Because they'd have the book open on their work-bench while they were soldering. Children don't look after books well.
And the math can clarify your understanding no end.
No surprise there. Then John Larkin tries to tell us about "design".
-- Bill Sloman, Sydney
Ah, you read as well as you math, I see!
Tim
-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Design Website: https://www.seventransistorlabs.com/
g in the lab was a huge waste of time & did the project at home. From the l imited amount I've seen of unis they seem to miss out that cost effective w ay for students to learn. And they entirely fail to motivate students to do so.
Heh. We had to present working hardware. Only with software could we have d one that, and there wasn't much software on the course.
NT
ent approaches. I don't relate to the learn it all in uni then practice a l ittle style, and watched people that did look like rabbits caught in headli ghts when asked to design & make something.
ng in the lab was a huge waste of time & did the project at home. From the limited amount I've seen of unis they seem to miss out that cost effective way for students to learn. And they entirely fail to motivate students to d o so.
Today, yes. It wasn't affordable to me then though.
Interest is almost everything. Without that nothing much happens. As said, academic study without practice or interest is poorly understood & quickly forgotten.
So the question is how to drive interest. One solution is to read about lot s of circuits, and along the way the student finds something of interests & wants to build it. Keep building interesting stuff & the need for academic learning crops up along the way. That's basically how I did it.
Pre-10 I was picking up scrap TVs & trying to fix them, rarely successfully . It resulted in plenty of parts though.
All this is impossible in Bill's world view of course. That universities don't even suggest students build things at home says a l ot.
NT
I look forward to the day Spice can access enough compute power to sim thin gs much better. More often than not it fails for me. Last time I used it ju st recently it thought transistor V_be didn't drop at low current, making i t not useful. A lot of stuff I don't Spice at all, I understand it better t han Spice does. I'm sure there's lots Spice knows I don't, I just tend to o perate in a space where it doesn't & isn't helping.
Having said that my LTspice is probably out of date.
NT
I'm not dissing math, I just knew that level of math wasn't relevant to me. A level maths otoh has been very useful.
NT
come mostly obsolete, and goods made from just one mixed signal IC with pro cessing & versatile analogue building blocks in one lump. Those would elimi nate most of the waste & the chips could be reused for other jobs, so possi bly the eventual solution to e-waste will have nothing to do with agencies or legislation. Who knows.
Yabbut they won't be custom ICs, that's the trick. Do the same for mixed sy stems that has been done with microprocessors. Pick a general purpose IC of f the shelf & program it to get it to do both your digital & analogue biddi ng. It costs in silicon, but wipes out so much else.
Imagine an IC with a moulded in row of connectors & all the microprocessing & analogue your product needs. All you need do is program it. No box, no p cb, no discretes, nothing. Sometimes there is zero manufacturing process fo r you.
Fast forward another generation: the connectors are all the same, it doesn' t matter what you plug in where, the chip works it all out automatically. U sers no longer need to know or care.
NT
John Larkin wrote
Well, different country, and they were right I used to copy diagrams and stuff with a transparent and fountain pen. One day the fountain pen... OK huge blue spot on a page in the book. Had to bring it back 'sorry'..
Yes
The problem with 'simulations' is that you work very hard, to create your 'thing' and when you shut down that PC you have nothing...... In my school days there were no computers and simulations, you spend time doing math on paper, then building things and trying to measure and see what happened was much more fun.
That is scary.
krw wrote
? 1968 0r 1969 build my first scope, DG7-32 CRT, 1 MHz or a bit more, tubes, 10$ parts? Most parts from scrap TV chassis etc..1969 build my second scope, DG7-32 all transistor, RTL logic, 5 MHz or more... trigger delay, what not. Also had some TV scope long before that with a TV CRT and HV with a car ignition coil, magnetic deflection, audio range... parts maybe 5 $ (CRT was free). In my last year at high school I donated it to the physics teacher, but he was scared of it, the CRT then went to the local TV repair shop... LOL Was a great teacher BTW, we would stay after classes to calculate electron orbits in a magnetic field.
later, 1975 or 1976 I build a 300 MHz wide scope, cost? few dollars, 10, 30 maybe. somebody donated the CRT.
