Connecting 50 ohm signal generator to non-50 ohm device

[@johnbellas490]

Tony This video was very helpful !! I am a Ham radio operator for years now and have slowly been setting up my repair lab/station with signal generators, oscilloscopes, spectrum analyzers, sweep signal generators, VTVM's and digital VOM's all old and some rather new, used equipment!! This video made what I thought in the back of my mind VERY CLEAR to me!!I would like a few more of these types of videos that explain things like this in a semi lab type setup with demonstrations like you had set up here !!! Many thanks ----- John Bellas KC2UVN

[@patrickoconnell3896]

Thanks for the video. The tuner side of the stereo is still my weakness and your videos help a lot. Please keep them coming.

[@caspianwendell]

Keep videos like this coming, you are the best teacher on the tube. I am an amateur myself and need all the help I can get to help pass the time in my retirement age

[@tiagomello]

One of the best explanations I ever had for this topic.

[@janeklof2769]

A thousand thanks to your very good and easy understanding of impedance - matching ! Well done !

[@sirsuse]

Great video! Congrats on 10,000! You've earned it.

[@robharley9838]

Thanks for diving in to a somewhat mysterious subject I've long "scratched my head" over. Information on Impedance matching of equipment is not well summarized anywhere that I can find, so this shed lots of light on the issue! I'm a new subscriber and love your videos, thanks for sharing.

[@georgethomas9436]

Great explanation. Well worth watching more than once. Congratulations again on the 10,000 subscribers. I think I would go with building the impedance boxes.

[@mikemix1361]

Impedance matching is a great topic. Congrats on 10k! Keep up the good work.

[@srtamplification]

The Keysight generator that I use has a 50ohm, High Z and custom output setting. Makes things real nice. Great video!!! I don't know how I missed this one.

[@srtamplification]

Thanks for the video. It reminded me of the 75ohm and 300ohm matching tranformers we had to use on TVs back in the old days to play Atari games on.

[@jamesmdeluca]

Greetings: Your 50 ohm terminator may have the same cause of intermittency as some terminators I have purchased. They were made with 50 (or 75) ohm resistors crimped to the BNC center pin but not firmly attached to the BNC shield; the opposite end of the resistor may have merely been folded over and stuffed inside the end cap rather than having that lead crimped between the end cap and the pertruding portion of the connector that is usually crimped to the shield of a coaxial Cable; no kidding! 0 dbm is 1 milliwatt into 600 ohms. Originally from Bell Telephone system equipment.

[@davidhamm5626]

Very well explained. I have a signal generator, but I am not sure of how to use it. This series , as it goes on, will help me, I think !

[@joshshine2489]

Very helpful Tony....thank you and yeah, I'd like to see more

[@alfredneumann4692]

Good explanation! In the 80s i had to build match-parts. It's a hell of measurement with a very good resistance meter (wayne-kerr) and one million metall-resistors (1%). Cheerio from Germany.

[@radio655]

Nicely explained! The commercial 6db pads to my knowledge contain a pi (or t) network, so three resistors instead of two. This helps to ensure that the generator sees 50 ohm and the load sees 75 ohm.

[@taldmd]

Great video! Worth explaining also, for a future sequel, that also cables and connectors are rated to a certain impedance, there are even variations of BNC 75 and 50 that have physical differences. And the consequences of impedance mismatching such as reflections, and why universally chosen impedance of 50 omh is that value and ... man, this is a huge topic.

[@W1RMD]

Great video! I especially like the notes you put on the page. They save a lot of time by not having to stop and explain. If you haven't already, I would like to see a video doing this only using ferrite toriod transformers.

[@TJET737]

Tony, my first comment to you ever since I've been watching your videos (from the beginning). 1st off, congratulations on your 10K subscriptions. I think you are obtaining such an audience because of the way you present things and the content provided. I've learned a lot over the last couple years watching your videos. You have a unique style in your presentations, provide excellent examples of what your talking about and deliver the content on a level we can all follow. Keep up the good work!! I'm sure it is very time consuming to compile and edit these videos. I appreciate the time (I'm sure everyone does) you put into sharing your hobby. Thank you again!!!

