top 10 uses for Function Generators

[k8943]

Suggest video on Top10 uses for function generators for someone who “does a bit of everything “.

Haven’t found anything useful on this anywhere.

As someone who does quite a bit of electronics but still doesn’t have one would like to have a better idea of what I’m missing rather than spend too much on a whim then have it sitting under-used on a crowded bench.

Help Dave!

[David Hess]

Top use?  Trade a function generator to the building's superintendent so that he will tell you how tall the building is and you can keep the neat clock.

[Berni]

The only commonly used test gear in general purpose electronics is a multimeter, scope and power supply.

Other gear tends to be indeed more rarely used, but when you need it it really saves you time and effort. A signal generator is only one such non vital but super nice to have when needed kind of gear. Other such gear might be a spectrum analyzer, network analyzer, synthesizer, LCR meter, electronic load, SMU, electrometer, curve tracer etc.. all the way to more exotic things like distortion analyzers, semicoductor analyzers, BERTs, pattern generators, atomic clocks etc..

Getting back to function generators they are mainly useful for providing input stimulus to a circuit under test. Could be anything from testing a transistor to sticking a PWM signal into  some module, sweeping the frequency response of something, put test tones trough audio gear etc... Its a very wide application piece of gear, but will certainly not be used as often as a scope.

[Cnoob]

Although Berni covered it.

I will just like to stress the points useful to test the AC Band width of DMM and performance of your scopes.

[tggzzz]

Top use?  Trade a function generator to the building's superintendent so that he will tell you how tall the building is and you can keep the neat clock.

I thought it was a barometer, but maybe that is just an old version of the story before atomic clocks became easily portable

At a job interview after university I was once asked how, if I was an ancient Egyptian, I would be able to tell when the people building the pyramids could tell when their shift had ended. I spent and enjoyable 10 minutes before getting around to mentioning sand clocks / egg timers.

[tggzzz]

Suggest video on Top10 uses for function generators for someone who “does a bit of everything “.

The video after that might consider top 10 used for a car.

[DimitriP]

...
10) connect it to the newly acquired oscilloscope and watch the waveforms while pondering  top 10 uses for an oscilloscope

[k8943]

Haha, you guys...!!

Should have known better than not to expect replies on a suggestion thread!

But STILL.... Have the impression that whenever make something the heavy lifting is done by reference designs and app notes such that haven't had to concern myself too much with whether some component might not be tweaked or the details of what is going on. Maybe it's better that way. But whenever do put something under the magnifying glass a whole universe appears and in light of that wondering what might be able to reveal with an AWG. When the book comes out, let me know!

[AndyC_772]

I will just like to stress the points useful to test the AC Band width of DMM and performance of your scopes.

If the only thing you're doing with test equipment is connecting it to other test equipment, you're doing it wrong.

It reminds me of photography forums I used to be on...people would post photographs of their new camera, not taken with the camera.

The rule for when to buy a tool is very simple: if and when you need it, buy it. If you don't, then don't.

(That said, I did once have a visiting customer comment that I "didn't even have a function generator", which was true at the time, because I had no need for one. That's no longer the case, of course, but maybe I should stock the lab with cool-looking but non-functional props just in case?)

[k8943]

Know very well what you mean about the camera thing.

I have one or two things I'd like to do with an AWG (most especially sweeping wherever there's a choke, a resistor capacitor pair etc.)  but have the impression you need to have had a career in electronics (silicon wafers in the gaps between your teeth etc) to have an intuitive sense of the gamut of possibilities. Besides thought Dave was running out of ideas for videos.

[Berni]

Well if you can't come up with any uses for a new piece of test gear then you most likely don't need it.

Because everyone works on different kinds of projects means that they all have different needs for test gear. Even just the choice of a scope can vary a lot. Someone might be just fine with a old 20MHz Hameg analog scope, someone else wants a Rigol 1000Z because they need one shot capture capability but not much more, someone else might need a Keysight DSOX3000 for the update rate and 1GHz bandwith, maybe someone else needs 8GHz of bandwith and advanced math features instead so they go for a Keysight S-series, or maybe you need 100GHz of bandwith and go for the massive Lecroy scope.