II had the CRT thing running on my bed, was a huge setup, all that EHT.
In the early seventies, color TV took up half the floor... Shortwave receiver whole table full... Same for the SSB transmitter. Then in the seventies PAL encoder and color video recording before anyone had one or could afford one. But then TV was my job...
tabbypurr wrote
That is much like an FPGA. You can then write your own processor in some HDL, or signal processing code
Dunno about connectors, there are many things like displays and sensors that require special wiring, some a lot of it.
Compute power is almost meaningless when the models are crappy, as almost all board-level SPICE models are.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC / Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics Briarcliff Manor NY 10510 http://electrooptical.net http://hobbs-eo.com
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC / Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics Briarcliff Manor NY 10510 http://electrooptical.net http://hobbs-eo.com
erent approaches. I don't relate to the learn it all in uni then practice a little style, and watched people that did look like rabbits caught in head lights when asked to design & make something.
king in the lab was a huge waste of time & did the project at home. From th e limited amount I've seen of unis they seem to miss out that cost effectiv e way for students to learn. And they entirely fail to motivate students to do so.
, academic study without practice or interest is poorly understood & quickl y forgotten.
ots of circuits, and along the way the student finds something of interests & wants to build it. Keep building interesting stuff & the need for academ ic learning crops up along the way. That's basically how I did it.
ly. It resulted in plenty of parts though.
Why?
lot.
If you want to be sure that student have built something themselves, it's n ice to be able to walk past from time to time while they are doing it.
One well known student sin is sub-contracting their essays to an essay-writ ing service. Anti-plagiarism software often catches that - the sub-contract essay-writers also cheat.
-- Bill Sloman, Sydney
ings much better. More often than not it fails for me. Last time I used it just recently it thought transistor V_be didn't drop at low current, making it not useful.
LTSpice routinely uses the Gummel-Poon model of the transistor. If the tran sistor Vbe didn't drop at low emitter current you were most likely putting current through the emitter that you hadn't noticed, but you might have plu gged in a strange Gummel-Poon model parameter - there are 64 of them.
User error is definitely a more plausible explanation than LTSpice getting that wrong.
s. I'm sure there's lots Spice knows I don't, I just tend to operate in a s pace where it doesn't & isn't helping.
Mine does remind me if it hasn't been updated recently.
-- Bill Sloman, Sydney
I lived at home in high school and first year of college, and had two rooms for my stuff. I left a small strip on my bed where I could sleep.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
Spice uses the standard diode and transistor equations, which you can tweak as needed. That's not a compute power issue.
What specific part did you use that didn't behave at low current?
LT XVII doesn't behave radically different from old versions, and the diode behavior is very likely the same.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
We have, and archive, the .asc files, with design notes documents. They are great to have available. We also document any related breadboards and first-article tests.
Why? The circuit doesn't care if I understand it or not. The customers don't either.
We make one all-analog discriminator, sort of a CFD, that locates the centroid of a roundish, sorta noisy pulse. The delay distortion of one section of the circuit cancels that of another. The cancellation was discoverer accidentally and tweaked in Spice, optimized for the expected range of input pulses. Works great.
My new high voltage pulse generator was similarly designed by instinct and simulation. It's still hard for me to understand.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
A lot of the gear in the EE lab was broken. The communal +100 volt supply had 20 v p-p ripple, and most of the guys didn't know that. Their amplifiers had interesting frequency response graphs.
I found the 3dB points on a scope, and then we left.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
Programmable mixed-signal chips have been marketed (Cypress?) but weren't very successful.
PCBs are about 100 years old, and they are still the best way to support and connect electronics. We just keep moving up the abstraction stack, soldering down opamps instead of transistors, FPGAs instead of TTL gates.
Soldering is even older.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
In my power-plane TDR, you can see the 50 ohm hardline and then, before the step hits the board edges, a bit that looks ohmic, about 1 ohm, and sorta flat with time. Is it flat? If the board were infinitely big, would this impedance remain constant?
(Ignore copper loss for this question, although it is significant on a real PCB. The black oxide adhesion treatment is ghastly for signals, good for planes.)
-- John Larkin Highland Technology, Inc lunatic fringe electronics
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.