|For novel ideas about building embedded systems (both hardware and firmware), join the 39,000+ engineers who subscribe to The Embedded Muse, a free biweekly newsletter. The Muse has no hype and no vendor PR. Click here to subscribe.
Episode 7: An Oscilloscope from 1946 Can Still Teach Some Lessons
|July 21, 2014|
(Go to the complete list of videos)
In this video Jack examines an oscilloscope from 1946. Using only 4 tubes it does a pretty poor job compared with today's units, but some of the design is really surprising and clever. Plus, the thing is a lot of fun to fool with!
Digital scopes have in common for about a decade or so. Before that they all use CRTs for the display, and before that they're all vacuum-tube based. Hi. I'm Jack Ganssle and welcome to the Embedded Muse Video Blog which is a companion to my free online Embedded Muse e-Newsletter.
Today we're going to take a look back in history the way scopes used to be. This is a Philco 7019 scope that premiered in the late '40s. That's $66, that's about $700 today. For that pricing, you get a low-end digital Tektronix. That's a hundred, maybe even thousand times better than this. Let's take a deeper look.
Let's see how it works. We'll start with the CRT which, of course, stands for cathode ray tube and it's a good name because there's a cathode at the left-end there which is heated by a 6.3 volts of AC, and it shoots rays, or in this case, a stream of electrons down the tube where they hit a phosphorous screen. Keeping the electrons go through four deflection plates which when a charge is applied to them moves the beam either up and down, or left and right across the screen.
This is a CRT out of the oscilloscope and the deflection plates are located right back in this area here. To see how they're organized, so I can draw them on the front of the screen, there's two that control vertical deflection and two that control horizontal deflection. And though they are positioned in the back of the tube, the electron beam goes streaming through here, goes between the plates as deflected by those plates and then it impinges on the phosphorous screen.
The input signal to the scope goes through a vertical amplifier which increases the amplitude up to a couple of hundred of volts necessary to deflect the screen. There's also an oscillator which goes into another amplifier that drives the horizontal deflection plates, and the isolator's frequency can be controlled by a potentiometer as is shown on the screen. It also produces a saw-tooth like ramp, as I've just drawn, and what it does is it moves the beam across the screen from left to right at some relatively slow speed, and then returns very quickly back to the left.
Suppose you wanted to look at a sign wave on the scope display. In order to make a stable display, you'd want to start the sweep from the left side of the screen at the same point on the sign wave every single time. So that ramp generated by the horizontal isolator would be synchronized to the sign wave itself.
So going back to the Agilent here, we're looking at sign wave but there's no trigger established so it's totally unsynchronized and it's random everywhere. As I vary the trigger level, see, there's a trigger level, and I bring it down into the wave form, this is now starting to wave, going across the screen at this trigger point right here. Let's see. As I bring it down, you can see it starts that wave going across it lower and lower levels. That's what triggering is all about.
You notice that the signal doesn't look anything like a sign wave although by adjusting the sweep right here, I can sort of get it to look like that. And there are couple of reasons for this. Number one is the Philco has no trigger at all on it. This is done at a totally adjusted, repetitive oscillator going across a horizontal access which is not synchronized to the input signal. And by changing the frequency of that oscillator, I can attempt to match that of the incoming sign wave, but it's very difficult to get any kind of stable signal at all.
This bright flat line is because the return when the sweep comes from the right back, it is not blank. It's not turned off. If you look at a schematic for a traditional television set or something, you'd see there's a blanking interval where on that return beam, the beam itself is turned off. But this scope is so small, that doesn't happen.
Sort of interesting, there's not even a reticule, you notice there's no division, no boxes on the screen. The manual suggests that you go ahead and make one.
My fancy Agilent, of course, has position controls. I can move the beam up and down. As a matter of fact, it's kind of nice when you get to the zero position, it just stops because it's seems you might want to be there. You have to keep pressing, turning a button to move it past it, that's a nice feature. Or I can move it left and right. Every scope has this, of course.
But this scope doesn't have one. There's no position control. Two gain controls, sweep break, fine sweep break, where this sweep comes from, intensity and focus, that's it. That's all the controls. Where is the horizontal, where is the vertical position?
In order to save money, they give you a magnet. And you position the magnet inside the cabinet in order to position the beam properly. See how that is moving the beam around because, of course, as we know from our basic electricity, a magnetic field can interfere with an electric field. And what's happening is this magnet is pulling the electrons around and squishing. Oh, kind of distorts, what fun, huh.
If you've ever done this to an old fashion CRT TV, you know what happens. You put a magnet against the screen especially a colored TV, old fashioned black, a CRT screen, it will distort all the colors because it moves the beam to an incorrect positions on the screen. So it's pretty crude but it can be effective.
You can also see this line burn on the screen. For 60 years, this beam has been sweeping back and forth and has burned the phosphor off the screen. That was always a problem with CRTs in the bad old days.
This is a complete schematic of the scope. Astonishingly, it only has four tubes and one of those is, of course, the CRT itself, another is just a rectifier and the power supply. If you look in the upper left-hand corner, you see the vertical amplifier. The signal comes in from... into the scope, is amplified and you can see it's applied right to the deflection plates inside the CRT. Down on the lower right, we have the horizontal oscillator, and the frequency of that is controlled by the selection of the capacitors by the selector switch, as well as the potentiometer which I had pointed out before.
This is the horizontal oscillator inside the little Philco scope. It should be a perfect saw-tooth, but as you can see it's quite curve which accounts for some of the distortion of the sign wave that we saw on the Philco earlier. Now, this is about a 120 volts. We're not talking about low-level logic signals here. And you'll note, this interval here is the return time when the beam sweeps from the right side of the Philco back to the left. You want that to be very, very sharp so the return is very, very fast. But as you can see, it's not that fast. As a matter fact, it takes about 200 microseconds for the beam to return which also accounts for some of the terrible display we saw.
At the beginning of this video, I showed you a 100 MHz signal with a 1000 Hertz AM modulation on it on the nice Agilent. This is almost the same signal being shown on the little Philco, except that there's no way you can see anything, it looks like a 100 MHz. So I crank this down to the slowest signal generator will go, 500 KHz with that 1000 Hertz AM modulation on top of it.
As you can see, none of the halfway megahertz is visible at all, because the bandwidth of this scope is only a 100 KHz. That's 1000 what a little $400 100 MHz digital scope would give you today. It's nothing. And, as you can see, it's very difficult to synchronize that signal unless we are constantly tuning this thing here. And because of that rounded ramp that we saw on the other scope, the spacing between those signals is not constant as it really should be.
So it can't really display those signals that are so important to us today. We've come a really long way and I, for one, would not want to go back.
And so that's how oscilloscopes were back in the olden days. Thanks for watching and don't forget to go over to ganssle.com for more embedded videos, over 1000 articles on the subject, and be sure to sign up for the free Embedded Muse newsletter.