Thank You! You just saved me from trying to fix a local oscillator circuit that wasn't broke due to a low level reading on my scope with 1x probe at 36 mhz. I switched to x10 and there was my correct level! I really enjoy your videos.
1X probes are not useless, they are very good for low voltage measurements, in order to keep the signal-to-noise ration as high as possible. Great video, though!
+electrocomm What he said was switchable x10 probes are useless as x1. You effectively have a hidden 470Ω resistor in series with your probe, and as shown your 'low voltage signal' will sink into the mud of noise @ 10MHz so it's useless for low voltage work as well ... more so.
+Peter Walker Agree, you might be right, the key word is "switchable", however, unless dedicated probes, these switchable 1X probes may not be quite useless; in a low BW (
excellent explanation, spot on. I had this explanation as a baby ship's radio officer in the late 80's in those days our radio and even some radar kit was lower then your average CPU FSB. - times change. This should really be compulsory stuff for new engineers yet as you say it's rarely mentioned, almost forgotten knowledge - tell us more Mr von Daniken !
Thank you Dave for this excellent video! I love all your videos but it is nice to pick things up like this which weren't discussed in the university (unfortunately). It may take some work but I'd hope the appreciation from the community encourages more in the future! Thanks!
They're great for me, because I'm just doing audio (20-20k Hz) right now using an old analog 30MHz scope. I do want to do radio at some point though, so a 1GHz scope and probes would be nice to have when I start with it.
I loved this video. I know a fair bit of theory and thought we were going to get into rise time limitations of the R/C compensation. I NEVER knew about the deliberately lossy transmission line. Dave, please do more theory vids.
Dave, One missing element in your trasmission line analysis is INDUCTANCE. Impedance (Z) is equal to the square root of the ratio of L /C. The whole problem of BW in your case is related to matching (optimal transfer of power) the input and output Z's to the trasmission line. The 9 Mohms helps expanding the BW but introduces resistive losses, hence the X10. A 50 ohms coax will not help without proper matching or termination. I think it is time to review transmission line theory. Cheers!
Brilliant explanation Dave. It would be great to continue this by explaining differential probes, how they work and possibly design one to build. Low frequency of course.
Very grateful for this video seeing as I'm currently playing with building an 80MHz RF front-end and was wondering how best to model probes for LTSpice to understand what voltages the ADC will be seeing. Nice one Dave!
Sometimes the route is much more enjoyable than the destination!!!! I'll not take for granted the X10 scale and will think thrice before I use the X1!!! Thanks so much!
Thank you for answering this. Had trouble finding this anywhere else. Didn't realize a coax had so much capacitance. This threw me off on a high impedance measurement when trying to use 1x mode, now I know why.
The x10 probe reduces the signal amplitude by ten times. This forces the scope to amplify the input signal by ten to get back to the original scaling (which means it will also amplify any input noise by ten as well). So the x1 probe is more sensitive and can capture low level signals with better resolution and less noise.
Another reason why x1 probes are designed that way (internal lossy transmission line) is protection of the sensitive input circuitry of the oscilloscope. A direct coax cable surely has the best bandwidth, but it will also allow anything to pass, including signal under test as well as any static HV peaks which might easily damage the ESD devices inside the scope. The design of ultra-high bandwidth/high gain voltage amplifiers combined with effective protection circuitry is a form of art in the electronics world.
I would say that everything is little simpler, since i had to design my own cable for oscilloscope probe. Most important feature of lossy cable is to have very low capacitance, so wire must be as thin as possible. If you try to replace it with regular coax you still can compensate everything for any bandwidth but input impedance will be terrible, what makes measurements impossible( done that and was puzzled why everything stops working when i attach my probe) probe impedance is basically cable impedance divided by multiplier. lossy cable allows you to achieve higher input impedance resistance does not seem to play significant role however it ensures that there will be no reflection because of improper termination and also it may improve impedance by turning cable itself into atenuator
Interesting to see how the pure truth not earns many thumbs ups. I add some other (probably even less popular) fact: the other reason of the high wire resistance is the material choice. The wire must be strong and flexible despite of the very small diameter, and also not soluble in tin during soldering, this is why it cannot be made from copper, but other materials with higher specific resistivity must be used.
100x is also useful for low voltage sensitive ccts, e.g. 32K xtal oscillators and micropower /low leakage stuff. For low voltage, low frequency use, a 90M resistor in series with the top of a 10x probe is a cheap option.
