OT: UK to move back to imperial units?

The mental math is analog, like a slide rule, not actual arithmetic. There are tricks, like crudely approximating a few constants and reciprocals and logs, that get within 25% or so of the right answer. That's close enough to decide whether it's worth grabbing a calculator or running Spice.

Like the old GR "lightning empiricism" game.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

Perhaps someone has finally bred a politician with a sense of humor?

John, esq.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

I do a bit of both. Memorizing a few facts helps too, e.g. for thermal vias, PTH copper is about 40-um thick and small vias (10 mil) are mostly copper.

I do a lot of "photon budgets", which are basically feasibility calculations mostly based on first principles and material properties. It's a lot easier to know how your design is doing when you know how good it could potentially be, so you know how close you are.

I've been working like a maniac the last couple of years, but once I get my third edition submitted, I'm hoping to string a bunch of photon budgets together into another book. It's a very useful skill that should be more widely distributed.

Yup. Wasn't that Philbrick? ISTR stealing the idea from a Pease column.

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

He's just being a Panteltje. We're used to it round here. ;)

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

We have debated here about whether to pave a copper pour with a lot of small vias or a fewer number of big ones.

Right, it was Philbrick. There's a section in one of Jim Williams' books about that.

There used to be technology in Massachusetts!

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

Yeah yeah, more standard troll babble. You have already seen my website, you know who I am and what I do. Oops, the logs say it was looked at from St. Petersburg 1-2 days ago, after my first post here for a long time. The location of the russian troll farms, sure you have nothing to do with them, I know, I know.

Like I said, they cast you all using the same form. Not so hard to understand why really, who would enlist as a russian propaganda troll if they were able to do something useful.

And don't try too hard to convince people here who I am, most of them have known me for many years. Many years before they assigned you to troll this group.

Dimiter

====================================================== Dimiter Popoff, TGI

======================================================

*** The lunatics are now truly in charge of the asylum. ***

Hey, this is the best line I have seen so far re the brexit madness! :D

Dimiter

====================================================== Dimiter Popoff, TGI

======================================================

You get more copper per square by using the smallest vias that get the full metal thickness down in the middle. With a 40-micron copper thickness, a 10-mil finished hole (250 um) has an unfinished diameter of

250+80 um, i.e.

pi/4*(330 um)**2 = 0.086 mm**2

and the finished hole's area is

pi/4*(250 um)**2 = 0.049 mm**2

so the hole area is 1 - 0.049/0.086 = 42% copper. Plated copper's alpha is about 380 W/m/K, and FR-4's is about 0.25 W/m/K. So assuming perfect heat spreaders on both sides, a 1.6 mm board with a rectangular array of these holes spaced by a distance d will have a thermal conductance per square metre of

1/theta(d) = (0.049E-6*380/d**2 + 0.25)/0.0016.

For holes on 1.5 mm pitch, this is 5300 W/K/m**2. For a square centimetre of thermal pad, we get 1E-4 square metres, so

theta = 1/0.53 ~ 2 K/W, not bad, and that's only a 6 x 6 array of holes.

Cheers

Phil Hobbs

You can get 0.2 W/K

--
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

Balderdash. I've been reading and posting to this group regularly since the mid 1990s under various guises and I don't remember you *at all* during all that time - despite having a *very* good memory for such things. If you're not Bill Sloman then you're Phil Alison; I'd put money on it.

--
This message may be freely reproduced without limit or charge only via  
the Usenet protocol. Reproduction in whole or part through other  
protocols, whether for profit or not, is conditional upon a charge of  
GBP10.00 per reproduction. Publication in this manner via non-Usenet  
protocols constitutes acceptance of this condition.

Thanks for clearing that up, Phil. I was beginning to wonder.

-- This message may be freely reproduced without limit or charge only via the Usenet protocol. Reproduction in whole or part through other protocols, whether for profit or not, is conditional upon a charge of GBP10.00 per reproduction. Publication in this manner via non-Usenet protocols constitutes acceptance of this condition.

** Very funny.

Neither Bill or I have ever used a "handle" or "sock puppet" and are not using them now.

I don't need to hide behind a stupid mask like you do nor think it one tiny bit fair to attack a poster while remaining anonymous.

Only cowards do that. Why don't you change your handle to " Complete Dickhead "

It has the same initials and is way more descriptive.

... Phil

Reece Mogg is old school english, but I don't have a problem with that. If he wants so specify rules in his own office, then fair enough, though the left wing msm will present it in the most prejudicial way possible, as usual...

Chris

The problem arises when he imposes it on the whole country (which, as a lawmaker, he has done), or makes the UK look stupid (which he does with his "metric is bad" stuff).

Presumably he wants to measure capacitance in Jars.

Ah. A fanatic that sees something and automatically presumes a conspiracy.

My production people don't like vias in pads, so a dpak or SOT89 or whatever needs to have a solid pad, then a region of solder-masked topside copper, then copper with a lot of vias. That will work for my dpak resistor, and I'd be happy to get somewhere near, say, 6 to 8 k/w overall.

That scheme doesn't work well for power pad parts that have leads all around. They just have to deal with vias in the pads.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

Empty vias in pads can suck up solder and leave voids, it's true. But there are lots of ways round that.

For one thing, large featureless pads typically get too much paste, which can cause open circuits in neighbouring leads of the same part. There's a happy medium someplace. ;)

Cheers

Phil Hobbs

I agree.

That is one thing that has been lost in the modern era of graphing calculators and spreadsheets everywhere.

People have lost the ability to sanity check their calculations to the nearest order of magnitude (as you always had to do when using a SR).

--
Regards, 
Martin Brown

** SR = slide rule - right ?

I still have mine from Sydney University Engineering days.

It's a Castell "Electro" in good working order and resides in it's original case from the very early 70s.

Honestly, I have almost totally forgotten how to use the darn thing.

Who remembers the old joke common in large computer installations where a similar rule was enclosed in a glass case on the wall ?

.... Phil

I don't think that is true. Sanity checking is something that you acquire with experience. Young people will get there, little by little. And then *they* start complaining that the *next* generation has "lost" that ability. It has always been thus.

Jeroen Belleman

** Sanity checking a SR result mainly depends on the ability to do mental arithmetic.

However, knowing what the right answer should look like IS mostly due to experience - but many calculator or computer simulation errors are so big accepting them is like believing the moon is actually made of green cheese.

..... Phil