[@allthegearnoidea6752]

I’m not surprised you have a good number of subscribers as you have excellent technical contents thanks for sharing it. I still think feeding none 50 ohm is the stuff of nightmares especially if trying to do accurate sensitive measurement. Thanks regards Chris

[@TheRadioShop]

Tony, You did very well on this one. Thanks

[@davidknight3643]

Hi Tony yes I found this very helpful please continue with these tutorials.

[@mfr58]

Nice one Tony. Reminded me of stuff I learned too long ago!

[@robertcalkjr.8325]

Thanks Tony. Congratulations on 10K! I never have all that I need...

[@timka880057]

Hey Tony, hope you had a Merry Christmas and a Happy New Year. Yes, I'd like to hear more about this and into the simplicity like you're doing. Mathematics is another thing I'm still trying to get a hold of. I used to know some of the formulas, like in mechanics and electronics by heart. I watch other videos on KZbin about calculating Impedance using vectors, leading and lagging voltage, current and reactance vs DC resistance and I'm getting back into understanding it better, it gets crazy when there's parallel and series caps and coils. The easiest part of course is DC resistance, volts, current and wattage. I learned the product divided by the sum in parallel DC resistances a long time ago but the reactance and impedance gave me quite some confusion back in the day when I was learning in the late 80's, but it became clearer as years have gone by. But I still have to touch up on my math. I had to go a different way where I could make ends meet because servicing stereos, mixer boards, and VCR s didn't quite pay the bills..especially when they'd end up buying a new tv or VCR. I should've went back to school and got into"" X-RAYs lol"" and robotics. But I didn't have the time or the resources at that time. Please do keep up the information, I've got a small shop I work in and since I'm disabled I've gotten into the hobby of restoring vintage stereos, radios, also guitar amplifiers, synthesizers and other musical equipment. I love it, electronics has always been a passion of mine... so is music since I'm a musician. Enjoyed the video thanks Tony!! :)

[@The1jonnyz]

Excellent video. Thanks for taking the time to make it!

[@kardeef33317]

I would like to learn more about this. And thx for the video.

[@capriracer351]

Great quality videos. For me in addition to the superb quality, I believe that you may actually live in a reasonably close proximity to myself, so that provides some interest as well since I did not realize that there were others close by doing this sort of thing. Keep making the interesting videos.

[@barryg41]

I like it! Thanks for the video. Happy 10K!

[@NickGaster]

Well done Sir, Thank You!

[@GeorgeEI7KO]

Great video Tony. Could you do a video on your 8 ohm dummy load please. Love your channel.

[@microreniassance2929]

Very good video ! Very helpful!

[@IZ0MTW]

Very useful!! I had to figure it out myself when I bought my first sig gen (Marconi 2022E). I’m glad to see that what my thoughts were are confirmed know by someone :) Btw my sig gen and I saw yours too can display the emf if you select it. Cheers

[@JelckedeBoer]

Great video and channel overall!

[@johnwaldburger8822]

Great explanation... more like this video please.

[@aduedc]

You can use transformers to change impedance for broadband impedance matching. Also, you could use PI restrictive network for wide band matching.

[@lulrik1]

Merci, very good and instructive !

[@steveott1884]

Thumbs up. More test equipment videos pls. Thx.

[@gregnewberry4813]

Excellent video. I'd like to see a video on doing a sweep IF alignment on a receiver if you don't have a real sweep gen. I have the same HP signal gen as you, plus a scope and function generator. Thanks.

[@Washburn-rr5eh]

I'm glad I found this video. Now I understand impedance matching a little better now. On your signal generator I notice you use the DBm setting when aligning radios. Would you mind explaining why you use that. I have the 8656B HP generator and I'm just trying to understand my equipment a little better. I would like to know more on how the various settings work on the generators. Especially the DBm settings. Sometimes I see -40 DBm, -70DBm or +10DBm and why you use those settings.