No, it's all to do with transmission line matching and compensation. The probe is transmission line optimised for x10 mode, so is all mis-matched when you switch to x1.
I've been using scopes for twenty years and never knew this. I was taught to always use 10x because of the higher bandwidth, and noticed it on the datasheet, but never needed to take the time to figure out WHY.
Electricians tape tends to fix that problem: set it, tape it, forget it. Also that strip of red tape is an easy way to know which one of the probes in the box is set to 1X
Hi Dave you never said anything about the 50 ohm resistor of the signal gen. That should be included in your calcs in order to be more precise on the -3dB point. I have seen a lot of your videos. As always they were extremely helpful. Thanks again.
Back for a "2nd suck of the sav". Good stuff! Ok, I'm a slow learner. Re: wiggle wiggle center conductor. I was told back in the day (elect shop) that the wiggle was to help [center] the conductor in the dialectic more consistently. On some coax's they will include a plastic/teflon? spiral to do the same thing. That being said I like the "flexibility reason" too. Could easily be both 😁 Thanks again for the gray matter re-charge. Cheers. 🇺🇸
I am no engineer. But what I picture when you're talking about this is Eddy currents going back and forth. Or electricity, A/C ...It can only travel so fast on such a thin wire with such a great resistance.
Hi Dave, You didn't do one thing when you damage this probe already: measure the whole lenght of this inner cable. You will be suprised how long is it!
I was right there with you until about the 24 minute mark. The scope probe as you mentioned is a distributed, lossy transmission line. You measured it to be around 330 ohms at DC / low frequency. It will be different at higher frequencies and as you mentioned is optimized for 10x mode and to achieve high bandwidth. The characteristics of the transmission line, however, are determined by geometry and materials and are the same whether you are on 1x mode or 10x mode. The transmission line doesn't look any different at 20 MHz (just picked a number) in 1x mode than it does at 20 MHz in 10x mode. Which means the effect of its interaction with the 15pF scope input capacitance would be the same at 1x mode or 10x mode. Also, what you are changing between 1x and 10x mode is whether the 9Meg resistor and ~15pF tip capacitance are in circuit or bypassed. So wouldn't the bandwidth limitation in 1x mode have to do with the interaction between the lossy line's RLC and whether or not the 15pF tip capacitance is in-circuit or shorted, and not the interaction between the lossy line and the scope's input capacitance? If not, what am I missing? Great video in either case. Thanks for posting.
If your signal contains frequency components higher than the probe's bandwidth (in x1 mode), the signal will APPEAR attenuated and 'rounded off' (on the DSO) but the signal itself is unaffected right? It's just a case where your probe isn't good enough to 'resolve' the higher frequencies (e.g., in the case of a square wave, the higher-frequency components are what makes the corners sharp).
Thanks for the video! But I can't understand how a passive circuit can have higher amplitude at the output than at the input. I mean the 'funny business' at 27:45, but I get the idea what was meant.
smaller voltages measurements most scope go up to 5mV/div, in x10 thats microvolts. HF x1 probes would be useful for that, a 20mVpp sine wave is at least visible in 4 divisions instead of half (2mVpp) at least that's what come to my mind.
I am not sure you have answered the question. The cable having a 3db bandwidth of 10MHZ explains the x1 losses yes. But the same cable is used at x10. Why does it magically not limit the bandwidth of the system when it is switched to x10. I feel you have given (very eloquently) half an answer.
Hi Dave, great video. Now I've finally got what's it all about with the center wire which I found to be as lousy as in your probe :) By the way, you use a great BBC like accent which lets me understand 99% of your lines, given my dumb ears. Thanks & all the best.
So basically what happens is that at x1 mode, more electrons have to travel at a time through the wire, because wire is too thin, it can't handle electron traffic. At a x10 mode, less electrons have to travel, so wire can handle the traffic. Have I got that right?
![[🔴LIVE] งานประกาศรางวัล MAYA TV AWARDS 2024 : มายามหาชน](http://i.ytimg.com/vi/nwmWYXXVwBY/mqdefault.jpg)
Thank You! You just saved me from trying to fix a local oscillator circuit that wasn't broke due to a low level reading on my scope with 1x probe at 36 mhz. I switched to x10 and there was my correct level! I really enjoy your videos.
Yes, highly recommended. I wanted to cover the x1 mode issue that Doug and everyone else seem to skip in their talks that focus more on x10 probes.
