[time-nuts] Re: Z3801A questions

[Jim Harman]

A traditional soft start circuit you could try would be to use a series
resistor and a 48V relay. 20 ohms would limit the current to the max of
your 2.9A supply. Connect the relay coil between the load side of the
resistor and ground. Connect the relay contacts so they short the resistor
when the relay is engaged. At startup, the load acts like a very low
resistance and the relay does not pull in right away. As the load
current drops, the voltage on the relay coil increases until it pulls in,
connecting the supply directly to the load. Hopefully the supply can handle
this smaller surge.

You may have to experiment with the resistor value. A larger resistor would
reduce the surge but it must be small enough so that the relay eventually
pulls in and the load starts up properly. Also consider the power rating of
the resistor, and whether you need a slow blow fuse to protect it in case
the load is shorted.

On Sun, May 28, 2023 at 4:44 PM Robert LaJeunesse via time-nuts <
time-nuts at lists.febo.com> wrote:

> Not cheap, but not particularly expensive, $50 gets you a new 48V / 5A
> supply:
>
> https://www.mpja.com/48-Volt-Power-Supply-5A-240W-Hengfu/productinfo/18437+PS/
>
> For a few dollars less, series up a pair of 24V / 6.5A supplies:
>
> https://www.mpja.com/24-Volt-Power-Supply-65A-150W-Switching/productinfo/33470+PS/
>
> Bob LaJeunesse
>
> > Sent: Friday, May 26, 2023 at 9:23 PM
> > From: "Mahlon Haunschild via time-nuts" <time-nuts at lists.febo.com>
> > To: "Discussion of precise time and frequency measurement" <
> time-nuts at lists.febo.com>
> > Cc: "Mahlon Haunschild" <k4oq.radio at gmail.com>
> > Subject: [time-nuts] Re: Z3801A questions
> >
> > Well, as expected, when connected across my 48VDC infinite current
> source (e.g. golf car battery) the Z3801A booted right up.
> >
> > Now all I need is a 48VDC 5A unregulated power supply.  Decades ago,
> these things were a dime a dozen.  Now they cost hundreds of dollars.
> >
> > Or maybe a suitable step-start or ICR between the Vicor SMPS and the
> Z3801A.  After all, the Z3801A only draws roughly 500 mA, steady-state,
> while the Vicor can source 2.9A.  Plenty of room for that sort of thing.
> >
> > Thanks for your support.
> >
> > Best regards
> >
> > Mahlon - N4ZK
> >
> > > On May 22, 2023, at 6:26 PM, Mahlon Haunschild <k4oq.radio at gmail.com>
> wrote:
> > >
> > > Ed,
> > >
> > > I think my 48V golf car battery bank is stiff enough for this :)
> > >
> > > Seriously, that is what I always intended to try first.  Fed through a
> suitable fuse/circuit breaker, of course.  Will give it a go and report
> back later.
> > >
> > > Best regards
> > >
> > > Mahlon -N4ZK
> > >
> > >> On May 22, 2023, at 2:19 PM, ed breya via time-nuts <
> time-nuts at lists.febo.com> wrote:
> > >>
> > >> Before doing a bunch of mods to the Z3801A PS, it would be wise to
> try running it from a linear supply with enough V and A to launch. SMPSs
> usually have a hard time with load startup surges. In PS experimenting on
> my 24V unit, I found that SMPS units of seemingly adequate size - even two
> paralleled - could not fire it up from cold. Neither could a linear one,
> unless rated for much higher output current - it would fold-back and stall.
> > >>
> > >> The bottom line is that you need substantially more than nameplate
> current to get these puppies up and running. They were built to run from
> telecom battery bus supplies, which can handle those critical few
> > >> seconds at startup.
> > >>
> > >> An old-school unregulated transformer/rectifier/capacitor supply is
> the closest match. This kind of setup can be sized (VA-wise) close to the
> normal running power needed, but is forgiving enough to handle the startup
> surge - use a slow fuse sized about right for steady-state VA.
> > >>
> > >> Ed
> > >> _______________________________________________
> > >> time-nuts mailing list -- time-nuts at lists.febo.com
> > >> To unsubscribe send an email to time-nuts-leave at lists.febo.com
> > _______________________________________________
> > time-nuts mailing list -- time-nuts at lists.febo.com
> > To unsubscribe send an email to time-nuts-leave at lists.febo.com
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe send an email to time-nuts-leave at lists.febo.com



-- 

--Jim Harman