Resistor/Fuse replacement (DEC H7104-D)
drlegendre .
drlegendre at gmail.com
Tue Mar 22 22:56:46 CDT 2016
Vincent & All,
Yes, I read that the series conforms to that particular UL fusing
requirement. My point was that I didn't see any promotion of or expansion
on the fusing qualities of the design. The UL spec almost seemed like an
afterthought. But then, I am not an engineer - so not used to parsing these
documents to such a level.
But you comment on fire prevention is duly noted. As the saying goes, "the
transformer fails open to save the fuse" ;-)
As for a resistor + fuse in series, I suspect you'd want to spec the fuse
for allowable currents - and then spec the resistor to withstand any
current in that range. Point being, you only +need+ one 'weak link' in the
design.
Thoughts?
On Tue, Mar 22, 2016 at 10:20 AM, Vincent Slyngstad <
v.slyngstad at frontier.com> wrote:
> From: drlegendre: Monday, March 21, 2016 6:50 PM
>
>> I don't quite get what makes this DigiKey part suitable for the role of a
>> fused resistor. I do see that it has specs for 'fusing behavior' but that
>> aside, I don't see that this series is marketed / sold as a "fusible
>> resistor".
>>
>
> I take "UL1412 recognised fusing" (sic) on the first line of the data sheet
> to mean that they do market them that way.
>
> One reason I question it, is the fact that the fusing ratings are only
>> plotted for like 40X or 50X expected current. Can the circuit under
>> protection be relied upon to produce those levels of current, even under
>> hard-fault conditions?
>>
>
> I read a little over 1000 seconds to fuse at 10W, which is only a few
> times the 2W rating. Admittedly, 20 minutes at 5X load amounts to a pretty
> slow fuse. I can only assume their concern is fire prevention, rather than
> circuitry protection.
>
> With regard to the suggestion of a fuse and a resistor, you'd need more
> room (likely not a problem), and a flameproof version of the resistor. I
> don't know anything about UL ratings, so I don't know if that could be made
> OK there or not.
>
> Vince
More information about the cctech
mailing list