Dave, you've taught me so much about electronics. Thank you. I bet no one at my college will ever cover this!
So many answer to questions I didn't know I wanted to ask until you asked answered them. Very groovy Mr Dave you got my sub for the ride
"Now, the first thing you should say is, well, bullshit; show us!"
Thats gold, love it!
1X probes are not useless, they are very good for low voltage measurements, in order to keep the signal-to-noise ration as high as possible. Great video, though!
+electrocomm What he said was switchable x10 probes are useless as x1. You effectively have a hidden 470Ω resistor in series with your probe, and as shown your 'low voltage signal' will sink into the mud of noise @ 10MHz so it's useless for low voltage work as well ... more so.
+Peter Walker Agree, you might be right, the key word is "switchable", however, unless dedicated probes, these switchable 1X probes may not be quite useless; in a low BW (
LaurMTB ツ as Dave mentions, for a true 1:1 "probe", just make your own out of a piece of coax cable.
You should go into teaching Dave, I've learnt more in this video alone than a whole year in a master's EE course, keep it up!
Yes, it's big subject actually, and you could spend ages investigating how it all works and is optimised for best performance.
excellent explanation, spot on. I had this explanation as a baby ship's radio officer in the late 80's in those days our radio and even some radar kit was lower then your average CPU FSB. - times change.
This should really be compulsory stuff for new engineers yet as you say it's rarely mentioned, almost forgotten knowledge - tell us more Mr von Daniken !
Thank you Dave for this excellent video! I love all your videos but it is nice to pick things up like this which weren't discussed in the university (unfortunately). It may take some work but I'd hope the appreciation from the community encourages more in the future! Thanks!
They're great for me, because I'm just doing audio (20-20k Hz) right now using an old analog 30MHz scope. I do want to do radio at some point though, so a 1GHz scope and probes would be nice to have when I start with it.
I loved this video. I know a fair bit of theory and thought we were going to get into rise time limitations of the R/C compensation. I NEVER knew about the deliberately lossy transmission line. Dave, please do more theory vids.
Dave, One missing element in your trasmission line analysis is INDUCTANCE. Impedance (Z) is equal to the square root of the ratio of L /C. The whole problem of BW in your case is related to matching (optimal transfer of power) the input and output Z's to the trasmission line. The 9 Mohms helps expanding the BW but introduces resistive losses, hence the X10. A 50 ohms coax will not help without proper matching or termination. I think it is time to review transmission line theory. Cheers!
Brilliant explanation Dave.
It would be great to continue this by explaining differential probes, how they work and possibly design one to build. Low frequency of course.
Holy crap, I learnt so much! Thanks Dave!
Btw: Could you do more videos like this?
Very grateful for this video seeing as I'm currently playing with building an 80MHz RF front-end and was wondering how best to model probes for LTSpice to understand what voltages the ADC will be seeing. Nice one Dave!
Low level signal measurement, e.g. 1mV/DIV. With x10 you trade signal level for more bandwidth, so your scope becomes 10mV or 20mV/DIV minimum
Thanks again Dave.... Your video's are much appreciated. I learn so much information each time. Keep them coming... Your a fantastic TEACHER.
Dave, that was hugely informative, thank you
I had so many issues with transmission line theory when i was working on my degree its a very difficult subject
Sometimes the route is much more enjoyable than the destination!!!! I'll not take for granted the X10 scale and will think thrice before I use the X1!!! Thanks so much!
Dave, this was great. You do tons of entertaining things, and you do them really well. Your educational stuff is absolute gold. Keep it up.
Thank you for answering this. Had trouble finding this anywhere else. Didn't realize a coax had so much capacitance. This threw me off on a high impedance measurement when trying to use 1x mode, now I know why.
The x10 probe reduces the signal amplitude by ten times. This forces the scope to amplify the input signal by ten to get back to the original scaling (which means it will also amplify any input noise by ten as well). So the x1 probe is more sensitive and can capture low level signals with better resolution and less noise.
Another reason why x1 probes are designed that way (internal lossy transmission line) is protection of the sensitive input circuitry of the oscilloscope. A direct coax cable surely has the best bandwidth, but it will also allow anything to pass, including signal under test as well as any static HV peaks which might easily damage the ESD devices inside the scope. The design of ultra-high bandwidth/high gain voltage amplifiers combined with effective protection circuitry is a form of art in the electronics world.
