These days, oscilloscope hacking is all about enabling features that the manufacturer baked into the hardware but locked out in the firmware. Those hacks are cool, of course, but back in the days of analog scopes, unlocking new features required a decidedly more hardware-based approach.
For an example of this, take a look at this oscilloscope beam splitter by [Lockdown Electronics]. It’s a simple way to turn a single-channel scope into a dual-channel scope using what amounts to time-division multiplexing. A 555 timer is set up as an astable oscillator generating a 2.5-kHz square wave. That’s fed into the bases of a pair of transistors, one NPN and the other PNP. The collectors of each transistor are connected to the two input signals, each biased to either the positive or negative rail of the power supply. As the 555 swings back and forth it alternately applies each input signal to the output of the beam splitter, which goes to the scope. The result is two independent traces on the analog scope, like magic.
More after the break…
If you’re wondering how this would work on a modern digital scope, so was [Lockdown Electronics]. He gave it a go with his little handheld scope meter and the results were surprisingly good and illustrative of how the thing works. You can clearly see the 555’s square wave on the digital scope sandwiched between the two different input sine waves. Analog scopes always have trouble showing these rising and falling edges, which explains why the beam splitter looks so good on the CRT versus the LCD.
Does this circuit serve any practical purpose these days? Probably not, although you could probably use the same principle to double the number of channels on your digital scope. Eight channels on a four-channel scope for the price of a 555? Sounds like a bargain to us.
Beam splitter ? sigh… A beam splitter or beamsplitter is an optical device that splits a beam of light into a transmitted and a reflected beam (wikipedia)
That’s the first thing I though, then I thought the whole article would somehow be about some optics
It’s almost like different fields use the same jargon to mean different things…
IIRC, Heathkit offered something like this back in the day.
I would call this a “trace splitter”, to avoid confusion with the optical component known as a “beam splitter”.
Also, referring to a “beam” in this context might lead people to conflate “trace” and “beam”. There are actual dual-BEAM oscilloscopes. These are fairly rare beasts. Their CRTs contain two sets of Y-axis deflector plates, and sometimes even two electron guns.
The much more common dual-TRACE analog scopes, on the other hand, simply have a more sophisticated version of the device presented here integrated into them. They usually allow for either rapid switching between channels during the course of a single sweep, or sweeping one channel and then the other.
So calling this cool little circuit a “beam splitter” is potentially confusing on two fronts.
Yep; they’d do the chopping at low sweep rates like up to 100µs/div, and alternate channels for an entire sweep at faster rates. I’ve used my home-made workbench computer to sample four inputs and put the samples in memory and then play them back on a single trace on the analog ‘scope, with a different offset for each trace as it stepped through, doing all of trace A, then all of trace B on the next sweep, all of trace C the next time around, then all of trace D, using one of the oscilloscope’s two channels for the Y, and using other input channel to trigger the sweep. This gave me effectively a 4-channel DSO, using the analog oscilloscope for the display. I’ve also done raster graphics with it, using the Z input on the back which is for brightness, which you can see at http://forum.6502.org/viewtopic.php?p=15348#p15348 . FWIW, I do have an Agilent DSO now, but it’s not one bit user-friendly. It’s user hostile! I hate it!
Yeah, a lot of DSOs seem needlessly arcane to set up and use.
Your homemade version sounds very cool, even if it lacks a bit in the ‘instant gratification’ department… ;-)
A couple of decades ago, I tried to use a particular Tektronix digital scope at work. I gave up after a few minutes because the user interface was not like any analog or most digital O-scopes I had used previously.
I’m not dissing all Tek O-scopes, just that one, it would have been nice to have such analog controls such as Volts/Div on a rotary knob instead of hidden in a soft key menu.
It was called a “Electronic Switch”
So do digital scopes – they cheat by drawing a line between points. If you switch the scope from “vector” to “dot” drawing mode, it will behave the same.
No, it’s not the same. Analog scopes display small-deflection (on the screen) signals better than large-deflection signals because the beam just spreads over more of the tube. If you imagine imaging the analog scope (which is how they recorded this stuff for data taking) the signal goes away as it gets bigger. This becomes a real limitation for faster analog scopes: the photons just aren’t there.
Dots versus vectors on a digital scope is just a display option. The signal’s there no matter what. You might end up with a similar issue on a digital scope with the difference between the small-signal/large-signal analog bandwidth, but that’s not related to its scale on the screen.
Could have used, oh, wait.
An Arduino?
No, wait wait, I got this…
A Raspberry Pi!
Right?
:D
Excel?
DOOM?
Not a hack?
B^)
Chop mode!
The Tektronix 500 series single beam scopes (includes a model 555!) took plug-ins of which the most common was the type CA. That had two modes — an unsynchronised switching between the two traces for slower signals, and a synchronised alternation mode that displayed one channel across the tube, then switched to the other for the next sweep. Fortunately the timebase trigger circuits were very good as the two traces were essentially non-coherent when using the ALT mode.
Very good oscilloscopes that were encountered in many work-places, but bulky.
I remember these! I rescued a Tek 545 with dual trace module when my school were going to throw it out, carried it nearly a mile home, parents were not impressed. Made an exceptional room heater as well!
Digital scopes do this all the time. It’s the entire reason why your max sample rate drops when you enable a second channel – for a single channel, they use all the available ADCs in a time-interleaved fashion. The fact that you can see the switching on the scope is just a display issue, you could trivially postprocess it to extract both signals.
You can think of it like time-division multiplexing like is mentioned in the article, or alternatively, IQ mixing: you’re multiplying one signal by “+1, 0, +1, 0” and the other signal by “0, -1, 0, -1”, which is the equivalent of combining two signals mixed by out of phase carriers with a DC offset applied to each one.
I contributed something similar back in Feb/2010
I used a DAC with 8 offsets and cycled through quickly to display 8 digital inputs.
I was doing a lot of RC servo work back then
https://hackaday.com/2010/02/11/use-an-analog-oscilloscope-to-display-digital-logic/