Hi, everyone. Thank you for joining Teledyne LeCroy today for our webinar on using your oscilloscope to make basic EMC and ESD pulse measurements with presenter Mike Hertz. A few housekeeping items before we begin. A copy of today's slides is available to you in the handout section of the control panel, which is to the right of your screen.



Also in the control panel, you'll see a section for questions. Please use this section during today's presentation to submit your questions. We will hold questions and answers at the end of the webinar to keep us running on time. This webinar is being recorded.



A link to the recording and slides will be sent to you automatically via email within the next 24-48 hours. Finally, as you exit today's webinar, a short survey will pop up. We would appreciate you taking a moment to answer these five questions so that we can continue to provide valuable content to you. So for those just logging in, thanks for joining Teledyne LeCroy today.



Our webinar is titled How to Optimize Your Oscilloscope for Basic EMC and ESD Pulse Measurements with presenter Mike Hertz.



This is part one of a new two-part series focused on how to get the most out of your EMC/EMI lab oscilloscope. We hope you'll join us for part two for more advanced techniques on November 9th. So a little bit about us before we begin. LeCroy was founded by Alabama native Walter LeCroy in 1964. Our corporate headquarters is in Chestnut Ridge, New York, but we have sales and service offices all over the United States as well as the rest of the world. Excuse me. Mike, could you just advance the slide one, please?



While we started as an oscilloscope manufacturer, focusing on physical layer tests, we have branched out into protocol analysis through several compatible acquisitions. In 2012, LeCroy was acquired by Teledyne Technologies, and we were renamed Teledyne LeCroy.



A little bit about our presenter, Mike Hertz. Mike is a field applications engineer with Teledyne LeCroy for over twenty years now and is based out of the state of Michigan. He holds a BS EE from Iowa State University and an MS EE from the University of Arizona.



Michael has several US patents for oscilloscope measurement design and has published more than 60 articles in the field of test and measurement. We know that there are a lot of demands on your time. We appreciate you joining us today. I'll now turn things over to Mike to start.



Hey. Thanks, Hilary. Thanks, everybody, for joining today.



Today, we're talking about EMC/EMI lab testing.



The way that oscilloscopes are used for EMC testing is typically to test the tester. So you're going to use an ESD gun. You're going to shoot it at some electronics under test. And before the pulse comes out of the ESD gun, you want to check it with a scope to make sure that the pulse is correct, and that's the main role that an oscilloscope has. Although, there are some other roles that we'll get to both in this seminar as well as the follow-on, which is going to be on November 9th, part two of more advanced topics for testing.



This is part one. This is focused on basic EMC ESD pulse measurements where we're testing the tester. We're qualifying that ESD gun or surge pulse generator. And then part two, we'll be looking at some more advanced techniques, and I'll talk more about that.



So here's the agenda. These are topics that I put together. I traveled around to different certified EMC labs and documented the use of an oscilloscope with the testing and some of the best practices and things that you can do to optimize testing and get the best quality measurements. Some of the topics we'll talk about are the test requirements and threshold selection. Threshold selection is a very critical point.



A lot of times, there are even some people in certified labs that didn't know all the details about setting measurement thresholds in a scope, so I'll go over that in detail. And then, also, ESD verification step by step. There are some tips and tricks using level after pulse and parameter limiters.



And, also, a frequently asked question that comes up about the 10605 standard for road vehicle testing. So I'll show how to do the consecutive parameter measurements for that surge testing.



And then another FAQ that comes up about what sample rate and V/div you should use to capture ESD pulses. So I'll show some guidelines for recommendations on getting the best quality waveform capture with those. And then we'll take a look at some getting started resources and take questions. So any questions you've got as we're going along, go ahead and just type them into the question field, and at the end, we'll go through all the questions and get everything answered.



First, starting with conducted immunity test requirements.



I put together a quadrant here that shows emissions along the top, immunity on the bottom. On the left-hand side is radiated, and on the right-hand side is conducted.



The way that oscilloscopes are used is typically in conducted immunity and also in radiated immunity. Within conducted immunity, that's tests like ESD pulse testing, electrical fast transients, which we'll talk about in part two, and surge testing. So, those are all within conducted immunity. And then within radiated immunity is when you've got the scope in a reverberation chamber and you're bombarding it with radiation from an antenna.