Blew my mind. I didnt have a clue about that. Never bothered to measure probe resistance. Excellent.
Now I am 10x more optimized to watch another EEVblog!
10x means divided by 10
I would say that everything is little simpler, since i had to design my own cable for oscilloscope probe.
Most important feature of lossy cable is to have very low capacitance, so wire must be as thin as possible. If you try to replace it with regular coax you still can compensate everything for any bandwidth but input impedance will be terrible, what makes measurements impossible( done that and was puzzled why everything stops working when i attach my probe)
probe impedance is basically cable impedance divided by multiplier.
lossy cable allows you to achieve higher input impedance
resistance does not seem to play significant role however it ensures that there will be no reflection because of improper termination
and also it may improve impedance by turning cable itself into atenuator
Interesting to see how the pure truth not earns many thumbs ups.
I add some other (probably even less popular) fact: the other reason of the high wire resistance is the material choice. The wire must be strong and flexible despite of the very small diameter, and also not soluble in tin during soldering, this is why it cannot be made from copper, but other materials with higher specific resistivity must be used.
Dave's probing genius rocks! Thank you for the video!
100x is also useful for low voltage sensitive ccts, e.g. 32K xtal oscillators and micropower /low leakage stuff. For low voltage, low frequency use, a 90M resistor in series with the top of a 10x probe is a cheap option.
No, it's all to do with transmission line matching and compensation. The probe is transmission line optimised for x10 mode, so is all mis-matched when you switch to x1.
I've been using scopes for twenty years and never knew this. I was taught to always use 10x because of the higher bandwidth, and noticed it on the datasheet, but never needed to take the time to figure out WHY.
thanks Dave, I had no idea about probes, but you're right about those switches on the probes, they can drive you crazy.
Electricians tape tends to fix that problem: set it, tape it, forget it.
Also that strip of red tape is an easy way to know which one of the probes in the box is set to 1X
And again, I hope I learned something! That was useful and most informative - thanks a lot :)
Great video Dave, I'm glad you explained this. Yes, the big white board on the wall is much nicer.
very useful video.
A series on sources/corrections for measurement errors would be a good video/series.
Eric Wasatonic did a fascination series on building his own x100 probe as well.
Look at this series starting 25 Feb 2013
Good to mention this, anyone who works in electronics has to know this, otherwise they should never have been hired.
@25:18" It doesn't matter a rats ass" ... gold aussie Tech talk hehe love it
Love the explanation, Dave...
Hi Dave you never said anything about the 50 ohm resistor of the signal gen. That should be included in your calcs in order to be more precise on the -3dB point. I have seen a lot of your videos. As always they were extremely helpful. Thanks again.
Awesome stuff Dave, really enjoyed this one! Whiteboard looks great
Back for a "2nd suck of the sav". Good stuff! Ok, I'm a slow learner.
Re: wiggle wiggle center conductor. I was told back in the day (elect shop) that the wiggle was to help [center] the conductor in the dialectic more consistently. On some coax's they will include a plastic/teflon? spiral to do the same thing. That being said I like the "flexibility reason" too. Could easily be both 😁 Thanks again for the gray matter re-charge. Cheers. 🇺🇸
Great video. Very well explained. Easy to understand for even a challenged hack such as myself.
I am no engineer. But what I picture when you're talking about this is Eddy currents going back and forth. Or electricity, A/C ...It can only travel so fast on such a thin wire with such a great resistance.
Do people prefer that one to the old hand held one? I think it works much better.
Hi Dave,
You didn't do one thing when you damage this probe already: measure the whole lenght of this inner cable. You will be suprised how long is it!
I was right there with you until about the 24 minute mark. The scope probe as you mentioned is a distributed, lossy transmission line. You measured it to be around 330 ohms at DC / low frequency. It will be different at higher frequencies and as you mentioned is optimized for 10x mode and to achieve high bandwidth. The characteristics of the transmission line, however, are determined by geometry and materials and are the same whether you are on 1x mode or 10x mode. The transmission line doesn't look any different at 20 MHz (just picked a number) in 1x mode than it does at 20 MHz in 10x mode. Which means the effect of its interaction with the 15pF scope input capacitance would be the same at 1x mode or 10x mode.
Also, what you are changing between 1x and 10x mode is whether the 9Meg resistor and ~15pF tip capacitance are in circuit or bypassed. So wouldn't the bandwidth limitation in 1x mode have to do with the interaction between the lossy line's RLC and whether or not the 15pF tip capacitance is in-circuit or shorted, and not the interaction between the lossy line and the scope's input capacitance? If not, what am I missing?
