breadboarding fast, tiny stuff

[Fred Kruger]

Dimbulb had one of those, but it was always run down.

More likely, the spring broke from over-winding after all that up and down
wrist action.

 

[Phil Hobbs]

I think the real question is: Why do we need myriad formats when we
have tar? It probably predates them all. All those young whipper-
snappers keep re-inventing the wheel. Oh well.

Jeroen Belleman

(Now let's have some archive format history...)

Tar is not a compressed format--zip dates from the days of 360k
floppies, remember!

Cheers,

Phil Hobbs

 

[Jeroen Belleman]

Tar is not a compressed format--zip dates from the days of 360k
floppies, remember!

Well, that brings up another point: Archiving and data compression
are two independent functions. The whole philosophy of UNIX-like
systems used to be to have programs that did one thing well, and
that you could pipe together in various ways. This is amazingly
powerful, once you get the hang of it.

This is totally lost on today's developers, who all seem to make
programs that do-it-all-and-then-some. That leads to many different
implementations of the same sort of functionality, all incompatible
to some extent and all with different idiosyncratic bugs.

Jeroen Belleman

 

[oopere]

I'm offering free data and advice, and you're whining about the price.

And it's not a web page, it's an FTP site.

And my camera makes jpeg's, not gif's.


Did I leave anything out?

John

Great info!
Any suggestions on how to drive such fast devices? (i.e. how can you
speed up slow edges?) Is there any comparator out there? Perhaps a LVDS
receiver?

Pere

 

[Joel Koltner]

The whole philosophy of UNIX-like
systems used to be to have programs that did one thing well, and
that you could pipe together in various ways. This is amazingly
powerful, once you get the hang of it.

What's interesting is that pipes never really caught on in a GUI'd
environment.

The problem with "piping things together on the command line" is that there's
more to remember: "someunpacktool -jsdlf furball.pack | tar -xvfg" is just not
as easy for the "casual" user as, "Double click on the .Zip file and follow
the instructions in the dialog that appears."

This is totally lost on today's developers, who all seem to make
programs that do-it-all-and-then-some. That leads to many different
implementations of the same sort of functionality, all incompatible
to some extent and all with different idiosyncratic bugs.

This is certainly true, but honestly the "hard core" UNIX approach goes too
far to the other extreme. There is some happy medium between "pipe everything
together on the command line with infinite options and super powers" and
"double-click but, no, there are no options for configuring anything, if you
don't like what happens, get another tool."

---Joel

 

[Michael A. Terrell]

More likely, the spring broke from over-winding after all that up and down
wrist action.

Nah. That just broke the watch band.

--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

 

[John Larkin]

How do you feel about wedding rings? ;-)

Expensive short circuits.

John

 

[Phil Hobbs]

It has been called creeping featuritis. Its roots began in the early
1960's with FORTRAN II and PL1.

Nah, it's way older than that. Bigger and bigger fins on
cars...automatic carriage return on typewriters...pilot lights on gas
stoves...all the way back to buttons so you don't have to tie your
loincloth on.

Cheers,

Phil "In my day we used every durn _part_ of the brontosaurus" Hobbs

 

[Fred Bartoli]

Tim Williams a écrit :

I'm sure Terry Given or anyone else in the business could tell stories
about induction heaters and rings, or something like that. ;-)

Yup, or about 10kW 5V PSUs... ouch!

 

[James Arthur]

[...]

How do you feel about wedding rings? ;-)

Expensive short circuits.

John

Hi-Q resonators?

Cheers,
James Arthur

 

[Joel Koltner]

Hi Joseph,

Not entirely true. There has been many implementations of GUI over
CLI implantations that were very effective and provided piping
capabilities.

Ah, good to know. I should have written that as, "What's intersesting is that
pipes never really caught on in any GUI I've used..."

Do you have a link to some examples? I'm curious to see what it "looks"
like...

 

[Joel Koltner]

Hi-Q resonators?

That depends on where you bought your ring now, doesn't it?

 

[James Arthur]

But make sure you know what's in that ceramic material ...

This is the second time I've seen this issue come up in SED. Since
I'm the one who gave that CPU to John, and would be loathe to deprive
the world of him a picosecond before his time, I'll add my $0.02.

BeO is a marvellous thermal material, an insulator with the thermal
conductivity of aluminum (300w/(m*K)).