You want to find out if the electronics under test is susceptible to radiation, and if so, how much radiation does it take and at what frequencies in order to get the device to malfunction. And then that information is useful to make sure that the electronics under test meets certain radiation standards. So that's in radiated immunity. And then along the top, we've got emissions testing.



This is typically done with RF equipment like a spectrum analyzer that has very wide dynamic range and narrowband capability.



Typically, you wouldn't use a scope for the areas here in the gray, but there's some debug work you can do in the gray areas, and I'll be talking about that in part two as well. So right now, we'll be focused on mostly conducted immunity and starting out with ESD.



Here's a graphic image just showing the basic concept of ESD, which is if someone walks into the lab, they've got static on their hand, they touch a piece of electronics, it may damage the electronics. How much ESD can the electronics withstand? And that's what's being tested with the ESD gun.



Typically, ESD has very high voltage but low current, so there's not a lot of power. There may be enough power to do some damage, and that's what's being tested for. The voltage could be very high. With ESD guns, you might be going to 16 kV, 20 kV, 25 kV, etc., so very high voltages.



And you could even see visible sparks. With ESD testing, a lot of times, there's a spark gap where you'll see a spark emitted from the ESD gun that hits the electronics under test during the testing.



Here are the general test requirements. First, to generate an ESD pulse.



This is an example of one. I put an image of an actual ESD pulse capture here on the right-hand side, and some of the measurements of interest are to look at the rise time, which is this very first leading edge. The spec wants to ignore all the other leading edges that could occur, but just focus on that one leading edge. Measure its rise time. It's very, very fast.



And then focus on the peak current. So even though voltage is coming into the scope, most of the specs are interested in the current, so you just divide by the impedance of the input. So you've got voltage divided by resistance, gives you current, and then you can measure the current. Let's say it's 20 A or whatever the spec requires.



Pulse width, and pulse width is defined where that's measured at for various standards.



And then two other common measurements are to measure the current at two different time delays after the incident pulse. So the incident pulse happens here, and many of the specs are interested in how much attenuation has happened by the time we're out 400 ns, 800 ns, etc., to measure what the amount of attenuation is at that point. Some of the devices under test that are checked are things like automotive parts. You might want to check things like airbag sensors or cruise control units. Check telecom boards and other high-speed PCBs. If you've got some electronics that's going to be handheld, maybe it's something like a cell phone, tablets, or computers.



Different types of electronics are tested to make sure that if they're hit with the static, they're able to survive that.



Some of the ESD standards of interest are the IEC and EN 61000 standards, very popular, ITU, Underwriters Laboratories, Telcordia, ANSI standards, Bellcore, etc. I listed some of the more common ones, and the 10605-2001 and 2008 are very common. I get a lot of questions about those. MIL-STD, other types of proprietary military, proprietary automotive, etc., and CE Mark is another one.



So this is what it looks like when someone's performing the ESD test.



This was an actual ESD test setup, and the test engineer is operating an ESD gun. He's shocking a protocol analyzer, and it's transmitting data over to the computer. So what he's checking is to make sure that as he hits the electronics under test with the ESD pulse, does the correct data get transferred to the computer? And that's what the test setup looks like.



Another test that's performed is when the oscilloscope itself is the electronics under test. In this case, some of the more common touch points on a scope are tested for ESD. So if someone may have static, they've scuffed their feet a little bit, walk over to a scope on a bench, touch the horizontal position knob, and when they touch that, then it could, if the scope is not ESD protected, do some damage. So Underwriters Laboratories is going to check that out to make sure that the scope can withstand ESD as well because it's just another piece of electronics under test when it's being tested.



And another touch point is the touchscreen display on the scope. So here, the ESD gun is touching the touchscreen, and then it's going to check, can the touchscreen handle 16,000 V? We'll find out.



And then, when the ESD test is performed, this is an example of an official compliance report from Underwriters Laboratories that's doing testing. The electronics under test in this case is an HDO8108 oscilloscope.



They submitted a 57-page report. I just took a photo of the title of it to show what a real compliance report looks like. And it either passes or fails, and then there are details in the report showing what happened.



