This site is now in read-only archive mode. Please move all discussion, and create a new account at the new Victron Community site.
In the interest of furthering the knowledge base this Community provides, I'm asking a question here -and will post the answer below- that I receive not infrequently when assisting Victron (and other) customers with troubleshooting their various devices and ask them to check the fuse to ensure that it's not blown or half-blown:
"Is it really possible for a fuse to half-blow? I've never heard of that!"
The answer, though it may surprise many, is a resounding YES. The attached photo is of a 3.15A fuse from a Cerbo GX power supply wire, and it has "half-blown", and what I mean by that is: under 0A load, the fuse will pass full vbatt (13.6v in this case), with no voltage drop (at least, none within a hundredth of a volt).
As soon as even the smallest load (~5ma) is applied, however, voltage through the visibly melted fuse filament drops to roughly half vbatt, and the voltage drop gets worse as the load is increased. At ~2.5A, voltage through the filament has dropped to ~1.6v.
The lesson here is simply to understand that a cursory visual inspection of a glass fuse such as this (or a BMV, or CCGX, or any number of other products that use glass fuses) is not always enough to determine whether or not the fuse is good; indeed, even a basic continuity test will not always reveal this condition, as -depending on how the filament melted- the fuse may present perfect or near-perfect continuity when tested with a multimeter.
As such, when experiencing power issues with any device using this type of fuse, I always suggest replacing the fuse first, even if it seems to look okay, because looks can be deceiving when it comes to these.
While replacing the fuse, of course also visually inspect the contacts inside the fuseholder, as indeed even the smallest amount of corrosion or other contaminant present at the contacts can also lead to erratic behavior including sporadic loss of power under certain circumstances and/or total loss of power altogether.
Cheers!
@Justin Cook - Bay Marine Supply USA Not very sporting of you to first get some other insights and responese first. That is the best part of a forumn. (Only kidding)
@ejrossouw LOL!
Actually you're right, it would have been far more entertaining to leave it unanswered and see what kind of responses came in... but, then again, the Community is more a knowledge base than a traditional forum, so hard evidence must be given precedence over opinion here :-P
@Justin Cook - Bay Marine Supply USA I personally don't see why not, but this is just taking a wild stab at it. If the properties of the fuse wire e.g. has only slightly changed due to coming close to blowing, it may well leave it in a halway state? I'd however suspect this is more likely with some mass produced items. It is not always about volts and amps and resistance plays a part, especially with some sensitive measuring equipment. I'd always suggest swapping the fuse out for a test and not just checking continuity. Measuring teh resistive properties may also be useful?
@ejrossouw yep, exactly that, though I've found in practice that resistance test won't always show the fault either, until the degraded filament is under load.
I was really just posting this here because I commonly get disbelief when I tell people that a "half-blown" fuse happens... so this is a convenient place to post the question, the answer, and the evidence :)
I was planning to call you tomorrow to request an RMA on my BMV, but decided first to try a new fuse even though the old one looked just fine to me. New fuse == Problem solved.
As they are slow blow fuses they do not blow instantly anyway. They need a sustained over current to blow. There is clearly a point where the metal heats for long enough to cause deformation but not long enough to fully deform the metal to the point of breaking.
From a logical assumption perspective it seems conceivable this half-blown fuse could come about by a repeated short spikes rather than just one event. That being the case when seeing this
I would want to asses the system to see if there is anything that could be causing a repeated short spike.
