[time-nuts] UPS for my time rack

[Neil Schroeder]

Depending on your actual power throughput needs- PTCs can be a challenge as
Bob mentions.  You will have a tough time meeting the power requirements of
a Thunderbolt with many you'll see.   Now efuses (think TPS24/25xxx)  or
mosfets with a controller (think LTC43xx) can be a good choice but don't
hesitate to use a real honest to god fuse that blows out and waits for you
to come replace it if that is what you would need

I personally do as suggested in the thread above to some degree- I do AC to
DC conversion once for all my time and frequency stuff then have batteries
local to important gear and (in older stuff) a diode(s) or (in newer stuff)
a real hotswap controller.  But whenever something is really important
there's whatever fuse is needed.

Finally if you are of the type to be adventuresome -
http://cds.linear.com/docs/en/datasheet/4110fb.pdf  or

- Google for Freescale's, ON Semi's or TI's *offline* UPS Reference
designs.  Yes they invert DC to AC, but you can drive the inverter gates
with a microcontoller's PWM output. meaning no matter how jagged your
utility's wave may be you can create a perfect, beautiful 50 or 60 Hz sine
of your own - devoid of most noise.

NS

On Wed, Oct 14, 2015 at 12:01 AM Bob Camp <kb8tq at n1k.org> wrote:

> Hi
>
> One option for the “fuse” part of the DC supply system are PTC resettable
> fuses. You *do* indeed need to be careful about voltage and current ratings
> on these gizmos. If the only objective is to “not have smoke” when there
> is a short,
> that can reduce the variables to a manageable level. If your wire will
> handle 20A for
> long enough to get the gizmo to limit and your load is typically < 5A
> there are
> parts you can find.
>
> A few cautions:
>
> The trip points are very temperature dependent. If you need to handle -20C
> in the winter
> and +40C in the summer, that will narrow things down quite a bit.
>
> Mounting matters quite a bit. If you go with SMD parts, be careful of
> traces that act as
> heatsinks. This is one case that the part needs to get hot.
>
> There are to many variables on most of the spec sheets to simply pick one
> and move on.
> The only good way to do it is to get several and run repeated tests on
> them Min carry current
> will occur at your highest temperature. Worst case trip will occur at your
> lowest temperature.
>
> Consider that the “carry current” may not be the real limitation.
> Resistance in the supply lead
> of an OCXO is a bad idea. This may limit your current well below the point
> that the fuse actually
> trips.
>
> This sounds like a pretty scary list. To some degree it is. These parts do
> have their place. That
> does not mean they work everywhere and anywhere.
>
> Bob
>
> > On Oct 13, 2015, at 2:54 PM, Poul-Henning Kamp <phk at phk.freebsd.dk>
> wrote:
> >
> > --------
> > In message <4FD0F30EBAEF49609DF207E3EE61C15B at pc52>, "Tom Van Baak"
> writes:
> >
> >
> >> I used to rely on one massive UPS (along with natural gas generator)
> >> for my entire lab. Eventually I found it more reliable and convenient
> >> to have localized power backup. By local I mean backup for a single
> >> shelf, or even a single instrument.
> >
> > The big gain is avoiding the DC->AC conversions.
>
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>