This is an example of a calibration setup, and here the oscilloscope, rather than getting tested with an ESD gun pointed at it, is going to be doing the testing to make sure that the ESD gun outputs the right pulse shape. And this is required for anybody who's doing official ESD testing. They're going to want to take an oscilloscope, hook it up to a current shunt target, have an ESD gun, send a pulse into the target, and then measure with the oscilloscope what that pulse looks like. And once the pulse shape is verified, then at that point, testing can begin, and someone can take that gun that's been calibrated using the scope and start applying it to electronics under test.



And this is from the IEC 61000 standard, and it shows the ESD simulator. So here's the ESD gun. It's connected through some shielding around the scope and the attenuator and cable chain. And this attenuator and cable chain is typically a 26 dB attenuation, and the purpose of it is to make sure that an oscilloscope doesn't get hit with that high voltage.



It's going to reduce the voltage down into a smaller amount because all the standards require that the oscilloscope is in 50 Ohm coupling, channel coupling. So the default would be 1 M?. It says switch it to 50 Ohms. And when you're 50 Ohm coupled, then you're going to have, usually a maximum of 5 V range on the scope channel.



So if you have an ESD simulator outputting 25,000 V, you have to get that voltage down into something that all the oscilloscopes in the scope industry can handle when they're 50 Ohm coupled and they're going directly into the channel, no probe, no 1 M? attenuation, etc.



To do that requires an attenuator and cable chain, which is part of the required setup for ESD testing. And then an oscilloscope here, it's shown greater than 2 GHz bandwidth.



Some standards are greater than 4 GHz, some are 16 GHz, many are 500 MHz or 1 GHz. It varies by the standard. In this case, the IEC 61000 was requiring a 2 GHz scope bandwidth.



And the way this is set up here, there's a tripod that's holding the ESD gun steady, but a lot of times people just hand hold it, take the ESD gun, and point it at the electronics under test. In this case, because this is the portion where the oscilloscope is used to make sure that the pulse is correct coming out of the ESD gun, then it's set up pointed right at the current target, which is on this vertical calibration plane. And on the other side is the scope that receives the signal, and the signal is going to look like this. So, it's designed to have a very fast rise time, comes up to a peak.



Here, it's hitting 15 A peak current. Then it comes down, comes back up again. So here's a non-monotonic shape to it, and then it settles down. And the IEC standard is saying, what's the current at 30 ns after the incident edge, and what's the current at 60 ns afterwards?



It's labeled here as just I30, I60.



And Ip is the peak, and these are the standard requirements of what needs to be measured with the scope.



And one thing to notice is sometimes people ask me why the ESD gun has a pulse that's shaped like this, and it's to simulate the impact of somebody who has static. When their fingertip first touches the electronics under test, then you get that initial pulse that comes up with a high peak, and it's got a fast decay. And then the secondary pulse that comes back up. This is the charge in the rest of the human body that's working its way toward the fingertip and then comes out and hits the electronics as a secondary wave, and that is why ESD guns have that shape.



The pulse characteristics, typically, the rise time on a fast event like this is between 500 ps and 1 ns, so the scope will measure what the rise time is. Typically, in that range, it could be faster depending on the spec and on the gun. And capture a single pulse, and some standards require to capture 10 in a row. I'll show how to do that in an easy way using a scope in this presentation.



And then also measure that pulse, verify the rise time for positive pulses. You can configure an ESD gun to put out negative pulses. You can measure the fall times on those to get those values. And then you could use, depending on what the standard requires, 1 to 4 GHz or more bandwidth.



And, some of the questions to look at is how is the rise time measured on this pulse?



And also, the 10% to 90% rise time measurement first requires that you know what 0% and 100% are, and the scope needs to figure that out. And in order to do that, it leads us up to the next item, which is ESD threshold selection, and this is a very important concept that has to be set up correctly in order to get good measurements.



First, I'll start with the IEEE definition.



All oscilloscopes by default are going to use this IEEE definition. This is the IEEE 181 standard, and what it says is it's looking at a clock pulse shape. ESD pulses do not look like clock pulses, so I'm going to show in the next couple of slides how to handle this problem because all oscilloscopes are going to measure according to IEEE 181 by default unless you make some changes in the scope.