Berylliosis is a dreadful but rare condition occurring if and only if
1) Be is inhaled, 2) in small enough particles, 3) by a person who's
sensitive or allergic to Be (estimated as 1-10% of the general
population).

The CPU in question is actually a TI486SXL2-G66-GA. It has a deep
bluish-purple tone, and is indeed a fabulous x-acto hone.

I don't know for a fact whether it contains beryllium, but I rather
think not. Here's why, and even if so, why AFAICT any danger would be
minimal:

I. Be CONTENT DOESN'T APPEAR LIKELY
1) BeO is white, and expensive. This CPU is purple, and cheap (a
budget '486 clone, sold on price).

2) The dangers of BeO have been known since the '40s, and I recall
reading warnings in Motorola literature pointing out which of their
products contained beryllium *within* the product, so that they might
be properly handled come disposal time. (Basically, no special
precautions except *don't grind it up.")

As Motorola was aware enough to warn people of Be *within* their
semiconductor assemblies by datasheet warnings in the '60s, I'd think
TI remiss not to warn people of products with cases made of the stuff
in the '90s. Especially since these things are ground up and
reprocessed by recyclers for their gold; it'd be a BIG DEAL--important
to publicize this--if the parts contained BeO.

3) Googling then and now, I find no matches suggesting TI has used the
stuff, much less in their bargain 'x86 CPUs, much less as a casing.

4) BeO is usually used as the heat-spreader, a thin thermal interface
sheet within a product. That is, the die mount. (Exception: certain
older, white, ceramic RF power transistors, whose entire cases were
BeO, IIRC).

II. LOW RISK OF EXPOSURE
5) When used as a hone, I doubt much dust could be made. a) The
reason it's used as a hone, after all, is that it's so much harder
than the knife blade, and b) mine's been used for years & shows no
evidence of wear whatsoever--no glazing, no grooves / gouges /
polish. None.

6) Most (but admittedly not all) times I hone with oil, which would
capture any dust.

7) Even all the 1-6 above failing, the exposure levels would be very
low. The rate of berylliosis and Chronic Beryllium Disease (CBD)
wasn't that high even in machinists who were grinding the ceramics and
machining beryllium alloys back in the day, steeped in the stuff.
Small consolation if you're one of the genetically unlucky hyper-
sensitive individuals, true, but most people aren't affected.

III. LOW RISK OF HARM FROM EXPOSURE
8) As stated already, only 1-10% of the population is susceptible, so
multiply any above risk by that factor.

So, I'd feel a tiny bit better knowing for sure, but I don't think the
risk is significant for this use of this CPU. I view it like lead
solder: don't eat the stuff, don't breathe the fumes, and you're fine.

Here, don't run a diamond wheel dry on this guy whilst huffing the
dust. Informed opinions to the contrary welcomed. Meanwhile, I'm
still using mine.

Cheers,
James Arthur

P.S. Oh, and BeO is not subject to RoHS, for whatever that's worth, so
you know it's safe. (NOT !)

 

[James Arthur]

III. LOW RISK OF HARM FROM EXPOSURE
8) As stated already, only 1-10% of the population is susceptible, so
multiply any above risk by that factor.

Clarification: 1-10% of the population are estimated "susceptible",
i.e. beryllium-sensitive, but the risk of problems is lower still--
only a fraction of these, when exposed (to an adequate dose of
sufficiently fine particulate), develop CBD.

Cheers,
James Arthur

 

[James Arthur]

That depends on where you bought your ring now, doesn't it?

And how sweaty your hands are when you put it on, too!

Cheers,
James Arthu

 

[Allan Herriman]

The Eclips ecl gates (the MC10EL and EP parts) swing almost 0.9 volts
in about 220 ps, which is perfect for slamming the gates of modern
phemts. Analog Devices makes some comparators that are a lot faster,
ballpark 40 ps with ecl swings. I'll have to try driving a phemt from
one of them, just to see.

Some of the lvds line receivers will swing 3.3 or 5 volts in roughly
600 ps.

