real word reasons to use an oscilloscope as a real tool?

[tggzzz]

...whacked it into the breadboard...

And there's the source of your problem.

Congratulations on battering this particular instance of your circuit into temporary submission.

[madires]

...whacked it into the breadboard...

And there's the source of your problem.

Congratulations on battering this particular instance of your circuit into temporary submission.

One of my first thoughts was that it's a breadboard problem. But I haven't seen any oscillation and the problem with breadboards is that they got too much capacitance between the contact strips. If you got a good explanation, I'd happy to hear it

[smbaker]

That is why I am so eager to see real world examples.

About a decade ago, I put together a CNC sherline milling machine, and I implemented a joystick option so that I could manually operate the milling table at a specific speed. I used the joystick to control direction and a 555 timer to control speed. As I turned up the speed control, the milling table would start to move erratically. I double and triple-checked the 555 timer circuit to try to find my mistake.

Finally, I got the scope (a B&K precision 20Mhz analog) out and probed the circuit in a few places. It became immediately obvious what was wrong -- I had selected too small a filter capacitor on my power supply. There was nothing wrong with the timer circuit itself, rather it was the power feeding the circuit that was the problem.

These days I often reach for the scope instead of the mulitmeter, just because it provides more information.

Scott

[tggzzz]

...whacked it into the breadboard...

And there's the source of your problem.

Congratulations on battering this particular instance of your circuit into temporary submission.

One of my first thoughts was that it's a breadboard problem. But I haven't seen any oscillation and the problem with breadboards is that they got too much capacitance between the contact strips. If you got a good explanation, I'd happy to hear it

That's only the least of it.

Much more important in high speed circuits is too much inductance, and dynamically variable resistance.

Work out the inductance of your wires (assume 1nH/mm) the change in current that flows during a transition, the transition time (signal repetition time is completely irrelevant), and plug the values into the elementary equation V= L^di/_dt

Work out the inductance and capacitance, and calculate the resonant frequency.

When you have correctly concluded that high speed circuits are unlikely to work reliably on breadboards, investigate the other techniques that work adequately.

[Ampere]

I'm working on an Arduino project which relies on two PWM signals. The duty cycle of both signals and ratio between the two duty cycles are controlled by their own respective potentiometer. I.e. The first pot adjusts the duty cycle of both from 25%/25% to 50%/50% and then the second pot changes the duty cycles from 50%/50% to 50%/75% or 25%/50%.

I might be a beginner, but I would think that even a pro would test the circuit how I did. I connected a probe to each output and watched how the signals changed in response to the inputs. I guess I could have used my DMM to monitor the duty cycle, but then I would have needed another DMM and I probably wouldn't have noticed the bug that occurred when the duty cycle rolled over past 100%.

[vk6zgo]

People will sometimes go to extreme lengths to avoid using an Oscilloscope.
At one employer,we made vehicle simulators.

The simulator's display had a "Test" function where you could check all the control settings & adjust them during initial set up.
Several of these controls output a PWM signal which was integrated passively,fed to a couple of Op Amps followed by an ADC* which was part of a "Mainboard" which interfaced with a PC.

The Op Amp circuit had adjustments of voltage range & offset,which had to be" tweaked" to both "centre" their output on,& make best use of, the ADCs range of acceptable inputs.
 
The standard way of setting these up was to look at the "Test" display on the PC screen .
This involved "dancing" the adjustments in & out till they were centred on,& just filled the display window.(which was a bit like the "pretend" analog scale on a DMM.)
I said:-
"Why not set one unit up in the "classic" way,then look at the signal into the ADC with an Oscilloscope,keep a record of the voltages,& preset all the others using a 'scope?"

They wouldn't even consider it,& continued wasting hours doing it the hard way!

* No,I don't know why they didn't just sample the PWM directly---I just worked there!

[dolphin42]

What have I used a scope for, engineering-wise?
* Ensuring that pulses have a fast enough transition in a remote sensing application, where a semi-passive input circuit supplies a small current over a lopng line in the presence of line voltage.
* Making sure that output stages are being driven properly. (solenoid drivers, switching supplies, audio amplifiers, motor drivers, etc etc)
* Watching communication busses for slew rate and glitches (and with the new capabilities of even simple digital scopes, doing simple protocol debugging - of course the new scopes can do a lot of heavy lifting there that used to take building circuits to decode the buss to watch it work)
* Checking that LCD drive signals really are always exactly 50% in simple arrangements, and with at least two traces, depending, checking the switching voltages on a muxed one.
* Simple logic analysis of a digital circuit, is the clock alighed properly with the signal to be clocked in? Does the PWM go all the way from off to on as you vary the range?
* In any servo circuit (motor control, power supplies of the regulated persuasion, etc) is the contro,. of the circuit calm and controlled? Or is it unstable and oscillating around the center of the control voltage, and losing energy to switching losses that it needn't)
* When any sort of inductive load, are the freewheeling clamps doing the job of keeping the output devices from blowing up? (A more general case I've already stated above indirectly)
* Are unwanted signals in a circuit from adjacent signals? Parallel wires will tend to couple a signal in one wire onto other adjacent ones.
* Is power supply instability causing the circuit to misbehave when loaded to its' maximum?
* Where is signal loss happening in a chain of amplifier stages? (Less of a problem in the IC age, perhaps but still a valid problem)

Anytime I develop a circuit, I just "have to" watch it work. If I can't see it working, even if it works right off the bat, I get a little twitchy about it.