And IEEE 181 is saying that it's expecting a clock-like shape where the signal rises up. It overshoots over the top of 100%, and then it comes back down again, and it may go below 0% into the negative territory and then back up to zero, and the 100% level on the pulse is considered top. This is an industry term: top and base. Base is at 0%, and the signal actually goes above 100%. It goes below 0%, and it uses that steady-state high and low value, which are determined in the background by oscilloscopes by looking at statistical mode.



And so here, this is the main mode of a statistical distribution of voltage values, and the top mode and bottom mode form top and base, and that's 0% and 100%. So if you want to measure rise time from 10% to 90%, that's always referenced to 0% and 100%, and the 0% and 100% by default are determined this way, which does not work correctly for ESD pulses.



So I'll show how to resolve that. This is a real clock pulse that I acquired on a scope, and I set up a histogram showing the vertical distribution here of voltage values and the main mode is somewhere along this trough at the top and the base value is where the signal dwells the longest on the lower half of the waveform. It's not the negative peak, but it's a steady-state value.



And we can see on the histogram distribution where all oscilloscopes will find 0% and 100%. So when we're dealing with an ESD pulse like this one, I took a real ESD pulse, and I did the same vertical histogram on the scope. And here, it's showing that all oscilloscopes in the scope industry by default are going to put the top value incorrectly for an ESD pulse along this secondary pulse.



And on ESD testing, the goal is to measure between 0 V and the peak. And so this type of measurement will result in a measurement error for ESD testing.



And this is a worst-case scenario where I took a real ESD pulse, and I had the scope show me where it found top and base at. So this is what all oscilloscopes in the industry will do by default, is define top and base where the 0% and 100% are. And here, because the waveform has decayed and then it dwelled for a long time at this voltage, then all oscilloscopes in the scope industry are going to put top incorrectly.



It's correct according to IEEE, incorrect according to the ESD standards where the top is labeled here. So, the scope actually put this blue line for top, the blue line for base, and it's measuring the rise time down here, which is not what the standard specs want.



What the standard specifications want for ESD testing is to measure between ground and 100%. So they would want the 100% level where base is found by scopes to be up here at the very peak of the waveform, and then they would want 0%, which is what oscilloscopes would find as the base measurement. Instead, the ESD standards want that down here at 0 V.



So in order to resolve that, it's very important to set the threshold instead of being in default percent, which is used in top and base like in the image here, we can see the blue guidelines are being applied to the rise time. It's just showing a small portion of the rise time. It's measuring that as 494 ps.



But by setting the thresholds to zero to max, this will set 0% at 0 V and the max will become 100%, then the 10% to 90% rise time is correctly measured at 854 ps.



So 494 ps versus 854 ps, that's a 73% difference in the measurements. That's very significant.



Huge error percentage, just based on one setup that a lot of people are not aware of. So very important to set the thresholds, and that's how to do it. This is a comparison where I took an ESD pulse from a real ESD calibration setup.



The ESD gun was hooked up to the vertical calibration plane through the current shunt target, and we can see measurements side by side. And the first measurement, zero percent to max, is measuring 992 ps, which is correct. And the second measurement, 336 ps, is off almost by a factor of three. So it's a very significant measurement difference and done side by side on the exact same pulse.



A similar measurement here for the standards that require pulse width, I've got a pulse width measurement set up here and we can see this vertical line is showing the width difference.



And when it's just very slightly higher in voltage, then it measures the width at 13.2 ns.



But then very slightly lower, it's 23 ns, so it's a big difference. And this was used in percent zero to max versus default IEEE percent as the threshold selection. So you can see here it makes a very big difference in the measurement result.



And in this case, if the threshold were to go even lower, then it's changed the pulse width measurement from 2.1 ns to 50.3 ns, so that's over a 2,200% error. And so it's very important that the thresholds are set up correctly because you could have massive amounts of error just from a very slight difference in the threshold setting.



The next item is to talk about some shortcuts and some tricks you can use to meet different standard requirements.



One of them is to use EMC level after pulse.



And what looks like a cursor on the screen is actually not a cursor, it's a marker indicating a measurement that's being done automatically. This is parameter one is EMC level after pulse, and it's got the delay set to 20 ns, and it's measuring what the vertical value is there. And then here as I step the delay along to 30 and 48 ns, then we can see the difference in the EMC level after pulse measurement.