[@xraytonyb]

That's kind of a two part question: dBm stands for decibels with reference to 1 milliwatt. The decibel is a way to represent a level or reference on a logarithmic scale. 0dBm would equal 1 milliwatt. The formula for dBm = 10 x Log^10(milliwatts). When you learn the dBm scale, it makes it easier to calculate a level. For instance, every time you increase a signal by 10 dBm, you are making the signal 10 times bigger. So 0dBm is 1 milliwatt, 10 dBm is 10 milliwatts, 20dBm is 100 milliwatts, 30 dBm is 1000 milliwatts (or 1 watt), etc. Additionally, 3dBm is an increase by a factor of 2, 6 dBm is an increase of a factor of 4. As for what level to use, the signal level is determined by the circuit you are testing. The antenna input of a tuner will only require a tiny signal (sometimes measured in dBf, or dB referenced to 1 femtowatt), while the IF section will require higher signal levels in the -dBm range.

[@ot7e]

Great video very well explained, I like it! Congratulations on the 10,000 subscribers. 73

[@mr.salisbury2435]

this was great and cleared up a lot for me

[@scottgilbert7927]

Great video, but do we know that the SG is set to put out double the voltage reading it displays? I think that the 50 ohm output is set by a reactive network, similar to your LC impedance matching network that you showed, not a resistor, so you would not have a true voltage divider network. You have to look at the schematic of the SG to confirm that. Like you pointed out, the LC network like the matching network you built does not dissipate heat, but creates a non-reactive output impedance. The output impedance network for the SG is most likely an LC network like a filter circuit that has a 50 ohm output impedance and acts like a 50 ohm resistor because the output voltage and current are in phase if there is a 50 ohm termination, so it looks like there is a 50 ohm resistor at the output of the device. But the output network is not dissipating power like an ordinary resistor so it is not dropping voltage due to power dissipation as a resistor would. I think the real explanation, or at least another way to look at what is going on is that if you connect the 50 ohm cable directly to the 1 megaohm input of the scope, you get a total reflection of the voltage and current waves at the connector, which is why you measure double the voltage you would expect, because you have a reflection coefficient of 1 at the load end of the cable. The forward and reflected voltage waves having 50 mV magnitude each add together in phase to equal 100 mV across the 1 megaohm internal termination of the scope. The 1 megaohm looks essentially like an open circuit at the load end of the coax cable, which would cause effectively a 100% reflection of the incident voltage. If you place a load resistor termination having a value much closer to 50 ohms in parallel with the 1 Mohm, you will see what looks like a voltage divider effect between the 50 ohm transmission line and the termination resistor, but is really a reduction of the reflection coefficient at the load. A 50 ohm termination would yield a zero reflection coefficient, so the voltage across the 50 ohm termination would be 50 mV as expected. When you calculate the reflection coefficients based on the termination resistor values other than 50 ohms, you measure the exact same voltages as if you had a purely resistive voltage divider based on 50 ohms and the termination resistance. While I am not a design engineer for test equipment, I know that radio transmitters use LC networks or inductive baluns to create the 50 ohm output impedance, which do not dissipate power other than for losses. But it doesn't seem likely that the SG would be designed to output twice the voltage that its readout displays on the assumption that the signal should be fed into a 50 ohm load. However it is important that the user of the SG is made aware that the voltage at the load depends on the load resistance. Also, I believe the commercial attenuators you showed use a resistive pad consisting of three resistors in a Pi configuration, not a simple two-resistor voltage divider network because you can't get the desired impedance transformation easily with those. The resistor values in the pad can be chosen to give the desired input and output impedances, but of course with attenuation of power as you said. The advantage there is that the resistive pads give the same impedance values independent of frequency (up to 3 GHz anyway). So thanks for coming up with the non-dissipative L network based on the inductance and and the variable cap How you built that impedance match would be worth a video in itself.

[@mrkattm]

Congrats on 10K,, reading through some of the comments I think there is still some confusion. Perhaps if you went over the maximum power transfer theorem and the voltage divider equation folks might see that points you were trying to convey.

[@moshet842]

You can use a series of progressive value inductors to deal with different parts of your radio spectrum and you can get it to a point where it remains consistent throughout the desired band.

[@gearheadted9210]

excellent video,thanks!

[@bigmac4948]

very helpful video. thank you!

[@ptronix]

Great video as always, would be interested if you could do a video about your owon spectrum analyzer, I have the siglent which I love, so would be interesting to see how they compare

[@CameraNut1000]

Good video. Thanks.

[@davidkclayton]

I would be curious to know how it compares to a matching Transformer circuit

[@robertdodge8587]

Thank you all of the tech stuff is fascinating, could you publish some diagrams and tech notes? Thanks.

