For novel ideas about building embedded systems (both hardware and firmware), join the 40,000+ engineers who subscribe to The Embedded Muse, a free biweekly newsletter. The Muse has no hype and no vendor PR. Click here to subscribe.
February's giveaway is a nice little DMM. Enter the contest here.
Episode 5: Siglent's SDS1102CML Scope - A Heck of a Value. Part 2
I'm Jack Ganssle and welcome to The Embedded Muse video blog which is a companion to my free Embedded Muse E-newsletter. In Part 1 of the series I looked at the Siglent SDS 1102CML 100 MHz digital oscilloscope. This is Part 2, and we're going to look into some of the features a little bit more deeply.
The scope does allow us to take all kinds of automatic measurements, and it's really quite a deep variety of measurements that we'll do. I mean, here's an example. I can set each of these buttons to be a different measurement. So here, for example, I can decide if I want to measure voltage on any of the sources, any of the channels using Channel 1. I can measure all of these different kinds of voltages, peak-to-peak, max, min, average, you name it. It's very extensive. I can do that for each of the channels. I can do the same thing in time. Take a bunch of different kinds of measurements, period, frequency, width, all of that stuff.
In addition to all of the measurements that it takes, it does support cursors as do most scopes, nowadays, and it's pretty standard sort of stuff. You can set cursors for time, voltage. You can set the channel. Here I'll go ahead and set the cursors on Channel 1, and I'll go ahead and set time cursors. Then I can select Cursor A or Cursor B. I'll use Cursor A and with this little all-purpose knob I can move Cursor A around. You can see it moving here. And over here you see the time for Cursor A and the delta, time between Cursor A and Cursor B, which is nice. I can press this button and now I'm moving Cursor B. Very simple, very easy to use.
And like most scopes today, it'll do all kinds of mathematics. It does the basic four arithmetic operations plus, minus, multiply, and divide. Personally, I never use these except with the rare occasion on using minus operator to calculate the difference between two signals. But it does do FFTs, fast Fourier transforms, which we're doing here. We're doing it on Channel 1. It gives a split screen although you can change that and have the FFT displayed as a full screen.
And here we see the peak. That's where most of the energy is. I can hit the Cursors button again and tell it that I want to use the math channel for cursoring and here's Cursor A. If I move Cursor A here, I can put it right on top of the peak, and you can see right there. Remember that's a 1 MHz sine wave. And sure enough, that peak where most of the energy is, is indeed at 1 MHz. So that's a nice feature.
One of the really nice features about this scope is that it includes filters on each of the vertical channels. By turning the filters on, I can select one of four types - low, I-pass or a band pass. In this case I'll select up here, and I can then select the frequency. Now I have a 20 MHz sine wave going in. If I turn this filter down way low, here's 15 MHz below the 20 MHz signal, it goes away. It's completely filtered out. As I crank the filter up, then the signal comes back in. Now as you can see, if I do this slowly the queue, or the edge at the edge of the filter, isn't particularly sharp. But still it's a nice feature to have. It's a little bit confusing because this symbol says to me a low pass filter, and yet the control seems to be having the opposite effect. So I guess you just have to read this a little bit differently.
Now, I'll point out one of my least favorite features of this scope, if you want to call it a feature. Here we've got a 1 MHz sine wave going in, and as you can see, it's jittering around in the horizontal axis. And it took me a while to figure out what's going on, but it turns out that if we change the input to a square wave, that jitter basically goes away. And so my theory is that there's a little bit of uncertainty on the trigger level. Now it turns out that an awful lot of these low-priced scopes I've looked at have the same problem, so it's probably just the nature of the beast.
What about rise time? That's something that really important to us digital engineers. Here I've got a square wave going in, and I've fed it through a shaping circuit so that the rise time is actually about a nanosecond. What does it look like on the scope? I'm going to crank the speed up here. We're up for the 2-1/2 nanosecond limit, and you can get a pretty good idea what that rise time is by looking at that. But why not be lazy. I'd go ahead and use the automatic measurements because they are so nice. I'll shoot for Channel 1, and the rise time is right here. And you can see it's measuring about 2-1/2 to 3 nanoseconds which for a 100 MHz scope is astonishing. That's a great number, so it does a great job in that department.
Now I've got a 100 MHz signal going into the oscilloscope, and you can see that the sine wave is a little bit distorted. And that, of course, is because it's a 100 MHz signal, and it's a one Giga samples being taken every second. So that there aren't nearly enough points in there to properly draw a correct sine wave. And we can see that when I turn on Channel 2. You know, as the shape gets even worse and that's because with two channels turned on the sampling rate drops to 500 million samples per second. This is entirely natural and you'd see this on any scope with the same specifications.
Here the Siglent is displaying a sine wave from a very spectrally pure signal generator. It's 50 MHz, and it's pretty much what you would expect. As I turn the sweep right down, slower and slower, we see these artifacts beginning to be displayed on the screen, until we get to these lower sweep rates where we get what almost looks like a beat frequency being displayed. As I turn the signal generator just a handful of kilohertz off the 50 MHz center, we see that the artifacts change enormously. It's obviously not real, and in fact it isn't. This is a result of the digitization process the Siglent is doing, and it's not unique to that oscilloscope. If we look at the Agilent, which doesn't show that effect because it's a much faster scope, it samples at a much higher rate. But if I crank it down instead of reduce the sample rate, and I'll just freeze the display, and then I'll crank it back up, you can see the same sorts of artifacts starting to appear there. And the point here is that you really have to know what your test equipment is doing to use it effectively.
So what's the bottom line? At $379, I think this is a fantastic value in a scope. You can get cheaper scopes, especially one of the U.S.B. scopes, like this analog discovery which is a nice model, but you won't get anywhere near the bandwidth offered by the Siglent. And for me, at least, I really prefer a bench instrument with separate knobs and its own display rather than cluttering up the PC screen. If a 100 Mhz is good enough for your work, I highly recommend this device.
Do you need to eliminate bugs in your firmware? Shorten schedules? My Better Firmware Faster seminar will teach your team how to operate at a world-class level, producing code with far fewer bugs in less time. It's fast-paced, fun, and covers the unique issues faced by embedded developers. Here's information about how this class, taught at your facility, will measurably improve your team's effectiveness.
Free Embedded Muse newsletter - twice/monthly, hard-hitting technical info that goes to 42,000+ engineers. Click here to sign up or enter your email here:
Latest Embedded Muse: Rresponses to two articles in Muse 415 and book suggestions.
Advertise with us! Reach 200K embedded developers per month. More info here.