then calculate the output z.

page 40 of this manual: www.keysight.com/upload/cmc_upload/All/6C0633120A_USERSGUIDE_ENGLISH.pdf

I guess it would really depend on the particular signal generator (and how much voltage you're talking about). There is no universal yes/no answer.

The 50 ohm output impedance is used to provide a good source match to the line and load. Why 50 ohms, here's a good answer for that: www.microwaves101.com/encyclopedias/why-fifty-ohms

+Beretta96Dan The SG has to be loaded with a 50 ohm load in order to meet its specification. A 1X probe used with any scope will only give you less than 10MHz of bandwidth. Another problem you have is that the 196 only has 100MHz of bandwidth, which means that 100MHz signals will read 3dB down, and only if you use a 50 ohm thru termination and no probe.

Yes! I still have to work on the lighting, but for the first time out with the new cam, I think it came out ok.

***** The video is great, Alan!

Hey - there's no such thing as a stupid question (only stupid answers, and I'll try to avoid giving them). Assuming that your switch simply adjusts the readout on your generator, you can set it to High Impedance, and then calculate the resulting amplitude by multiplying the reading by RL/(RL+50).

+w2aew "Assuming that your switch simply adjusts the readout on your generator" -> Can we assume that some generators are able to short their 50 ohm load, then?

+fabts4 I have seen some where you can change the output impedance from 50 ohms to 600 ohms. I don't think I've ever seen one where you can change it to (near) zero.

The short answer to "why a 50 ohm termination is needed at the end of the coax" is to prevent signal reflections that can distort the signal. This is explained more fully in this video: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-g_jxh0Qe_FY.html For low frequencies - below a few MHz, it really isn't that important since the length of the line is much much shorter than the signal's wavelength, and reflections don't pose any problem. For for higher frequencies, or for low frequency signals with high frequency content (such as fast rise/fall times), termination may be important. The function generator has a 50 ohm resistor in series with it's output to terminate and reflected signals that are coming back from an unterminated end. This ensures that these reflected signals to re-reflect back towards the load again. Since the generator has a 50 ohm series resistance, and the expectation is that the far end load will be 50 ohms, this sets up a basic 2:1 voltage divider between the output amplifier of the generator and the far end load. Again, for low frequency work, you can omit the 50 ohm far end termination. Just keep this in mind when setting the desired output amplitude. If the generator doesn't have a way for you to tell it that what the far end load impedance is, then you'll just have to remember that the output will be twice the set value if the far end load impedance is high.

@@w2aew Thank you, I did watch the linked video and your explanation helps alot. I forgot about the in-depth content of that video. Sidenote, I actually frequent your videos on a different account which has more kid friendly subscriptions so the sidebar videos stay clean for when my kids and I view them. Take care.

The coax cable impedance really only begins to matter at high frequencies. Remember, the cable impedance is simply the impedance seen by the signal as it travels down the line (see my video on transmission lines and terminations). ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-g_jxh0Qe_FY.html

Dear w2aew, Thank you for your reply. I'm watching the link. Thank you for your nice tutorial again!

It is almost always best to terminate at the load.

Yes, excellent point. Anytime you get above a few MHz you have to start thing about transmission line effects, terminations, etc. I did a video on quarter wave lines as part of the video on tracking generators.

Yes. It's always a good idea to measure the signals at the device under test, just as it's a good idea to measure/check power supply voltages at the circuit and don't always trust the power supply display.

@Ralph Reilly I'm not sure why do you think I need to know, but it is one year old post.

50 would be closest - but remember, all this is doing is adjusting the calculation that the generator does for the output amplitude to account for the voltage division of the 50 ohm output impedance and the load. In either case, if you want a specific amplitude applied to your load, MEASURE it AT the load and adjust the generator accordingly (regardless of what the generator is telling you the amplitude is).

@@w2aew thank you for responding!

The omega symbol in the vertical indicator is typically use to indicate that this channel's 50 ohm termination is being used. I would imagine it is the same thing on the Rigol.

If the generator is set to Hi-Z, then it is expecting to see a high impedance load. Thus, the amplitude displayed on the generator would reflect that. But, since the load is actually 50 ohms, the actual voltage displayed on the oscilloscope would be 1/2 of the value displayed on the generator.