Another shortcut is the EMC time to half value parameter. So it starts out with a peak value, and to find the amount of time it corresponds to reach half of its value, you could use a built-in measurement, take a shortcut, and measure what is time to half, and it just reports it as an automated measurement.



And then another technique you can use is because you could have a situation where there are multiple peaks that cross through the same threshold.



In order to avoid measuring that second peak because the ESD standards ask you to measure the first peak, they don't want the second peak. If the oscilloscope can measure all the peaks through the threshold, the way to sort them out is to use the parameter limiter, and you could select that. You just want to measure the first peak. You want to ignore the second one. And parameter limiter here, I set the number of peaks to one to just report that first one. So that's a way to make sure that you don't make a false measurement by measuring the second peak through the threshold.



And taking a look at another situation, this is a real ESD pulse.



Here's the threshold crossing, and you could have a peak that crosses through the threshold multiple times. So you've got the main peak, it comes up, comes back down through the threshold, it's got a secondary peak, comes back up a third time and then a fourth time each crossing through the threshold, and I put a magnified view of that situation here where we can see each of those peaks that's crossing through. And at this point, we've got measurement statistics that are showing the min, max, standard deviation, etc., for the width measurement.



And it's hard to tell which width was the first one, which is the only one that ESD specs are interested in. And which one are these other three that were correctly measured by the scope but incorrectly applied for an ESD measurement? And so the way to sort that out is you could use a parameter limiter, set the number of measurements that you want to be one, and it'll just take that first measurement, and it's reported as 1.181 ns, which didn't show up at all as the min, max, or mean or standard deviation of the measurement. So it's good to be able to sort that out, and this is a technique you can use to bring out just that first and correct width, rise time, etc.



Alright. The next area is to talk about verification and test setup. This is a photo I took in a certified lab of what it looks like to do the ESD pulse test.



And on one side of this partition is the ESD gun sitting on some foam. It's got... its grounded on the vertical calibration plane. The current shunt target's here, so you can take that gun, shoot a pulse into the current shunt target. It's coming out the other side, and this is that attenuator chain that we saw in the 61000 spec that has a series of attenuators.



It's got a 20 dB attenuator that's hooked up to the current shunt target. It's going through a high-performance SMA cable. It's coming into a scope that has a 20 GHz bandwidth. It's doing testing that requires 16 GHz or higher for a specialized ESD standard.



And then on the scope side, it's got another 6 dB attenuator, so it adds up to a total of 26 dB attenuation, which is the amount of attenuation you're probably going to want to use for most ESD standards.



And on the scope, it also captured multiple pulses. So I got it set up here to capture 10 ESD pulses, and it's doing a trend of it. I'll show you how to set that up shortly.



Here's a photo showing a test engineer. He's holding the ESD gun. He's shooting pulses in there, and then the scope captured them, one by one.



And here is on a different day with a different setup, but the same test engineer and the same gun. And he's firing it into the scope here, and it's captured an ESD pulse. So this is what it looks like for an ESD pulse capture setup.



This is a portable setup. So if you want to travel somewhere and you're bringing an ESD gun, then you can get a miniature vertical calibration plane to use. And here, the ESD gun's being fired into that, coming out the other side of it and into the scope for a capture.



This was in another ESD lab, and we can see the backside of the vertical calibration plane is being captured by the scope here. And then, here's the front of it where he's got the ESD gun, and he's hand-holding it here and firing it into that current shunt target.



And then here's an example, what that looks like here with the ESD test setup, some other equipment running in a different area.



So on the scope side, then, this is a scope screen capture that I took in the certified lab. Measuring the rise time of an ESD pulse, the levels are set percent zero to max.



The max value is being applied, and here, it's just rescaled that voltage waveform into a current. You could do it directly from a channel or you could do it in a math operator. Both work fine. It's measured. The max value is 32.8 A peak current through the waveform.



And then here's the EMC level after pulse at 400 and 800 ns, and this is the current at T1. This measured 2.01 A, and it started out at the peak at 32 A and is already down to 2 A after 400 ns, to give an idea of how fast the decay is. And then here on parameter four, then it's down to 1 A after 800 ns or the current at T2.