And once you have enough experience, to know what a circuit should be doing, the same thinking applies. And when you get a broken circuit that used to work, while static checking is often faster (e.g. are the caps open in a switcher, is often the best first test, with the circuit off) sometimes you gotta see it working in real time.

(THis is my first post in this forum, but I can't resist chiming in LOL even if I'm mostly repating things others have said)

[Smokey]

...   At first I thought it was just good to have because they look cool and girls go crazy over guys with them!  ...

No, no no... It's not the scope.. The girls love the MULTIMETERS!

[fivefish]

Wait... a scope is a real tool for doing useful work?

I thought the point of buying a scope is so that we can hack it and then find all the bugs and imperfections, weird behavior, freezing scenarios, obscure combinations and permutations of switches and settings that will cause it to crash so we can have something to talk about.

[grifftech]

Is this guy for real ? he obviously has neber taken a TV apart from 30 years ago and found a diagram in it that showed oscilloscope waveforms to loof for on different points and even gave you a couple of alternative waveforms with a description of the problem they indicated. I could not live without a scope, it's like having eyes in your circuit.

I think that you mean never .

[crispy_tofu]

...   At first I thought it was just good to have because they look cool and girls go crazy over guys with them!  ...

No, no no... It's not the scope.. The girls love the MULTIMETERS!

Those probes are in the wrong jacks... 

[HighVoltage]

If you're trying to attract girls, I'm 99% sure an oscilloscope isn't going to help you. Try a motorcycle, or a little knowledge / interest in silver jewellery.

Well, I had one 23 year old German girl working in my lab and she loved the scopes as much as I do. Even asked if she could take one home to play with. May be there are more of them out there but I have only met one so far.
This is a funny thread!

[madires]

Much more important in high speed circuits is too much inductance, and dynamically variable resistance.

Work out the inductance of your wires (assume 1nH/mm) the change in current that flows during a transition, the transition time (signal repetition time is completely irrelevant), and plug the values into the elementary equation V= L^di/_dt

It's around a few mV in my case. This shouldn't be the cause.

When you have correctly concluded that high speed circuits are unlikely to work reliably on breadboards, investigate the other techniques that work adequately.

I don't disagree, but 1MHz digital I/O isn't particularly high speed. A 74HC4050 used as unidirectional level shifter works fine with the same breadboard and same circuit. I've had circuits on the breadboard running around 10MHz without much issues. Though I wouldn't go much higher

Anyway, identifying or specifying the maximums of a breadboard is a good application for a scope.

[Smokey]

...Those probes are in the wrong jacks... 

That's what she said?

[tggzzz]

Much more important in high speed circuits is too much inductance, and dynamically variable resistance.

Work out the inductance of your wires (assume 1nH/mm) the change in current that flows during a transition, the transition time (signal repetition time is completely irrelevant), and plug the values into the elementary equation V= L^di/_dt

It's around a few mV in my case. This shouldn't be the cause.

Show your working, including length of wires, impedance (not resistance) of loads, type of driving logic, transition times.

When you have correctly concluded that high speed circuits are unlikely to work reliably on breadboards, investigate the other techniques that work adequately.

I don't disagree, but 1MHz digital I/O isn't particularly high speed. A 74HC4050 used as unidirectional level shifter works fine with the same breadboard and same circuit. I've had circuits on the breadboard running around 10MHz without much issues. Though I wouldn't go much higher

Since you quote "1MHz", you clearly don't understand the issues. The problems will manifest themselves at 100kHz. Or 1Hz. Or 0.1Hz. Let me put it more directly: what do you think is the highest frequency present in a 1MHz digital signal, or a 0.1Hz digital signal?

Please check you answer by reading and understanding http://www.sigcon.com/Pubs/pubsKeyword.htm#rise%20time

[madires]

Since you quote "1MHz", you clearly don't understand the issues. The problems will manifest themselves at 100kHz. Or 1Hz. Or 0.1Hz. Let me put it more directly: what do you think is the highest frequency present in a 1MHz digital signal, or a 0.1Hz digital signal?

I think you misunderstood me, I'm very well aware about rise times, pulses and Mr. Fourier.

[crispy_tofu]

...Those probes are in the wrong jacks... 

That's what she said?

Red is in the 10A jack, black is in the volts-ohms jack. 
What about the common?

[smbaker]

It's good you clarified, otherwise we could have misinterpreted and thought you had an unusually long thumb.

[c4757p]

Subtle as always, timb

[Simon]

...   At first I thought it was just good to have because they look cool and girls go crazy over guys with them!  ...

No, no no... It's not the scope.. The girls love the MULTIMETERS!

Have a look around ebay and look at how crazy the chinese ones go for power inverters, a 5KW one can get you up to 3 girls drooling over it at once.

[timb]

It's good you clarified, otherwise we could have misinterpreted and thought you had an unusually long thumb.

I do. It's both! There was this industrial accident and... It's a *long* story. You know what else is long? My pen is. It's hard to write with sometimes. You know what else is hard? Diamonds.


Sent from my Tablet