[ARC5] [Milsurplus] HV Fuses: You Were Right, I Was Wrong.. more
Peter Gottlieb
nerd at verizon.net
Sat Jun 16 10:08:07 EDT 2012
I work with high power DC systems running at up to 1100 volts. We typically use
1400 volt DC rated fuses, the smallest is rated 160 amps. Internally they are
made of pure silver foil with multiple lines of holes punched across the ribbon
leaving thin filaments. Thus, each of these weak areas is a "zipper fuse" where
overcurrent will open one strand, then higher current concentration the next,
until they all blow open. If an arc forms actoss that segment, current will
flow in the next one and the same thing happen. I don't remember how many
zipper fuse segments there were, something like 6-8. The fuses are sand
filled. The way the fuse elements (zipper sections) are put in series would
tend to validate the idea of putting multiple fuses in series to increase the
voltage rating. (I'm not a fuse engineer but it seems to make logical sense)
We did multiple short circuit testing runs on large systems with something like
20 fuses each (ouch, they aren't cheap fuses!) since the interrupting current
was very close to the 100kA fuse specs. I would then disassemble the fuses to
examine the failures. Remarkably, it was rare to find more than one zipper
section in each fuse to have opened. Thus, I would assume that the testing and
design of the fuses anticipated a much higher inductive break than we used (it
is hard to get large inductances that can handle many hundreds of thousands of
amps). I would think that with so much iron in these old transmitters there
might be really significant inductive kicks and serious arcs across opening
fuses. So, if it was my gear and I really cared about it, I would either get
the right fuse or string together enough lower voltage ones to make the required
rating. When I had my Harris transmitter setup I accidentally opened the HV
fuse the first time I tried adjusting final bias. This was a 3 amp 5000 volt
fuse and looked about the size of a 6 inch pencil. I was lucky to be able to
get several samples from a friendly fuse manufacturer sales rep (the company I
worked at bought millions of dollars of fuses every year) since they were
special order, 100 quantity minimum, and over $80 each. After that I was *much*
more careful.
Peter
On 6/16/2012 9:46 AM, Bruce Long wrote:
> FYI McMaster Carr has two types of fuses rated 1000 vdc, one is designed for
> commercial solar panel use - is somewhat expensive at $25 each and comes in
> several current ratings
>
> The other type is intended for use in multimeters and is less expensive Both
> types are spec-ed with the maximum interrupt current
>
> I have not been following this thread close enough to remember the voltage in
> question but I suppose it would be in excess of 1000 vdc. Still these fuses
> would work better than conventional fuses and two or three in series might
> well do the trick
>
> --------------------------------------------------------------------------------
> *From:* Peter Gottlieb <nerd at verizon.net>
> *To:* jfor at quikus.com
> *Cc:* arc5 at mailman.qth.net
> *Sent:* Saturday, June 16, 2012 9:38 AM
> *Subject:* Re: [ARC5] [Milsurplus] HV Fuses: You Were Right, I Was Wrong.. more
>
> Yeah but you'll feel even worse if you spent $80 to protect that $500 tube and
> STILL had the $500 tube blow first!
>
> No guarantee the "proper" fuse will perform any better. I^2t calculations were
> rarely ever done back then, and even if they did, what could they do about it?
> Their choices were fast blow, normal, slow blow and no-blow (forego fuse) and
> for some components it made little difference, they were going down no matter
> what.
>
> ...P
>
>
> On 6/16/2012 9:25 AM, J. Forster wrote:
> > It might be worth the $80 if you are protecting a $500 tube!
> >
> > I remember an old comment about expensive devices blowing to protect the
> > fuse....
> >
> >
> > -John
> >
> > ==========
> >
> >> I saw a spec sheet saying the 1 amp fuse wire has 510 feet on the spool.
> >> Still,
> >> for the high cost of it (and then you have a 400 year supply), the copper
> >> wire
> >> solution is looking mighty nice.
> >>
> >> Peter
> >>
> >>
> >> On 6/15/2012 11:58 PM, Roy Morgan wrote:
> >>> On Jun 15, 2012, at 8:42 AM, J. Forster wrote:
> >>>
> >>>> If it were me, I'd try and get some 'fuse wire'. I don't know if it is
> >>>> still available.
> >>> From a later post, it does seem to be available, but pricey: though
> >>> the
> >>> price per roll did not say how much you get on a roll.
> >>>
> >>>> ...If you can't find fuse wire, consider fine magnet wire with the
> >>>> appropriate fusing current.
> >>> On that point, I find in my notes file on fuses the following:
> >>>
> >>>> fuses.txt
> >>>> From K1LKY
> >>>>
> >>>> From: "Dr. Barry L. Ornitz"<ornitz at tricon.net <mailto:ornitz at tricon.net>>
> >>>> To: "Tom Rauch"<w8ji at contesting.com <mailto:w8ji at contesting.com>>; "Old
> Tube Radios"
> >>>> <boatanchors at theporch.com <mailto:boatanchors at theporch.com>>
> >>>> Sent: Thursday, December 09, 2004 1:45 PM
> >>>> Subject: Re: Which begs the question...