Tho collecting test equipment can sometimes become a bit of a hobby in itself. Mainly because you sometimes need extra equipment to be able to test and repair whatever you just bought. Can be a bit of a expensive hobby sometimes, but you do also learn quite a bit while repairing test gear on a component level. Buying old broken test gear and repairing it can also be a cheap way of outfitting a lab with high quality gear.

[k8943]

I guess that's it then.

Will carry on without until the need hits me over the head (or until someone makes an inspirational video for those poor fools who've bought them already & don't know what to do).

In the meantime, love using the miniVNA pro.... I guess that's an AWG of sorts!

[David Hess]

Top use?  Trade a function generator to the building's superintendent so that he will tell you how tall the building is and you can keep the neat clock.

I thought it was a barometer, but maybe that is just an old version of the story before atomic clocks became easily portable

It depends on the class.  The last time I took physics, it was a perfectly spherical clock instead of a barometer.

...
10) connect it to the newly acquired oscilloscope and watch the waveforms while pondering  top 10 uses for an oscilloscope

See below.

Will carry on without until the need hits me over the head (or until someone makes an inspirational video for those poor fools who've bought them already & don't know what to do).

I consider a function generator to be a basic piece of test equipment.  If you have two of them, then they can be combined to become a swept function generator.

[bsudbrink]

I seem to recall another thread...  Ah!  Yes, here:

https://www.eevblog.com/forum/beginners/real-world-use-of-a-function-generator/

[tggzzz]

Top use?  Trade a function generator to the building's superintendent so that he will tell you how tall the building is and you can keep the neat clock.

I thought it was a barometer, but maybe that is just an old version of the story before atomic clocks became easily portable

It depends on the class.  The last time I took physics, it was a perfectly spherical clock instead of a barometer.

I'm probably confusing it with a spherical cow.

[bsudbrink]

I'm probably confusing it with a spherical cow.

Udderly ridiculous.   

[djacobow]

I notice new function generators do not work nearly as well as old ones at holding open doors. Kids today and their crap build quality!

[schmitt trigger]

True story:
Long ago, the company’s general manager liked to walk into the lab and peer into work benches.
Not being an engineer, he would make pesky questions.

One way we would deal with him is to connect an oscilloscope to an AWG whose output would be the wackiest we could generate. In addition, a DMM or frequency meter would be a nice addition. Plenty of paper schematics strewn around, some breadboards full of components and the show was fully set.

That on its own made having on the lab an expensive Agilent AWG worthwhile.

[k8943]

The video after that might consider top 10 used for a car.

Here's the thing: we could make a video on the top 10 uses of a car and not be so far from the mark.

Just that it wouldn't be of any interest to most people in western civilisation who have grown up in families with cars. (Though it might nonetheless help them re-evaluate their purchasing priorities!)

1) getting to work (see car sharing);
2) shopping (foraging diverse locations and bringing back stuff that might be cumbersome to carry);
3) helping people to get places (old people, young people...);
4) getting out into the country (if you lived there you'd have one already - or a horse);
5) getting back late at night (see Uber);
6) you can sleep in it (and you should, if you're drunk);
7) win friends and influence people (certain kinds of people, certain types of car);
8 ) participate in a community effort to change the climate;
....

Don't know many who grew up in families with a waveform generators next to the microwave. Hence the video suggestion (which is obviously pretty dull for those who are already at the bar).

Now this been in the back of my mind for a day or two could probably - and probably will - sketch out a pretty decent list myself (full of newbie lack of perspective) but most won't bother and besides videos are FUN!!!

[tggzzz]

Top use?  Trade a function generator to the building's superintendent so that he will tell you how tall the building is and you can keep the neat clock.

