Re: measuring output of adapter
- From: real-address-in-sig@xxxxxxxxxxxxxxx (Rowland McDonnell)
- Date: Sun, 9 Aug 2009 15:09:53 +0100
Adrian Tuddenham <adrian@xxxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote:
Rowland McDonnell <real-address-in-sig@xxxxxxxxxxxxxxx> wrote:
[...]
Mind you, I've used 1N4148s as reverse bias protection in various places
(but in the ancient past)...
Designers often forget that with continuous ratings up to 75 volts and
75mA, they can be used as convenient small power devices. Their single
pulse ratings can be up to an amp.
Yeah, that's what I needed it for. There was this low duty cycle spike
that I wanted clobbered *fast*. Quicker than a power diode and oh look
there's one under my hand here in the lab, I'll shove it in, will you
look at that? It works! And it's not died like the bloody 4000 series
CMOS PLL IC has been on a regular basis (either that spike I mentioned,
or me frying it with my natural static - just me *looking* at the damned
things when unplugged seems to kill 'em. Never blown up anything else
due to static, mind - just 4000B series CMOS ICs).
Personally, I prefer to insure against catastrophic failure by using big
lumps of silicon and simple protection devices. Too often the
sophisticated protection systems get caught out by some unexpected
combination of events - and big devices run gently will last a lot
longer and not cook the components around them.
The business with the protected FETs I mentioned was to protect the big
lumps of silicon when the design around them failed to do so - and with
protection circuitry integrated on the same bit of Si as the power
device, all wrapped up in the one box, well, you might as well just
treat it like a normal FET (which was the idea), just one that's nigh on
impossible to blow up. Or so the theory went, anyway.
Doing it that way, the overall design doesn't usually turn out to be
much more expensive and it gives a comfortable feeling of being
bomb-proof. The concept of pushing output (or PSU) devices to their
limits originated in valve equipment, where it made more sense (and
where the limits were less absolute).
When I did my MSc, we had an electronics lecturer who was keen on
pushing ICs beyond spec limit on speed and suchlike.
I ignored all his suggestions on exceeding spec with the devices on
offer. Any gadget I soldered together worked reliably (*I* use devices
well within spec, especially wrt current and power and speed). The same
could not be said for pretty much everyone else on the course.
<heh> I recall sorting out a circuit someone else had made - needed
decoupling capacitors on the power rails of all the ICs but had none.
Veroboard construction, no space left on the component side to put the
caps in, so I soldered some small disc ceramic caps on the track side,
being careful about it. `You can't do that!' I was told by the cute
chick who'd made the circuit in the first place (I'd been instructed to
rescue her by the lecturer in charge - who, me, complain? ... I suspect
the lecturer was banking on that effect kicking in.)
Yeah, well, `unconventional' does not mean `in any way impossible or
even necessarily dubious'.
Rowland.
--
Remove the animal for email address: rowland.mcdonnell@xxxxxxxxxxxxxxx
Sorry - the spam got to me
http://www.mag-uk.org http://www.bmf.co.uk
UK biker? Join MAG and the BMF and stop the Eurocrats banning biking
.
- References:
- Re: measuring output of adapter
- From: Rowland McDonnell
- Re: measuring output of adapter
- From: Adrian Tuddenham
- Re: measuring output of adapter
- From: Rowland McDonnell
- Re: measuring output of adapter
- From: Adrian Tuddenham
- Re: measuring output of adapter
- Prev by Date: Re: Delays on startup
- Next by Date: Re: Why 2009 is like 1984
- Previous by thread: Re: measuring output of adapter
- Next by thread: Rolex Oyster Perpetual Lady Datejust Ladies Watch 179174-MAJ
- Index(es):
Relevant Pages
|
Loading