Thank you!

You need something that can show/calculate the true-RMS value - because if the waveform is not sinusoidal, then the 0.707x factor is meaningless.

Usually not - in most cases, you'd want to AC-couple a test signal that you're injecting to a circuit.

Thank you sir

To answer your questions: 1) The termination provides a "matched load" for the transmission line. To understand why, it will help to understand how transmission lines work (and why they have a characteristic impedance). This video should help with your understanding of that: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-g_jxh0Qe_FY.html As you will learn from this video, a 50 ohm transmission line is NOT the same as a 50 ohm resistor. It only "looks like" a 50 ohm resistor to signals that are *changing* - so, crudely you could think of it as a AC-coupled 50 ohms. Again, the video should help clear this up... 2) When you change the function generator to high-Z, this does *NOT* change the output impedance of the generator. Let me say that again... it does *NOT* change the output impedance of the generator. This is a VERY common mis-understanding. The output impedance of the generator is *ALWAYS* 50 ohms. When you change the Output *load* setting on the generator, all this does is causes the generator user interface to change the output amplitude appropriately. Since the output impedance is 50 ohms, there is a 50 ohm resistor in series with the output at all times. So, the voltage at the load will depend on the load resistance because this load resistance forms a voltage divider with the generator's internal 50 ohm resistor. The *load* setting on the generator tells it how to calculate the voltage appearing at the load by properly calculating the effect of the voltage divider (50 || load_resistance).

@@w2aew I see now. Really liked your transmission line video!

If you draw a schematic of your setup, including a "model" for the generator output and scope input, it'll be clear to see the problem with the common ground. A bridge rectifier expects an AC input at the input, where the input signal on each side swings above and below ground, while the output of the bridge typically has one side tied to ground. But, when you drive the input of the bridge with the function generator, that effectively ties one side of the input of the bridge to ground - so that node will no longer swing below ground (because it IS ground). A simply audio transformer can be used between the function generator and bridge rectifier input to solve this issue for your measurement purposes.

You can generally use a 50 ohm terminator on a tee instead of a through terminator for HF frequencies and below, because the stub length will be insignificant. However, don't forget that even low frequency square waves and digital signals will have frequency components in the very high frequency range due to the fast rising and falling edges. If your signal's rise/fall times approach 1ns or less, then you'll start to see distortions due to using the tee instead of a through termination.

The SIGLENT SDG1032X is probably a good buy.

It's most commonly used as a "source terminator" when the source has a low output impedance.

When the source impedance is lower than say 50 ohm. So it acts like the 50 ohm resistance in your schematic here.

This simply means that the reading assumes a high load impedance. I bet if you connect a 50 ohm load, the the actual output would be 1/2 the setting.

Thanks for such informative videos... Sounds like you have a bad cold.... Hope ya feel better...

You're dB calculations are off, but you have the idea correct. The generator has to produce twice the raw voltage (into 100 ohms) in order to achieve the desired voltage into 50 ohms. Remember that power is calculated from RMS voltage, not peak. Considering sinewaves only: 2Vpeak = 1.414Vrms, 20mW (+13dBm) across 100ohms, and 1Vpeak = 0.707Vrms, 10mW (+10dBm) across 50ohms.

The 50 ohm resistor that I connect to the scope is *in parallel* with the scope's 1Mohm input, and since it is so much lower than 1Mohm, the result is very nearly equal to 50 ohms. This 50 ohm combination then appears *in series* with the 50 ohm resistor that is inside the generator (in series with the output amplifier). Then end result is that the function generator's 50 ohm output resistance is in series with the 50 ohm termination at the scope, thus making a 2:1 voltage divider.

@@w2aew this is what I was able to figure out myself. My problem is that the schematic doesn't look like a parallel. Could you explain how it is a parallel connection with the scope? It looks like a serial connection to me.

@@Infinitesap The 1Mohm input impedance of the scope is not drawn explicitly. The 50 ohm thru terminator shown at about 3:10 in the video has a 50 ohm resistor from the center of the coax to the shield (ground). Thus, when connected to the scope input, this resistor is in parallel with the scope's 1M input impedance (since the scope will now be measuring across this 50 ohm resistor.)