John,

Have you ever experimented with CML logic gates instead of ECL? I
mean something like the NBSG16M. The swing is somewhat reduced (about
400mV) but they seem pretty fast (at least from the datasheet).

http://www.onsemi.com/PowerSolutions/product.do?id=NBSG16MMNG

Thanks,
Allan

 

[John Larkin]

John,

Have you ever experimented with CML logic gates instead of ECL? I
mean something like the NBSG16M. The swing is somewhat reduced (about
400mV) but they seem pretty fast (at least from the datasheet).

http://www.onsemi.com/PowerSolutions/product.do?id=NBSG16MMNG

Thanks,
Allan

We use an NBSG16VS as a variable-swing pulse generator. It gens a
fast, programmable-level pulse that we feed into a couple of Hittite
distributed amps, to finally get a 0-6 volt pulse that drives an
optical modulator. It all works nicely, very clean and fast. All three
chips are power-pad mounted with heatsinks on the back side of the
board. The NBSG costs $32, and the Hittites are about $200 each.

PCB:

ftp://66.117.156.8/AmpTop.jpg

and output into 50 ohms, 1 volt/cm:

ftp://66.117.156.8/NewSq2.jpg


The NBSG would be a nice TDR step generator.

John

 

[oopere]

We use an NBSG16VS as a variable-swing pulse generator. It gens a
fast, programmable-level pulse that we feed into a couple of Hittite
distributed amps, to finally get a 0-6 volt pulse that drives an
optical modulator. It all works nicely, very clean and fast. All three
chips are power-pad mounted with heatsinks on the back side of the
board. The NBSG costs $32, and the Hittites are about $200 each.

PCB:

ftp://66.117.156.8/AmpTop.jpg

and output into 50 ohms, 1 volt/cm:

ftp://66.117.156.8/NewSq2.jpg


The NBSG would be a nice TDR step generator.

John

Superb waveform! Are the hittite amps DC coupled or do you clamp the
output somehow?

Today I have been playing with a 74LVC00A which was available in some
drawer to generate "narrow" pulses from a square signal in the quick and
dirty way:

----
---·-----| \
| |NAND---NOT--
|-NOT-| /
----

In -------- -------
-----| |-------|
_ _
Out------| |------------| |----

Output pulse width is rougly 2.5ns with a leading edge rise time of 1ns.
It fills the spectrum quite nicely up to 1 GHz. (Currently targeting the
0-1GHz part of UWB spectrum -impulsive flavour).

The NBSG16VS would probably allow to take this some steps further. Have
also looked at the MC10EP05. However, from the ON-Semi datasheets I
can't figure out how fast the input edges should be. Is there some
schmitt-trigger built into the inputs? How fast will the output edges be
for say 2ns input edges? Do I only have to care on the time it takes
through the V_IL-V_IH zone?

Pere

 

[John Larkin]

Superb waveform! Are the hittite amps DC coupled or do you clamp the
output somehow?

All ac coupled. This thing gates the NIF laser beamlines, which run at
an absurdly low duty cycle. We generate a 0-30 ns wide pulse at 960
Hz. The laser itself fires maybe a few times a day.

Today I have been playing with a 74LVC00A which was available in some
drawer to generate "narrow" pulses from a square signal in the quick and
dirty way:

----
---·-----| \
| |NAND---NOT--
|-NOT-| /
----

In -------- -------
-----| |-------|
_ _
Out------| |------------| |----

Output pulse width is rougly 2.5ns with a leading edge rise time of 1ns.
It fills the spectrum quite nicely up to 1 GHz. (Currently targeting the
0-1GHz part of UWB spectrum -impulsive flavour).

Take a look at the NL37WZ16 triple buffer. With all three sections in
parallel, it will put 5 volts into 50 ohms in about 600 ps. Some of
the LVDS-to-ttl receivers are about that fast, but not quite so fierce
a drive.

The NBSG16VS would probably allow to take this some steps further. Have
also looked at the MC10EP05. However, from the ON-Semi datasheets I
can't figure out how fast the input edges should be. Is there some
schmitt-trigger built into the inputs? How fast will the output edges be
for say 2ns input edges? Do I only have to care on the time it takes
through the V_IL-V_IH zone?

The spec sheet typ EP05 risetime is 130 ps. These things aren't
schmitts but they do have very high voltage gains, so they don't need
a super-fast input edge. Something pokey like 1 volt per ns should do
fine.

John

 

[Terry Given]

I can't name any off the top of my head, but perhaps xemacs qualifies.

and then some!

http://xkcd.com/378/


Cheers
Terry (who doesnt wear a watch, as one nearly got him killed)