[@xraytonyb]

I have a drive posted on my Patreon Channel with some of the info you see on the videos.

[@georgethomas9436]

I've watched this before and like revisiting it.

[@sincerelyyours7538]

Thanks Tony! Impedance matching for me has been a mystery since my college days. Intellectually I understand it, but achieving it is more difficult than it sounds, especially in one-off home designed circuits. Electronics profs tend to think that their students automatically understand this concept but trust me, we don't, at least I didn't. Still a mystery for me is how to calculate or measure the impedances of different areas of a circuit so all the various stages work best in concert with each other and not against each other. Can you elaborate on that a little bit for us?

[@warplanner8852]

The voltage on the scope measures Vpp which is 2.828 times Vrms. So then there's that as well.

[@jw228w]

trying to understand impedance matching. if you are outputting 50mv from signal gen,but scope is showing 100mv,isnt that more efficient?or does that 100mv in a load like a radio receiver create more problems like heat? and another question. what is the history/reason that the electronics industry settled on 50 ohms in the first place thanks john

[@Slartibartfas042]

@jw228w: Hi John - Pretty much complicated question asked very easily. (But definitely not to be answered as easy to understand!) All that mess up was done because you have 2 different purposes of signal sources. There is the other section that just has to deal with very tiny amounts of power and has to transport the smallest amount of power most efficiently and the cheapest way they could from the receiving Antenna to the receiver's electronic parts. And there you have the 75 Ohms subsection of RF Industry - the Consumer electronics industry of Radio and TV receivers. And finally there is one source that wants to transmit as much power as could be transferred with reasonable voltages and currents over cables, very often Coax-cables. That can be seen on any kind of transmitter like CB-Radio, Ham-Radio, Mobile-phone output, TV and Radio stations at their broadcasting side. They need to transmit as many *power* as possible with least amount of technical effort. That is the 50 Ohms subsection of RF Industry. Why were these different load impedances chosen for the 2 different purposes? Case 1 (75 Ohms): Dealing with very little power, so neither voltage nor current the connections have to handle are any problem at all. So just take that impedance where you can produce cables and transmission lines as cheap and efficient as ever possible. Here is 75 Ohms advantage. You get most efficient transport of your signal from the Antenna to your Receiver with the least losses and relatively low amount of materials to use for. Now on the other side you have Case 2 with 50 Ohms line impedance: Here we have all devices that have to *transmit* signal into the air. Starting with TV/ Radio transmitting stations, CB- or HAM Radio stations, Mobile Phones, your WiFi station and so on. Why? Think of the transmitted power. You want to transmit some decent amount of power as that is the measurement for "how good receiving stations could hear you". The more power you get into your antenna that will radiate it the more signal will arrive at the receivers! So, you want to transmit for example 100 W into your Antenna. As the transmitted Power "P" is P=V²/R (Voltage V squared divided by Impedance) and therefore the Voltage your cable and the electronics will have to deal with is : V=SQRT(P * R) (Square-Root of Power times the Impedance). This makes for the relatively small power of 100 Watts a Voltage of about 70 Volts for 50 Ohm system whereas with 75 Ohms it would be 86 Volts. Now think of a transmitter that has 1000 W - that would be a voltage of about 274 Volts! (SQRT of 75000) Finally: Is 50 Ohms the optimum for transmitting signals? NO! The best for pure transmitting systems in theory would be at round about 30 Ohms (for example according to mosers-on-tour.net/hb9lcd/kw-antennenzubehoer/koaxialkabel/ who relate to various sources on the internet). But.... what about the resistance of maybe 0.5 Ohms for each connector on the path from your transmitter along to your Antenna? You have at least 2 of them, making 1 Ohm of Resistance along. That's 1/30th of the power...(would mean with 30 Ohm system you would burn 33.333 Watts in 2 Connectors/Plugs!!!) Even if we just assume it would be Resistance of about 0.1 Ohm per Connection that would make a pretty much "hot -plug" system! ;-) And, BTW: Same statement is true for the pure Resistance of the cable wire itself (I am meaning the pure DC Resistance of the Copper in the Cable. That would sonsume the power as well and put it into heat)! Every 0.5 Ohms you would burn another 33.333 Watts of your pretty 1000 Watts total output power of the transmitter. *Summary: Circuits having to deal with no high RF powers are using Coaxial cables with 75 Ohms Line impedance for almost best signal transmission to the receiver, everything that has to be capable of delivering some RF power (and even 1 Milliwatt has to be considered as such!) usually is using a compromise inbetween low loss and voltages that can be very well dealt with, and that compromise is 50 Ohms* Line impedance.

