Design Con 2015
Breaking News
Blog

First the Basics, Then the Complicated Stuff

NO RATINGS
1 saves
View Comments: Newest First | Oldest First | Threaded View
<<   <   Page 2 / 4   >   >>
GSKrasle
User Rank
CEO
Re: Widely forgotten, never taught, or just unappreciated...
GSKrasle   6/18/2014 5:04:14 PM
NO RATINGS
boblespam,

Exactly right! Those questions are there, but white font.

I use this not only when talking about the basic concept of wattage not being a linear function, but also to illustrate the importance of doing sampling correctly, particularly the need for (as close to as possible) simultaneous sampling. Also, why a switching regulator is more efficient than a linear one, and why "peaky" current draw from a battery with internal R loses efficiency (sometimes, an LDO actually gives longer life in e.g. memory hold-up).

I can send you the original visio, but it's pretty effective when just sketched-out.

boblespam
User Rank
CEO
Re: Widely forgotten, never taught, or just unappreciated...
boblespam   6/18/2014 5:13:45 AM
NO RATINGS
GSKrasle: I guess the hidden questions are what is Wmin and Wmax ?

Nice quizz, I will probably use it (with your permission) to demonstrate my customer why he should synchronize his software to measure U and I from a traction battery to evaluate the energy coming in and out of the battery for a State of Charge calculation.

AZskibum
User Rank
CEO
Re: "That's the only way it would compile"
AZskibum   6/18/2014 12:22:20 AM
NO RATINGS
the student had commented out the entire program, which is why it didn't produce anything.  When he pointed this out, the hapless student said: "But that's the only way I could get it to compile!"


The digital hardware designer's corollary to that is the newbie who thinks that after he finally gets his Verilog code to compile, that it's likely to do what he wants it to do!

GSKrasle
User Rank
CEO
Widely forgotten, never taught, or just unappreciated...
GSKrasle   6/17/2014 10:40:14 PM
NO RATINGS
And I'm not referring to myself, no matter how applicable....

 

I have been faced many times with a few of issues for which I appear to be the only drummer. I've seen then sink projects, indirectly cause the loss of incalculable wealth, and I've even lost jobs, either because I was not "convincing enough" or "too insistent." I'm sure lots of other folks have encountered these too, but it's all the more frustrating as I KNOW they were part of the curriculum! 

Number one is Nyquist. If you don't give him his due, his hoary ghost will invariably bite you in the butt. So many times I see data acquisition systems with no provision for an anti-aliasing filter, and often, those same systems are the targets of earnest but expensive and ineffectual "improvement." No, going to a 16-bit converter or a 1MHz rate won't help if you have broadband noise out into several MHz. A cheap capacitor is often all you need. Yes, it will attenuate high-frequencies, and slow the edges, but that is what it is SUPPOSED to do. A DAQ system is not an oscilloscope. And don't say to me "well, then, why doesn't a SCOPE need an input filter?"

Number two is dynamic range: a 12-bit converter with a +/-10V range should be preceded by an amplifier (or be switched to a diffferent gain) if you expect to measure a 1mV signal with any precision!

Watts are not linear. You can't filter (average) the I-signal and the V-signal and THEN multiply them! Here is a "quiz" I give to candidates, and sometimes to interviewers:

 

There's no trick, no esoteric knowledge, no calculations necessary, but 80% of engineers make the same mistake, that is, unless i "turn on" the two "hidden" questions.

Number five: expensive wire is only worth it if you actually USE the shield. Shields aren't magic, and attention to grounding goes a long way in improving measurements!

Just because two connectors feel like they hang-togeter is not sufficient to conclude that they are the appropriate combination to use. Manufacturers list compatibilities in their datasheets for a reason!

Protect your inputs! Just because the input impedance is listed as high (CMOS) doesn't mean you can neglect the case where the device is unpowered but connected to something that IS; "nobody would EVER unplug the tester while the DUT is still ON" is a cop-out.

There are lots more, but I need to chill-out a little. &^(**^()*^&$$!

antedeluvian
User Rank
Blogger
Missing in action
antedeluvian   6/17/2014 4:42:04 PM
NO RATINGS
I apologise that at I not supporting my own thread. Unfortunately life has intruded. I will try to contribute as time allows.

Max The Magnificent
User Rank
Blogger
Re: Things everyone should know!
Max The Magnificent   6/17/2014 11:25:32 AM
NO RATINGS
@Randa11: When hanging a door get a set of shorter and thinner screws than the ones you eventually intend to use.

That's a good one!

Max The Magnificent
User Rank
Blogger
Re: First post!
Max The Magnificent   6/17/2014 11:24:13 AM
NO RATINGS
@Andrew: ...mostly building on my involvement in my college's IEEE chapter...

That's a good thing to do -- keep it up!

Max The Magnificent
User Rank
Blogger
Re: Observe that no current will ever flow. We all can see that, right?
Max The Magnificent   6/17/2014 11:22:36 AM
NO RATINGS
@JCreasey: ...the magic smoke that accompanies it...

Ah, the magic smoke ... how often I recall applying power to a prototype board and seeing a flash and basking in the acrid smell of magic smoke (and once it's escaped from a component, it's a bugger to get back in :-)

Randa11
User Rank
Rookie
Things everyone should know!
Randa11   6/17/2014 6:19:48 AM
NO RATINGS
How about you? Have there been any occasions when you discovered that you'd forgotten some fundamental concept? A more complete subject for this post would be " Things everyone should know, but, no one ever told me, and, it was too late when I realised." 1. Turn TRVs (Thermostatic radiator valves)up to max during the summer, when the central heating is off. This initially counter-intuitive idea is obvious when you think about it. During the summer the temperature in the room gets much higher than it does when you initially turn the valves to minimum in the spring. At the spring temperatures the valve is already trying to turn the flow off completely; in the heat of the summer the valve tries even harder to turn the flow off, resulting in a jammed or broken valve. 2. When hanging a door get a set of shorter and thinner screws than the ones you eventually intend to use. if you're a complete amateur like me you'll end up putting those screws in and out lots of times before you get everything right. Using smaller screws initially is a lot easier on your wrist, and does far less damage to the wood.

AndrewAPTemple
User Rank
Rookie
Re: First post!
AndrewAPTemple   6/16/2014 10:50:00 PM
NO RATINGS
Maybe you will be able to attend some of these conferences after you've finished being a student. Where abouts are you based?

Hopefully. Still mostly building on my involvement in my college's IEEE chapter. I currently live in Philadelphia, where I also go to college. 

<<   <   Page 2 / 4   >   >>
Radio
NEXT UPCOMING BROADCAST
EE Times Senior Technical Editor Martin Rowe will interview EMC engineer Kenneth Wyatt.
Top Comments of the Week
Like Us on Facebook

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
EE Times on Twitter
EE Times Twitter Feed
Flash Poll