I thought it was a barometer, but maybe that is just an old version of the story before atomic clocks became easily portable

It depends on the class.  The last time I took physics, it was a perfectly spherical clock instead of a barometer.

Ah, found my source from 1968, probably referring back to the late 50s. It wasn't in "Random Walks in Science", but the sequel "More Random Walks in Science".

[Brumby]

Like it!   

[Berni]

Hah nice one, makes a very valid point too.

This is still the way they teach these days, its encouraged to memorize what they want to hear from you rather than to actually think about the subject.

But at least at the university i went to i have noticed students prefer this way. Most classes ware structured around this sort of memorizing how to solve something, lab work had printed scripts to work along so it was almost like flowing a cooking recipe. But then in the final year there was one professor teaching sensor technology who did his lab work the other way around, on the first day of his class he walked in with a big cardboard box full of various sensors, asked everyone to make groups of 2 students and have each group pick a sensor from the box. The goal was to make the sensor do something usefully, this was generally regarded as getting the sensors output value into a PC and display it. Any way of doing it was allowed, but they are provided with PCB etching tools and PIC MCUs with debuggers if they want to use it. What followed is a solid few minutes of a room of students starring at a sensor in front of them and then a wave of questions what are they supposed to do. So the lecturer had to explain that they can google the partnumber on it to get a datasheet. That went reasonably smoothly since people of my generation know how to use the internet. What followed however was half an hour or looking at the datasheet followed by again a wave of questions what they are supposed to do. Eventually for some reason they got embarrassed of asking the professor for every single step and started showering me with questions and eventually even that stopped but all work on there sensor projects also seamed to halt, by the end of it everyone had half finished PCBs that didn't work or sorta worked. After all of it pretty much everyone was of the opinion this was the worse subject in all 3 years at this uni. While i found this being the most awesome lab work ever.

EDIT:
Oh and at the end of each year its actually mandatory to fill out a anonymous poll about each subject and its professor. There was some award for the best professor, but it was likely mostly used by the universities management to pressure the professors that students are unhappy about. So sadly that professor likely got a pretty awful score on that poll for his lab classes and probably won't be doing them in this open freestyle format much longer.

[bitseeker]

I remember students like this in the engineering program I went through. They're what I referred to as "textbook engineers." If it was in the textbook, they did well. If it required creativity or thinking about things other than what was explicitly written in books, they were lost. This included their senior projects, where most were published projects that I recognized from electronics magazines.

[AndyC_772]

I'm not entirely sure how I feel about this kind of project being set as part of an academic course. As some of you may know, I design sensor products for a living; sometimes these are an off-the-shelf component packaged up in a convenient way for the application, other times they're designed from the ground up starting with the underlying physics.

Learning how to google a data sheet, make sense of it and infer what supporting components a sensor will need is something which can be readily done in industry, and it's the sort of thing I'd fully expect to teach a junior engineer on the job. Being able to do it correctly is more about experience than theory, and you really don't need to take up precious course hours decoding some particular vendor's badly documented register set, or trying to solder a QFN without wrecking it. They're practical skills that I can teach someone quite easily.

In an academic course, I'd hope and expect that students will be taught the principles of operation of these sensors. How do they really work? What are the underlying physical processes going on? What factors influence their readings? What are the fundamental, theoretical limits on how well then can work? Why might a given type of sensor be a good choice in application A, but entirely inappropriate in application B? What knowledge and insight can a graduate bring to my company that I can't read for myself off a manufacturer's data sheet?

 

[tggzzz]

I'm not entirely sure how I feel about this kind of project being set as part of an academic course. As some of you may know, I design sensor products for a living; sometimes these are an off-the-shelf component packaged up in a convenient way for the application, other times they're designed from the ground up starting with the underlying physics.