And that's how it's set up. So those are four of the main measurements that you might want to do for the ESD test. And so peak current, rise time, current at T1, current at T2. This is how the measurements are set up. That's how to actually do the measurements.



A frequently asked question has been for ISO 10605-2008. They introduced the consecutive parameter measurement requirements. You need to do 10 in a row, and people have been looking for what's an easy way to do that on an oscilloscope. So I just created some quick documentation to show how to set that up. And what you need to do is just turn on measurement statistics when we have these four measurements set up that we were just looking at.



When we turn on statistics, then the num field is the number of measurements that were performed. It shows there have been 10 rise times, 10 maxes, 10 EMC level after pulses, and the statistics show the measurements that have taken place. And then you could turn on histicons that are showing the rise time. This shows rise time on the x-axis versus number of occurrences on the y-axis. So you get a statistical distribution starting to form for each of those ESD measurements that were taken.



And then you can trend them. So in an oscilloscope, you could take advantage of a math operator called trend and apply that with the input to the trend being the measurement that you're taking here. So we're measuring the rise time of an ESD pulse. We want 10 of them. We want to take a look at like a data logger would show the sequential values of each of those rise times. So what we're seeing in the green trace is rise time on the y-axis, and the x-axis is chronology. So it started out here at the zeroth event up through the ninth event, so it's got 10 total.



And on the left-hand side, we've got voltage versus time as that pulse is acquired. And on the right-hand side, we've got the measurement value itself, which is rise time versus chronology.



And so this is a technique that'll allow you to capture the measurement values in the order they occurred, and then you can extract that chronological trend to show what's transpired.



So here are the exact steps on how to save that because a lot of people have asked, well, how do you use that data? How do you extract the data? How can I get it off onto my computer or whatever you want to do with it? You can set up first, the measurement is rise time. The source is channel 1.



And the operator is trend, and the measurement is P1. So P1 is the rise time at channel 1, and then F1, which is a function, is the trend being applied to that measurement P1.



And then you can, then you've got your trend plot, which is this green waveform that's constructed of a series of measurements that were taking place. And then you can take that trend and save it, for example, as a CSV file. You have comma-separated values. You want to bring it into Excel or use it or just save it as a report or whatever you like to do, and you can save it here. And I'm choosing not the input channel, but trace F8, which is the function that is the trend of the rise times. So that's how you can extract 10 consecutive measurements or 100 or 1,000 or a million or however many you're looking to get.



And in this case, I trended all four of the ESD measurements, not only rise time, which is the green one, but also the other three measurements as well. And then we can see the evolution of those measurements over time.



And starting from the first measurement out to the tenth one, we can see, for example, that the EMC level after pulse, which is this blue one, has quite a bit of variation on it. We can check the scaling on the y-axis.



Here, I separated the trends of the ESD measurements into different grids. I've got three histograms showing the number of occurrences on the y-axis versus the measurement value on the x-axis.



And the trend, by contrast, on the y-axis is now showing time, which is the measurement itself that's being measured versus chronology. And this time, I ran it up to 50 different measurements so that we could take a look at the changes from one measurement to the next.



And that's how to set that up as a trend. And then here, I'm selecting I want to save the waveform. I want to save all of them displayed, and I'll save all of them here too, to the E drive. We'll save it on a thumb drive. I want to use it for something else and save it all here as CSV files.



So the next topic is surge testing.



Surge pulses are similar to ESD pulses in that they have a very fast rise time and they have a slow decay, but the fall time is much, much slower than the rise time. They also have some similar measurements in terms of measurements such as rise time and max, but they've got some new measurements like area under the curve and charge, and I'll show how to do those measurements on a scope.



So this is a real surge pulse that I captured, and we're taking a look at measuring the rise time and fall time. So rise time, less than a microsecond. Fall time, 45 µs, so a much, much slower fall time.



And this is a real setup. I took a photo in a certified lab of a surge generator hooked up to the scope. Similar to ESD testing, the role of the scope in surge testing is to test the tester. So before a surge generator is hooked up to the electronics under test that you're doing the testing on, it's required to hook it up to a scope first, make sure that the scope measures the surge pulse as within spec. And once that's the case, then you can take the surge, hook it up to the electronics, and find out what happens.