@@w2aew thank that made it clear for me. Please make more videos. I'm really learning a lot from you 🙂.

Thanks this really made me understand. Could you make a video on how to use the sync and trigger in- and outputs on the fg/scope. Would also be awesome if you could show application with them on breadboard setups.

Yes, a voltage divider, precisely. The generator uses the output-Z setting to properly compute the division so it can accurately show you the amplitude. It's always a good idea to measure/verify the output with a scope, etc.

@@w2aew thanks as always Alan. Happy new year. Enjoy some good propogation

Does the generator have a setting where you can input the load impedance? If not, then the best practice would be to use a load impedance that matches the coax impedance (usually 50 ohms), and then always measure/verify the output amplitude with your scope.

Would I be doing the right thing using a variable terminator? Thank you

It would be best to have the termination match the coax impedance (i.e. 50 ohms) to prevent any reflections.

What I am trying to accomplish with the variable terminator is adjusting it so that the signal generator Peak-Peak voltage match the scope Peak -Peak readings. If I use a 50 ohm terminator on the scope, the reading are not the same however, if I use my variable terminator set to about 37 Ohms, the readings match each other. I tried this on both ends of the 50 ohm coax and the readings match the same. I'm thinking maybe the signal generator voltage display reading is off. I realize that matching the terminator to the coax is the best thing to do. I guess I should dis-regard the voltage reading from the signal generator and make adjustment according to the scope is my thinking while using a proper 50 ohm terminator at the scope input. Oh! buy the way, the signal generator is a GW Instek SFG-1013 which of course is a entry level signal generator and not a lab grade device. It's frequency and voltage output is rock solid and good enough for what I need. Thank you for any reply.........

If it appears that the amplitude is "off" by the same factor at all frequencies, etc., maybe you can make a modification (insert a series resistor) inside of the generator to make the voltage division more accurate.

You also have to take into account any DC offset. But, in general, the RMS value of a 50% duty cycle square wave is equal to 1/2 the peak to peak value. This video might help: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-ue0wtlrmCJE.html

Thanks Allan! Just discovered your amazing channel!! Am plowing through them at a good clip. Playing some multiple times. Playing and pausing others, and then running over to my scope. This is an amazing resource. Thanks so much!!!!

You may know that I maintain a complete index of all my videos, arranged by number and by topic. The link can be found on my channel's homepage, in the lower right corner of the banner graphic.

For low frequencies (below a few MHz), it can be either place. Above a few MHz, it's best to terminate at the end of the transmission line.

It will all depend on the impedance presented by the load (as well as your measurement setup when connected to the load). Be especially careful of coaxial connections, proper terminations and coupling techniques between the generator, the load, and the measuring device (scope, etc.).

who do i know what is the internal resistance of function generator if i don't have user manual , further for dynamic load how do i match my load resistance with internal resistance of function generator.

In the vast majority of cases, the internal resistance is 50 ohms.

The TTL output designation on the signal generator means that this output is designed to be compatible with TTL logic signal levels.

Yes, 'load impedance' changes the way the *displayed* amplitude is calculated. If your generator is expecting a 50 ohm load, you can use a 50ohm thru terminator, or simply account for the 2x voltage difference.

@@w2aew Thank you so much.

Both - it is an AC signal with a user adjustable DC offset

Yes, but that is the "surge" impedance. In other words, it appears like 50 ohms to rapidly changing signals, due to the distributed capacitance and inductance. At low frequencies, or where the line length is very short with respect to the wavelength of the signal, the coax simply looks like a pair of conductors.

@@w2aew Thanks this solved a mystery for me :-)

Signal generators almost always have a 50ohm output impedance - THIS is what limits the current. If it is designed to deliver 1Vrms to a 50 ohm load, that means that is generating 2Vrms at the gen side of the internal 50ohm resistor in series with the output. Thus, the voltage delivered to the 8 ohm load would be 2V *(8/58) = 276mVrms.

@@w2aew many kind thanks.. now i know my stuff works... what do u suggest,, to increase current output..

@@kaybhee6 If you need to drive such low impedance loads, like loudspeakers, etc., then you'll likely want to stick and audio power amplifier between your siggen output and the load. There really isn't anything you can do with the siggen since the output impedance is generally fixed.