Learning how to google a data sheet, make sense of it and infer what supporting components a sensor will need is something which can be readily done in industry, and it's the sort of thing I'd fully expect to teach a junior engineer on the job. Being able to do it correctly is more about experience than theory, and you really don't need to take up precious course hours decoding some particular vendor's badly documented register set, or trying to solder a QFN without wrecking it. They're practical skills that I can teach someone quite easily.

In an academic course, I'd hope and expect that students will be taught the principles of operation of these sensors. How do they really work? What are the underlying physical processes going on? What factors influence their readings? What are the fundamental, theoretical limits on how well then can work? Why might a given type of sensor be a good choice in application A, but entirely inappropriate in application B? What knowledge and insight can a graduate bring to my company that I can't read for myself off a manufacturer's data sheet?

I would expect a student engineer to:

• demonstrate that the basic physics/chemistry/mechanics of a device would be suitable for the task
• demonstrate using a chosen device to accomplish the task
• define why they did what they did, plus what they would do differently next time

I appreciate not all courses manage to achieve that. My degree course did, and from discussing the comparable course with students last year (open day on 40th anniversary of our graduation!), still does.

You need both solid theory and solid practice to be an engineer. No theory means you are a technician. No practice means you are an academic.

And to trap out a stale discussion, neither engineers nor academics nor technicians are better/worse than the others: you need all of them. Same is true for doctors and nurses.

[AndyC_772]

I completely agree, but bear in mind, someone just starting work needs to have a more academic bias if they are to end up a competent engineer after some time in industry.

If they start off with too little academic, theoretical knowledge, and then add pure on-the-job experience, then they'll become technicians - and may have missed the only opportunity to become the best engineers they could have been.

[Berni]

I'm not entirely sure how I feel about this kind of project being set as part of an academic course. As some of you may know, I design sensor products for a living; sometimes these are an off-the-shelf component packaged up in a convenient way for the application, other times they're designed from the ground up starting with the underlying physics.

Learning how to google a data sheet, make sense of it and infer what supporting components a sensor will need is something which can be readily done in industry, and it's the sort of thing I'd fully expect to teach a junior engineer on the job. Being able to do it correctly is more about experience than theory, and you really don't need to take up precious course hours decoding some particular vendor's badly documented register set, or trying to solder a QFN without wrecking it. They're practical skills that I can teach someone quite easily.

In an academic course, I'd hope and expect that students will be taught the principles of operation of these sensors. How do they really work? What are the underlying physical processes going on? What factors influence their readings? What are the fundamental, theoretical limits on how well then can work? Why might a given type of sensor be a good choice in application A, but entirely inappropriate in application B? What knowledge and insight can a graduate bring to my company that I can't read for myself off a manufacturer's data sheet?

This was just part of one semester of this subject. The lectures did go into the theory behind how all sorts of sensors work and there was lab work in characterizing and using sensors. But that part mostly followed the usual way of teaching. Tho the lectures did include good amount of experiments shown mid lecture. But what followed was this particular semester that has taken this different approach.

I'm not saying that all lab work is supposed to be done this way, but at least some of it should be done more like this. It prepares soon to be engineers for what actual engineering work is like. Engineers don't come to work and get a list of tasks to do, along with instructions how to do them. Instead they are given a task and that's it. Nobody constantly looking over your shoulder and guiding you, its part of your own job to figure out how to get the task done.

These "textbook engineers" end up being employed as technicians despite having a degree or ending up employed in a field that has nothing to do with electronics.

But if you don't teach them enough theory what you get instead is a "copy paste engineer". They get things done but all of the designs they come up with are actually a bunch of existing designs from the internet or literature slapped together. They have no clue how any of those design snippets actually work, but they know what the snippet is for and are smart enough to put them together in the right combination to do the job. Results in clumsy non optimal designs that still mostly work in the end.

[RoGeorge]

My Rigol DG4102 generator can output DC as well.  To my surprise, many times (if not most of the times) I am using it as an adjustable DC reference in the +/-10V range.   

I have a 3 channels digital power supply, too, but I like to use it only for powering the circuit.  For reference voltages I like to use the AWG generator instead, don't know why. 