And here, the surge generator has an attenuator hooked up to it, and then it's running through the cable into the scope and being captured.



Rise time, width and max, I won't talk about because those are very similar to taking a look at ESD pulses. But what's new here is the area and the charge. And the area, you can measure there's a measurement parameter in scopes. That's the area parameter.



And when it's being applied to channel 2, it's measuring in microwebers, and it's given a value here. And to convert that into charge, you can just scale it with 50 Ohms. So here we've got a 50 Ohm parameter constant on parameter seven. And then by dividing the area under the curve by the parameter constant, it's given the charge in microcoulombs, and it's 18.2 µC being measured on parameter eight.



So that's how to measure surge pulses.



And a question that has come up a lot, so I just created a section on it, is to know what sample rate and V/div you'd want to use for ESD testing.



And I ran an experiment to find out what the optimal settings are to use.



And I started with an ESD pulse sampled at 40 GS/s and then zoomed in on it over here. So this zoom area is highlighted on the original trace, and then this ESD pulse sampled at 4 GS/s, this one at 2 GS/s and at 1 GS/s. And the corresponding zoom of each of those color-matched zooms over here is zooming in at, these are all on the same time scale.



These are all on the same time scale, but the sample rate is varied. And on an oscilloscope, you can manually set the sample rate up to the max sample rate of the scope down to very, very low sample rates. And to know what the correct one is to use, this experiment shows us. So when we've got five or six sample points on the rising edge here, we can see the shape of that edge is pretty similar.



When we're down to 2 GS/s, then now it looks like somewhere between two, three, or four samples depending on how we count it along that rising edge. And here's somewhere between one to three samples depending on how we count the number of samples along this rising edge.



So the measurement result, I hooked up the connected rise time measurement to each of these different traces.



And the first one had a, this is a reference measurement at 40 GS/s measured at 839 ps, so that's 839.



When we sample at 4 GS/s, it's 865 ps on the same waveform, so that shows a 3% measurement error from the sample rate reduction.



Sampling at 2 GS/s, now it's just divided by two, went from a 3% measurement error to a 28% measurement error.



And then the fourth one at 1 GS/s shows a greater than 84% measurement error, on the same pulse. The only thing here that changed is the sample rate.



So the sample rate makes a big difference on signal integrity measurements like rise time on an ESD pulse, for example.



So what sample rate is needed? And, I ran a test here where I varied the sample rate until I found an optimal point. And if you can sample the ESD pulse at a rate of 10 sample points per rising edge, then you can get the measurement to be within 0.5% of the final answer. So here, the rise time when it's 10x oversampled on the rising edge is 843 ps compared to 839 ps on the reference measurement when it had 50 sample points on the rising edge. So this experiment shows that if you can get 10x oversampling on the rising edge, then you're going to be within 0.5% of the final answer, so that's great.



I wanted to point out because each ESD pulse has a slightly different shape that in order to do a true apples-to-apples comparison, isolating just sample rate instead of having four different pulse shapes to measure the rise time. I just took one sample at the max sample rate that the scope had, which was 40 GS/s, and I decimated it with a math operator called sparse. And what sparse does is it takes the original waveform and just throws away some samples, so I threw away extra samples to decimate it from that 40 GS/s down to 4 GS/s, to two, down to one. And this was all from the same scope capture, which allows you to isolate only the effect of sample rate and not the effect of variation of ESD pulses from the ESD gun.



So that's how to set up an apples-to-apples comparison on sample rate if you ever want to run an experiment on sample rate.



Another very fundamental and important criterion is to look at the V/div on a scope. So sometimes people think that the scaling of the V/div is just a visual effect, but it actually has a huge impact on signal integrity.



And if you can capture the waveform and fill most of the grid on every oscilloscope in the scope industry, then if that scope is an 8-bit vertical scope, then you can maximize the use of the 256 quantization levels that 8-bit scopes have. So, 2 to the power of 8 is 256. If the scope is a 12-bit vertical scope, like a lot of LeCroy scopes are, then you'd have 2 to the power of 12 or 4,096 quantization levels vertically. And the more the scope can maximize the waveform that's being captured to fill the ADC's range, the more of those counts of the ADC you can use in the capture, so it makes a difference in the signal integrity. So, for example, if you had an ESD pulse and you can fill most of the display, then you'd have all 256 quantization levels of the 8-bit scope or all 4,096 of a 12-bit scope captured.