[@donaldchapman8392]

John, The reading on the SG assumes that there is an impedance matched output - in other words, the manufacture displays the output voltage of 50 mv assuming you are connected to a 50 ohm load. The ACTUAL output of the SG is 100 mv - and if you have it connected to a 50 ohm load that voltage gets divided equally by the INTERNAL 50 ohm resistor in the SG and the 50 ohm load leaving 50 mv across each load. When you connect it to a scope directly that voltage divider that was 2 - 50 ohm loads in series becomes a 50 ohm (internal SG resistor) and the 1 MEG ohm input impedance of the scope in series - most of the voltage gets dropped across the 1 meg ohm load hence it now shows the full output voltage of 100 mv.

[@erikdenhouter]

I want to add a note on what is written here, and what is said (or not said) in the video. What is forgotten is explaining the reason WHY impedance matching is important: efficiency of power transport (power exchange). Since it is impossible to add a graph here, I will use a few lines of calculations to show that the exchange of power is at its highest when internal and external impedances match. P = I^2 x R (power = current squared times resistance) I = V/R (Current = voltage divided by total resistance (R-internal and R-load are in series)) Put these two together, and get: Power in the load resistor (Pl) = (V/(Ri + Rl))^2 x Rl Assume internal resistance = 50 Ohm and internal voltage source SG = 5V Next you will see that if we vary the load (15, 35, 50, 70, and 100 Ohm), the power in the load resistor will peak with 50 Ohm: Pl, power in the load resistor: (5/(50+15))^2 x 15 = 0.089 W (5/(50+35))^2 x 35 = 0.121 W (5/(50+50))^2 x 50 = 0.125 W < peak (5/(50+70))^2 x 70 = 0.122 W (5/(50+100))^2 x 100 = 0.111 W From here it is better to see that when only a 1 MegOhm scope is attached, you might get the full voltage from the source, and you might conclude that that is better, but that 1 MOhm will course only a very tiny current, so no power of any significance.

[@dhpbear2]

22:20 - Hmm. My guess was a bit oversimplified: just a 25-ohm resistor from one connector to the other. This L-network looks totally dependent on frequency.

[@radio655]

To avoid reflection effects, one would need to present a 50 ohm load to the generator and a 75 ohm source to the load. The best way to achieve that is via a pi or t shaped network. That is what normally in the commercial pads. The l network is very popular with ham radio operators to match very high impedance end fed antennas to low impedance transmitters without too much loss.

[@SeanBZA]

@@radio655 Added bonus with L network is that it is automatically a low pass filter, so that harmonics are reduced as well.

[@user-su5sq5ib3i]

Good video thanks

[@gerardogennai4486]

Hi, the explanation of the BNC T and the 50 Ohm termination is fine, but it uses a 1MHz wave. Is it the same with higher frequencies ? Let´s say 50MHz, 300MHz or we have to take in account other efects ?

[@jspencerg]

Thanks!

[@johnny7254]

The oscilloscope IS showing the correct voltage -- at 4:44 in the lower left corner of the o'scope's screen it says "50.0 mv

[@BEripsta]

Nice video! But couldn't you just use an OpAmp as a voltage follower? Or is an OpAmp just not suited for these higher frequencies?

[@bahagursoy8138]

Dear Sir I have question about 50-75 ohm impedance cırcuit. If we use a this maching circuit should we connect a load impedance output of 75 ohm side or O scope input in paralel. Also what is the copper wire of dia. Thanks for you Baha

[@bharathkumarkrishnamurthy7576]

Very nice explanation..it is really helpful. I have a SG with 600 ohm o/p impedence and peak voltmeter with 600 ohm i/p impedence, when I source 1 v(pk-pk), my volt meter measures 2 v (pk-pk)..Any reason behind this?