[tggzzz]

I completely agree, but bear in mind, someone just starting work needs to have a more academic bias if they are to end up a competent engineer after some time in industry.

If they start off with too little academic, theoretical knowledge, and then add pure on-the-job experience, then they'll become technicians - and may have missed the only opportunity to become the best engineers they could have been.

Just so. Exactly. There are exceptions, but they are so rare as to be almost mythical.

And the most irritating such technicians won't even recognise that's what can happen or has happened - and will prosyletise thie world view.

Those considerations are why it is so helpful to find what a candidate has done in their spare time, since that neatly complements the coursework and illustrates what they think they are capable of doing.

[RoGeorge]

a perfectly spherical clock

What do you mean by that?

(Google doesn't fetch me any "perfectly spherical clock" either, well, it has something, but not physics related.)

[David Hess]

a perfectly spherical clock

What do you mean by that?

Perfectly spherical roughly means ideal and without extraneous behaviors or limitations, like the canonical "perfectly spherical cow".  So a perfectly spherical clock measures time perfectly, has no drag, rebounds with no energy loss, etc.

[NorthGuy]

Ah, found my source from 1968, probably referring back to the late 50s. It wasn't in "Random Walks in Science", but the sequel "More Random Walks in Science".

Interesting. The most advanced student managed to get "almost" full grade, almost as good as others.

[RoGeorge]

I don't get what's the fuss with that student.  There was no rope available there, and no ruler to measure the rope.

Zero points, no doubt.

[GregDunn]

The problem with waiting until "on-the-job" to teach practical stuff is that it leads to people like we (actually, the company - we strongly recommended NOT hiring contract labor for the job) got to handle some software development on a time-critical project a few years ago.

They had all apparently graduated with "software engineering" degrees, which meant they understood programming paradigms, could write long screeds in Java, and tell their favorite IDE to build an object file.  What they did not know was how to deal with a design which wasn't in their received wisdom book of tricks, how to invoke custom scripts in a different language, or how to debug something which wouldn't run on the IDE.  They also didn't know how to pass what they learned on to their successors when the 18-month contract ran out, so if one of us got transferred to a different project in the meanwhile, they were back to square one using our documentation (which they couldn't interpret in the first place).  We were forbidden to do any of their work for them, or supervise the handoff to the project - that was done by their local supervisors who knew even less than they did. As we had expected, it was a disaster.

So I support anything in an engineering curriculum which forces the student to adapt their theory to real-world situations as part of the completion of a course.  None of us in EE could have built a class project in the first place if we didn't learn how to read a data sheet, compare specs and try out components in a test rig.  One of the most important things we learned in school was that the part specs, the requirements and the actual components often occupied three different spaces, and part of the solution involved dealing with apparent contradictions and hopefully finding a conjunction of the sets.  If I hadn't already done that as part of my course work, I would not have been welcomed into a R&D environment to build some pretty cool hardware - the project managers didn't have time to run a remedial course for engineers.

[tggzzz]

The problem with waiting until "on-the-job" to teach practical stuff is that it leads to people like we (actually, the company - we strongly recommended NOT hiring contract labor for the job) got to handle some software development on a time-critical project a few years ago.

They had all apparently graduated with "software engineering" degrees, which meant they understood programming paradigms, could write long screeds in Java, and tell their favorite IDE to build an object file.  What they did not know was how to deal with a design which wasn't in their received wisdom book of tricks, how to invoke custom scripts in a different language, or how to debug something which wouldn't run on the IDE.  They also didn't know how to pass what they learned on to their successors when the 18-month contract ran out, so if one of us got transferred to a different project in the meanwhile, they were back to square one using our documentation (which they couldn't interpret in the first place).  We were forbidden to do any of their work for them, or supervise the handoff to the project - that was done by their local supervisors who knew even less than they did. As we had expected, it was a disaster.

