General Analog/Digital Oscilloscope Question

[DustinB]

So I have an analog tek 2246 on the way and I was curious about something. I hear that digital scopes are required for small non-repeating signals. Is there a real world example of something I could not measure?

Let's say I'm working with the ENC28J60 ethernet chip, on one of the pins it send 2 pulses at start-up. Will I be able to see this with my analog scope?

 

[crutschow]

If it's two non-repetitive single pulses, then that would be hard to see with an analog scope. Analog scopes need repetitive signals to view high frequency or fast waveforms.

It's a little late to worry about viewing such signals, after you've already purchased an analog scope.

 

[DustinB]

Ouch, well that sucks.

 

[Nigel Goodwin]

Ouch, well that sucks.

For most purposes an analogue scope is fine, and for some purposes probably better? - really it's good to have BOTH!

 

[DustinB]

I figured it will work for most anything, but I had a problem recently trying to tell if a chip was initializing. The recommended method was to put a scope on it and see if it pulsed twice to indicate initialization. I purchased the scope before realizing it had to be a digital scope.

I guess I'll use this until I start EE school and come up with some money.

 

[BrownOut]

You should still be able to test it even with the analog scope. Set up your vertical amplifier for the measurement, and then choose "normal" trigger mode and "single shot" on your trigger options. Make sure the level is set to fire on the pulse amplitude ( may need to experiment a little here ) Then turn up the CRT brightness and start up your chip. If the pulses are there, the scope will trigger ( you'll see the light for single shot, which came on when you chose this option, extinguishes when the scope triggers ) If the brightness was turned up, you might see the pulses, depending on the timebase setting. It might help to be in a dark room. Experiment a little with your scope setup and see if you can capture the waves.


Don't blink though

 

[crutschow]

How fast are the pulses?

Years ago, before digital scopes, I looked at fast single-shot pulses on an analog scope by forming black construction paper into a tube and taping over the CRT to block ambient light. Then you turn up the CRT brightness, as BrownOut suggested, place you face over the tube to block as much light as possible, and observe the pulse. The CRT phosphor generally has enough persistence to show a faint outline of the pulse for several seconds after it passes.

 

[Nigel Goodwin]

The classic way (in the distant past) was to use a scope camera to take a picture of the pulses, and was why scopes had a 'single sweep' mode (where it did one single sweep only on receiving a trigger pulse).