[@gregwmanning]

Gday Tony, Is there a physical/mechanical difference between 50 & 75ohm BNC adaptors (BNC to SMA or BNC to N-Type etc)? If they are different how can you tell them apart and can you interconnect them? Thanks Greg

[@xraytonyb]

The short answer is yes, 50 and 75 ohm connectors are different, but they will still mate to one another without damage, however, they may not connect securely, causing noise in the signal. That said, the 75 Ohm BNC plug does not have extended dielectric around its outer spring fingers. The air has a higher impedance constant than solid dielectric material. Also, the 75 Ohm plug’s center pin has the same diameter in the rear area (where the pin is crimped) as in the front mating interface area. The 50 Ohm BNC plug’s center pin has a thicker diameter in the rear area where it is crimped. Furthermore, the 75 Ohm connector’s dielectric is made of Teflon because of its higher impedance properties than Delrin, which is typically used in 50 Ohm connectors.

[@KC2QMA]

Any chance you could share the spreadsheet file? Great videos!

[@phyzxengrmoore6928]

Liked and subscribed!

[@blackarrow8683]

What kind/name of coax cable are you using?

[@baghdadiabdellatif1581]

Thank you

[@coldfinger459sub0]

Thanks for the video that really explains some things I only read about but did not have a full understanding of. Right now I’m using my cell phone as a signal generator and I have a old 1960s or early 70s BK signal generator and my junior college teacher is going to give me one of their old ancient nonworking RS sweep generator. Just the caps need to be changed and do a calibration and I’ll have a freeRF generator

[@Electrolab28Ag]

I'm trying to understand this but having trouble, I'd like to learn more. However, If the SG is set at 50mv and you get 100 at the scope... Can you not set the SG to 25mv and get the 50mv at the Scope? Don't know and that's why i'm asking. and trying to learn.

[@xraytonyb]

Without a load attached to the SG output, you will always read about 2 times the amplitude displayed on the Signal Generator. This is because you are essentially reading an open circuit, as the scope has a high input impedance. Since the scope is very high resistance and the internal resistor is only 50 ohms, most of the voltage will be across the scope and very little will be across the 50 ohm internal resistor. When you place a 50 ohm load across the output of the signal generator, it creates a voltage divider between the internal 50 ohm resistor and the external 50 ohm load. The voltage actually doesn't cut in half, its just that half will be applied across the internal resistor and the other half will be across the load resistor. The scope is only reading the voltage applied across the load resistor and not the internal resistor. This is called a voltage divider. Hope this helps.

[@microreniassance2929]

@@xraytonyb I think what may be the answer he's looking for here is if you want 50mV into your DUT, don't be misled by the scope with out a 50ohm termination. If 50 ohm is the DUT and the SG, you're good to go despite the scope read. If for some reason you only want to read 50 on your scope, just dial down the SG to get there. But the signal will not be 50 going into a 50ohm DUT.

[@btouw8558]

I think these things are better explaned on a whiteboard

[@cottydry]

Isn't there a simpler method? Ignore the SG's readout and dial down the amplitude until the actual output measured at the DUT is what you want.

[@old64goat]

I AGREE

[@moshet842]

I thought those T junctions by default correct the problem without having to put a 50 ohm termination on the other side of it. At least that is what I was told.

[@davidausterman5915]

They have no "impedance" built into them so you still need a termination load- unless they are some specially built units, but I have never seen anything like that (which means nothing lol)

[@PackardDog]

Fantastic. Thank you for the video!! Do you happen to have a part# and/or source for the 75 ohm diy box.

[@xraytonyb]

I purchased the box from Antique Electronics Supply ( tubesandmore.com ) . The coil is home made from a piece of #12 solid wire and the capacitors are just a ceramic cap and trimmer.

[@PackardDog]

@@xraytonyb Thanks! I appreciate it!

[@daveturner5305]

MORE More more - eventually it won't be as clear as mud.

[@poduck2]

My fluke 6060A/AN will automatically convert dbm to mv, so that math is easier.

[@NICK-uy3nl]

Why not just adjust you generator output to match the voltage you want by working your formula backward !

[@maximacom-BelgradeSerbia]

very useful video.thank you !