> >>>>
> >>>>> Tom, W8JI, had an excellent discussion on protecting tubes and
> >>>>> high voltage power supplies under fault conditions. In this he
> >>>>> wrote:
> >>>> Resistor content clipped out..
> >>>>
> >>>>> Tom also wrote:
> >>>>>
> >>>>>> I'm guilty of using thin enameled wire as a HV link fuse,
> >>>>>> and I have on occasion used carbon resistors to locate an
> >>>>>> unknown fault (like which tube is arcing?). But for reliable
> >>>>>> protection in commercial or tell-the-world-to-do-it-this-way
> >>>>>> articles a bit more sensible engineering would be better.
> >>>>> I have done this too, using 32 gage and smaller wire. Always
> >>>>> working near chemists, I had ready access to glass tubing to
> >>>>> run this inside. One trick was to use the spring from a
> >>>>> ballpoint pen to put the wire under tension. Thus when the
> >>>>> wire fused, the spring would quickly separate the two ends,
> >>>>> resulting in less arcing and ionization. Like Tom, I would not
> >>>>> want to do this with a commercial product design for production.
> >>>>>
> >>>>> A good approximation for the fusing current for copper wire
> >>>>> with the length much greater than the diameter is:
> >>>>>
> >>>>> I = 10244 * d^1.5
> >>>>>
> >>>>> where: I = fusing current in amps,
> >>>>> d = wire diameter in inches.
> >>>>>
> >>>>> So for 32 AWG wire, the diameter is 7.95 mils, and the fusing
> >>>>> current is approximately 7 amps.
> >>>>>
> >>>>> 73, Barry WA4VZQ ornitz at tricon.net <mailto:ornitz at tricon.net>
> >>>> At:
> >>>> http://www.railcar.co.uk/mechanical/data/fuses.htm
> >>>> we find that no. 42 plain annealed copper wire used as fuse wire is
> >>>> rated at 1.5 amps, with a blowing current of 2.5 amps. (these may
> >>>> be BWG sizes!)
> >>>>
> >>>> Rating Blowing current wire size
> >>>> 1.5 2.5 42
> >>>> 3 5 38
> >>>> 5 8 36
> >>> Roy
> >>>
> >>> Roy Morgan
> >>> k1lky at earthlink.net <mailto:k1lky at earthlink.net>
> >>> K1LKY Since 1958 - Keep 'em Glowing!
> >>>
> >>>
> >>>
> >>> ______________________________________________________________
> >>> ARC5 mailing list
> >>> Home: http://mailman.qth.net/mailman/listinfo/arc5
> >>> Help: http://mailman.qth.net/mmfaq.htm
> >>> Post: mailto:ARC5 at mailman.qth.net <mailto:ARC5 at mailman.qth.net>
> >>>
> >>> This list hosted by: http://www.qsl.net
> >>> Please help support this email list: http://www.qsl.net/donate.html
> >>>
> >>>
> >>> -----
> >>> No virus found in this message.
> >>> Checked by AVG - www.avg.com <http://www.avg.com/>
> >>> Version: 10.0.1424 / Virus Database: 2433/5072 - Release Date: 06/15/12
> >>>
> >>>
> >> ______________________________________________________________
> >> ARC5 mailing list
> >> Home: http://mailman.qth.net/mailman/listinfo/arc5
> >> Help: http://mailman.qth.net/mmfaq.htm
> >> Post: mailto:ARC5 at mailman.qth.net <mailto:ARC5 at mailman.qth.net>
> >>
> >> This list hosted by: http://www.qsl.net <http://www.qsl.net/>
> >> Please help support this email list: http://www.qsl.net/donate.html
> >>
> >>
> >
> >
> >
> > -----
> > No virus found in this message.
> > Checked by AVG - www.avg.com
> > Version: 10.0.1424 / Virus Database: 2433/5073 - Release Date: 06/16/12
> >
> >
> ______________________________________________________________
> ARC5 mailing list
> Home: http://mailman.qth.net/mailman/listinfo/arc5
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:ARC5 at mailman.qth.net <mailto:ARC5 at mailman.qth.net>
>
> This list hosted by: http://www.qsl.net <http://www.qsl.net/>
> Please help support this email list: http://www.qsl.net/donate.html
>
>
> --------------------------------------------------------------------------------
>
> No virus found in this message.
> Checked by AVG - www.avg.com <http://www.avg.com>
> Version: 10.0.1424 / Virus Database: 2433/5073 - Release Date: 06/16/12
>
More information about the ARC5
mailing list