Electrical Engineering

A place to talk with friends from the EE community about vacuu...
Mar 11, 2020 15:59
@ThePhoton Punched tapes. Yes. Standard stuff.
Mar 11, 2020 15:53
(Nostalgia mode) That pen plotter sounded like some futuristic albeit frenzied music when it was operating: The motors and ribbed pulleys made quite interesting sounds.
Mar 11, 2020 15:52
Also, a few years after my first experience with a camshaft sequencer, a somewhat nutty family friend figured out a way of using the IO cards on a FORTRAN computer to "design" the layout for an advertising sequencer's camshaft. Early CAD/CAM with limited resources, I suppose: They had a pen plotter at the same university lab in those days, so it's not as though they couldn't have programmed their computer to print out a diagram.
Mar 11, 2020 15:47
@ThePhoton If you had 1000 individually controlled lamps, easier to use a camshaft instead of relays. The same skills were involved in making those cams, as in creating the rollers for mechanical musical organs.
Mar 11, 2020 15:43
So I'll let you in on the secret of why I know this camshaft stuff: The light sequencers which would become obsolete in the USA or parts of Europe were often bought at scrap value and configured for use in advertising signs in India. They were operational here for perhaps decades after they had become antique curiosities in your part of the world.

When I was in middle school, I got a chance to fiddle with one of those setups belonging to a family friend. It had stopped working - merely because nobody had replaced the brushes for a decade or more, and they'd worn down.
Mar 11, 2020 15:38
@ThePhoton :-) :-) :-D :-D
Mar 11, 2020 15:19
The bulbs and even the neon tubes needed frequent replacement, though - leading me to suspect that nobody was really bothered with interrupting current at zero-crossing.
Mar 11, 2020 15:17
I'm about to appear very aged by this comment:
The popular camshaft sequencer designs used the camshaft as the common, and fairly robust springy "brushes" as the contacts.
The advantage used to be that the contact brushes and camshaft were both self-cleaning, just by the friction and its polishing effect.

One big DISadvantage of high power camshaft-sequenced advertising signs was that, depending on the power involved, the repetitive arcing could ruin AM radio reception for everyone in the vicinity.
Feb 27, 2020 19:42
@Fermiparadox 10 KHz PWM is pretty trivial for any LED I've ever used. I've driven open-air LED / photodiode pairs up to 200+ KHz with just deadbug wiring.
Feb 27, 2020 19:41
Jan 29, 2020 05:08
Cluck cluck.
Jan 24, 2020 17:03
@NickAlexeev Yeah, TI has some amazing ULV motor control parts.
Jan 23, 2020 17:07
Perhaps this one from TI might fit your requirement: ti.com/lit/ds/symlink/drv10964.pdf
Jan 23, 2020 16:45
@NickAlexeev Hmm, I can't identify the TI parts for this. However, ST has the STSPIN250 etc, which seem suitable: st.com/content/dam/technology-tour-2017/…
Jan 23, 2020 16:32
3V motor and some kind of single-chip PWM + speed control part would be great. If you can fit it in your budget, TI has some low-cost, low-current single-chip motor controllers IIRC, that do the PWM, back-EMF readout (including readout of measured speed via SPI, I believe), and even bells and whistles like motor speed ramping.

Next choice, 3V motor, PWM for speed control, and back-EMF

That is unless your device has a boost convertor already used for something else, and you have an underutilized 6V / 9V rail with power to spare :-)
Jan 12, 2020 09:26
While I don't have the answer about your specific diode, often two manufacturers, and even two "vintages", of an identical diode, will have different epoxies used: Different thermal conductivity, more advanced polymers etc., and the physical package dimensions will change.

Also, thermal insulation is not a desirable parameter of a diode's package epoxy, thermal conduction is: You want the heat from the junction to dissipate out as fast as possible - otherwise that's how thermal runaway happens.
Jan 11, 2020 04:17
@TonyStewartSunnyskyguyEE75 Oh, she is a brilliant model. Perfect posing, timing, at such a young age. I loved working with her!
Jan 11, 2020 04:04
Ahh, cool!
See also: https://www.instagram.com/anindoghosh/ <-- More of my commercial photography work.
Jan 11, 2020 03:55
@TonyStewartSunnyskyguyEE75 Which site did you see?
Jan 11, 2020 03:54
@TonyStewartSunnyskyguyEE75 Hi, and thank you!
Jan 11, 2020 03:49
@ScottSeidman Likewise! :-)
Jan 10, 2020 17:44
@W5VO Umm hold on a second. I seem to remember that old-style through-hole diodes used to be made with thermally conductive potting epoxies, almost invariably. Thermally insulating epoxy would cause the junction to overheat pretty quick. Is that not the case anymore?
Jan 10, 2020 17:35
@W5VO Thank you, W5. Means a lot to me to hear that.
Jan 10, 2020 17:34
@ThePhoton Those, plus the press kit hand-outs of C-level and board members for various press events, plus cover images for company annual reports, that kind of stuff.
Jan 10, 2020 17:33
Now here we go, tempting me to do some of my destructive testing ;-)
You guys remember the number of times I blew up LEDs for testing such hypotheses
Jan 10, 2020 17:32
@W5VO I've downgraded to no EE at all, for a few years now. However, the addiction to technology pulls me back in!
Jan 10, 2020 17:31
@W5VO Fair point on the epoxy.
Jan 10, 2020 17:31
Hi Photon! I'm thrilled to see all of you familiar nicks!