So I support anything in an engineering curriculum which forces the student to adapt their theory to real-world situations as part of the completion of a course.  None of us in EE could have built a class project in the first place if we didn't learn how to read a data sheet, compare specs and try out components in a test rig.  One of the most important things we learned in school was that the part specs, the requirements and the actual components often occupied three different spaces, and part of the solution involved dealing with apparent contradictions and hopefully finding a conjunction of the sets.  If I hadn't already done that as part of my course work, I would not have been welcomed into a R&D environment to build some pretty cool hardware - the project managers didn't have time to run a remedial course for engineers.

I don't disagree with the above, but it then leads to the question of where the balance should be.

Even if it was practical to teach the problems which occur in large (time/people) projects, I don't think it would be desirable. But the students should certainly have a clue that "here there be dragons".

Similarly, I wouldn't expect them to understand how to build, say, high availability distributed systems. But I would expect them to be aware of the relevance of the Byzantine General's problem and the "eight fallacies of distributed computing".

[SparkyFX]

Suggest video on Top10 uses for function generators for someone who “does a bit of everything “.

The field is quite wide, so I don´t know if this would be Top 10, but anyway, besides testing other test equipment, like oscilloscopes and frequency counters
- OpAmp circuit input/output testing
- testing filters in general (although anything can be described as a filter function)
- slew rate testing of logic gates
- encoder simulation
- testing demodulation applications
- frequency source for microcontroller projects
- component testing

Of course you go by the datasheet in many of these cases, but in prototyping it might actually be useful to see where the limits are.

[David Hess]

My most esoteric use involves either my swept function generator or two of my normal function generators.  I create a swept sine wave and send the sweep trigger to my oscilloscope or the reverse.  Then one channel of the oscilloscope measures the function generator output and the other measures the output from whatever network I am testing.  Combined, that makes a low frequency VNA (vector network analyser).  Since the oscilloscope is operating with a triggered sweep instead of XY mode, measurements can be gated to pick a specific spot to measure.  With alternate delayed sweep, the measurement point is highlighted on the oscilloscope display.

[JPortici]

One of the uses of integrated AWGs:
- There is a signal that upsets my board (reveals a bug)
- Acquire portion of signal
- From scope's AWG menu select the option to create waveform from acquisition
- Select the range to taste / normalize
- Apply.
- Now you have your test signal, go on debugging!

You can do it with any AWG of course, that's the whole point.. but it is so convenient to do with those integrated in scopes.
Literally the only advantage over standalone function generators (well, syncronized triggers too)

[2N3055]

One of the uses of integrated AWGs:
- There is a signal that upsets my board (reveals a bug)
- Acquire portion of signal
- From scope's AWG menu select the option to create waveform from acquisition
- Select the range to taste / normalize
- Apply.
- Now you have your test signal, go on debugging!

You can do it with any AWG of course, that's the whole point.. but it is so convenient to do with those integrated in scopes.
Literally the only advantage over standalone function generators (well, syncronized triggers too)

I agree, same here. On Keysight 3000T you can also add noise to the signal easily to test for noise resilience on sensor input...

[Electro Detective]

A function generator and oscilloscope and other basic test gear will quickly verify Audiophool equipment specifications 

and save wood ducks claiming to have golden ears from weeping golden tears blowing a ton of money on nicely badged generic equipment,
with price mark ups that are beyond any moral or criminal integrity 

[RoGeorge]

This week only:

- searched for the resonant frequency of a NFC antenna located inside a Li-ion phone battery
https://www.eevblog.com/forum/rf-microwave/bamboozled-by-rx-voltage-nfc-antenna-resonance-(in-a-samsung-li-ion-battery)/msg2462037/#msg2462037

- improvised a Time Domain Reflectometer (TDR) to measure a coaxial cable's impedance, velocity factor and frequency response
https://www.eevblog.com/forum/rf-microwave/diy-50-ohm-bnc-cables-on-a-budget/msg2469279/#msg2469279