Or you could have seven if you don't fill the... if the ESD pulse doesn't fill the whole screen vertically, then the digitizer will continue to digitize everything that's on the vertical scaling. So it'll digitize nothing for half of the screen, and then it'll use half of the other digitization values to fill the portion where the waveform is. So, effectively, because half of the screen captured no signal, then the other half did all the scope capturing. So on an 8-bit scope, for example, you'd have 7 vertical bits.



If you captured an ESD pulse and it only happened to fill one-quarter of the screen, then you'd have 6 vertical bits or 64 ADC levels on an 8-bit scope, and the other three-quarters of the screen would still get digitized, but since there's no signal there, you'd be left with a worse signal-to-noise ratio on the digitization on every scope in the scope industry.



And the reason why that happens is because when the ESD pulse or any other analog, real-world waveform comes into the amplifier of a scope, then the V/div setting that you select affects what gets digitized by the digitizer.



And then that range, it only utilizes a small portion of it, sends that into the memory, gets processed, and you're doing measurements on a signal that has a lower signal-to-noise ratio.



So some of the getting started tools that you could use, I wrote a couple of app notes.



One of them is on ESD pulse testing. You can click on the link here. It'll hyperlink right to the website. You can download it if you want to share it with colleagues who weren't here today. I wrote another app note on radiated immunity testing that will be in Part Two, and I've also published five articles on the topic that are available, so you could do a search. Since they're not on Teledyne LeCroy's website, I only included the app notes, but there are other resources out there as well.



And a few closing comments. So, it's very important, because ESD standards require measurements between 0 V and the peak, that you can manually set the thresholds to zero to peak in order to account for that to take accurate ESD pulse measurements. And then there are some specialized techniques you can use, level after pulse, parameter limiters, trending, etc., to make your job easier when you're testing ESD pulses.



And then the other point was the sample rate and V/div setting actually change the measurement results, so it's good to maximize both of those or optimize both of those so that you can get the best signal integrity, and then you get good quality waveforms being captured, good quality measurements taking place.



And there's going to be a part two where we'll get into a bunch of other topics that we didn't have time to cover in one session. So we'll talk about electrical fast transient (EFT) testing using sequence mode segmented memory in the scope to capture and do debug work on EFT.



Also, radiated immunity testing, which was one of the other areas in the quadrant.



And also, some MIL-STD measurements, momentary power interruption, and how to do those measurements that are all EMC lab requirements.



And there are quite a few upcoming events. So I think Teledyne is doing more than 100 different webinars this year. It might be as many as 120.



There are many of them happening per week. They're on all different types of topics and not only on scopes, but also on protocol analyzers and other types of test equipment, TDR measurements, etc., S-parameters. Eric Bogatin has a series of webinars on here as well. So you can view on-demand webinars by clicking up here to view there are probably well over 1,000 of them that are archived on demand, but also new ones coming up each week.



There are new ones you can click on on this events page shown on the web link here. And so you can view them from the past, view the ones that are coming up. And then also, there is a relatively new tool called MAUI Studio that allows you to take a lot of the measurements that you do on a scope. You can take them offline and have the scope software running on your computer.



So you can find out more about that here at teledynelecroy.com/mauistudio.



And now we'll go ahead and open it up for questions. So, Hilary, let me know what questions we've got. We'll go ahead and address them.



Yep. Thanks, Mike. Thanks everyone for joining us today.



A couple of you asked about the slides. The slides are available in the handout section of the control panel. I will also send them in a follow-up email that you will get within the next 24-48 hours along with the recording.



Don't see any other questions.



So I will go ahead and close the webinar for today. My email and Mike's email are both available if you come up with something after today.



So thanks, Mike. Thanks, everyone else. Hopefully, we'll see everyone on the ninth or sooner for a different webinar.



Alright. Thanks, everyone.



Have a good afternoon. Thank you.



You too. Take care.