I've been building a career in corporate portrait photography. Mixed bag so far, but it's gathering momentum.
Jan 10, 2020 17:30
@W5VO My problem with an air-gap would be the loss of thermal sinking through the board copper - Air is a darn good insulator :-)
Jan 10, 2020 17:28
@W5VO Heyyyyy W5! It's great seeing you! Yes, it has been ages! How is the EESE gang doing?
Jan 10, 2020 16:03
In my experience, TI has been excellent in supporting older ICs for manufacture purposes, and to an extent even for small volumes. When they discontinue something, the notifications have been years in advance.

I haven't tested this in the last 3-4 years, though, as I was busy with non-electronics things.
Dec 17, 2018 10:27
@abdullahkahraman Makes sense. :-) I went through something similar.
Dec 17, 2018 10:26
@NickAlexeev Perhaps, like me (back when I dropped off), he found something more compelling to do with his time. I can empathize with that.
Dec 16, 2018 13:26
@AjayMishra Sometimes :-)
Dec 15, 2018 17:17
@abdullahkahraman And what happened to Steven?
Dec 15, 2018 17:17
@Marla Wait, what? Olin was suspended, why?
 

 The Renderfarm

A place to talk while we wait for our renders to finish. Site ...
Dec 14, 2018 14:08
@gandalf3 Thanks! I'm going to fudge around a bit with "fatten"
Dec 14, 2018 13:44
Exactly.
That's why I find the approach taken by Keyshot 8 to be rather elegant. Are the pitchforks about to come out now? :-D
Dec 14, 2018 13:43
The more complex the original shape, the more complicated it becomes to create the containing shell for the collision physics, if I am analyzing the problem correctly.
Dec 14, 2018 13:42
@gandalf3 Yes, that is so, but that's where our discussion started: How do we constrain those particles to be conforming to a certain outer geometry, without adding a collision geometry / containing shell of a certain minimum thickness.
Dec 14, 2018 13:39
Ideally, I would be looking for the exact same mechanism Keyshot uses, i.e. a duplication of a shape, then replacing the material with a volumetric geometry transform into particles of some other material and shape. Is there some such approach available in Blender?
Dec 14, 2018 13:37
@gandalf3 I am toying with a slightly different solution: The do-not-render collision object can be an outer shell to my desired shape: Basically a surface extrusion from all faces, of dimension slightly bigger than particle size. That should allow successful collision detection, while allowing the particles to exploit the entire target object volume.
Dec 14, 2018 13:34
@gandalf3 Oh, no, that isn't the shape I am working towards - My upload was a far simpler model a friend made. I'm working on shapes like partly hollow star shapes (the pendant necklace on that model).
Dec 14, 2018 13:23
@gandalf3 I suspect the 0:0:0 point of each particle is tested against the collision object's envelope, so Monte Carlo dictates that some cases will spuriously find the particle to be outside, if the particle is big enough - after which the particle floats off unconstrained :-D
Dec 14, 2018 13:14
I'm really sorry. I might be wasting your time on an unsolvable problem, and that is not my intention.
Dec 14, 2018 13:13
So the thin sharp bottom part of that skirt shape, for instance, would be devoid of any particles.
Dec 14, 2018 13:13
@gandalf3 Umm collision physics requires my collision wall thickness to be greater than maximum particle dimensions, unless I'm misreading something. Right?
Dec 14, 2018 13:11
OK, shrunken duplicate approach isn't giving me joy - the shape I want to use has concave portions, and scaling down causes the inner surface of the duplicate part to breach the concave surface of the "glass" part. Of course, I could use a "shell" type boolean to shrink proportionately on all surfaces, I suppose, but that would cause thinner portions of the shape to not have any particles at all.

Ugh, my head hurts.
Dec 14, 2018 13:09
I guess my question is not ideally suited for the main site, as it is becoming what is called a "unicorn question" on EESE, as I go along: The nature of the question is changing as your answers expand my mind :-)
Dec 14, 2018 13:07
@gandalf3 Wait, how do I make the particles remain contained within a complex shape after emission, other than using a "do not render" invisible container? From your suggestion, it appears that is possible.