Great video in either case. Thanks for posting.
Brilliant info...so what is the advantage of having ORIGINAL probes , say for a Tektronic 465M if it did not come with original probes?
Cheers
Your videos is a brilliant youtube mine. 👍🏼
I have an old analog 20Mhz scope and I measured the BW of the probes to be 500kHz on x1
No, you can roll your own. I've done a video with Doug on high voltage probe design.
And this is why scope probes must be handled carefully. Center conductor is so thin.
Did you what, watched like 2 of them? They are all great!
No, usually not. Yes, the odd shuffle was required in this one.
I like this DaveCAD-HD! It looks great.
Great tutorial thanks we learn somethings every day.!!
Excellent video Dave!
If your signal contains frequency components higher than the probe's bandwidth (in x1 mode), the signal will APPEAR attenuated and 'rounded off' (on the DSO) but the signal itself is unaffected right? It's just a case where your probe isn't good enough to 'resolve' the higher frequencies (e.g., in the case of a square wave, the higher-frequency components are what makes the corners sharp).
Hi Dave, can you make some quick video on measurement of smd capacitors and coils?
Learned a lot. Thanks.
Excellent! As always! Thank you!
You could use them both, sort of picture in picture white board.
Waling off at the end like a professional you tuber. How stylish.
😊 Again.
Where that 1k resistance parallel to the oscilloscope input comes from?
Xlint. Color me impressed 👍😁
"It doesn't matter a rats ass" LMAO
Reflected capacitance? Please explain!
Great video, thanks.
Thanks for the video!
But I can't understand how a passive circuit can have higher amplitude at the output than at the input. I mean the 'funny business' at 27:45, but I get the idea what was meant.
Didn't think about that, yeah, I've swapped sides!
you are missing the scope ground in the probe schematics
Mr. Jones, thank you! :D
Damn it that large white board was trippy.
smaller voltages measurements
most scope go up to 5mV/div, in x10 thats microvolts.
HF x1 probes would be useful for that, a 20mVpp sine wave is at least visible in 4 divisions instead of half (2mVpp)
at least that's what come to my mind.
Very useful video!
Better white board in theory, but the hand held one was just cool. Try break the hand held baby out occasionally :D
So, is it best to get dedicated separate high frequency 1x and 10x high frequency probes if you need to test high frequencies?
Man, I love these videos
He is a professional TH-camr. :)
Deep. But it drives home the importance of a good probe.
Oh the funny little hidden secrets you unveil for us. Thanks, Dave.
I am not sure you have answered the question. The cable having a 3db bandwidth of 10MHZ explains the x1 losses yes. But the same cable is used at x10. Why does it magically not limit the bandwidth of the system when it is switched to x10. I feel you have given (very eloquently) half an answer.
Very nice video, Keep it up.
So, if I need an x100 oscilloscope probe, the only sollution is to buy one. Is that right ?
Thanks for the information :)
Also, why on every probe state 10 Meg when actually you need a 9 Meg to match 10:1?
I like the end, "I could have told you that at the start."
Just going to go ahead and subscribe now. Thanks for the content.
Big thumbs up,,, that was good stuff.
Thanks !
Hi Dave, great video. Now I've finally got what's it all about with the center wire which I found to be as lousy as in your probe :) By the way, you use a great BBC like accent which lets me understand 99% of your lines, given my dumb ears. Thanks & all the best.
Great vid thanks.
What do you think about the iPad app oscilloscopes that are available for those who don't have the $ to buy a proper one?
Hey Dave, at 23:10 the video jumps from having a 15 pF tip cap to a 1k Ohm resistor. What is that about? What am I missing?
Hi sir, When you fill in the formule of Xc = 1/2pifC = 1/ 2pi10MHz15pF you get 1k ohm. So you can replace them for the equivalent reactance.
@@TheMrTxM Thanks for your reply :-)
New board much better. What is the AC impedance of the probe coax?
hi can I use 150 MHz probe on 50MHz oscilloscope?
Great! Thanks man.
What is the compensation network compensating?
So basically what happens is that at x1 mode, more electrons have to travel at a time through the wire, because wire is too thin, it can't handle electron traffic. At a x10 mode, less electrons have to travel, so wire can handle the traffic. Have I got that right?
Butifull explanation sir.