Fuses or just a lump of wire?

[AAH]

I was fed the following video by the youtube algorithm and I thought that it was quite interesting . It also made me check my own supplies of fuses to confirm that they were really reliable fuses and they met the expected blow curves. I was quite pleased that they did especially as I buy direct from a reputable manufacturer of fuses and fuseholders.
The lack of fusing from the fuses in the video is incredible. If you're putting fuses in your display which everyone should be you shouldn't necessarily try to save a few cents buying some "fuses" from the cheapest seller on Ebay. I buy from http://www.sellifuse.com/Stacp.html who have a fairly godawful site but a potentially better Aliexpress site.

View: https://www.youtube.com/watch?v=apQU_VuJlFU

 

[Notenoughlights]

A few years ago I bought some mini automotive fuses. They all looked identical, just with different coloured plastic. The fuse "link" was the same size, from 5A all the way to 30A.

 

[dkulp]

Very interesting. While testing and evaluating eFuses for my board, I did a bunch of testing with 5A fuses that my manufactures sticks on my boards. I blew over 30 of them from various batches of my boards and such to try and determine some of the "actual" characteristics and they all blew somewhere between 5.3A and 6.2A. I was somewhat surprised on the variance, but at least I didn't see anything as bad as in that video. Wow. Kind of scary.

The eFuses behave differently as they measure the current directly and don't rely on a wire heating up and melting. The eFuses I'm using can operate forever at 5.0A, but once they hit 5.3A, they will trigger very very fast (milliseconds).

 

[Katekate]

I only buy quality fuses for this reason. Narva stuff isn't expensive.

 

[Indigogyre]

After I shared that video with some other friends of mine they indicated his testing method was not the best method due to the type of power supply he was using. I do know the original poster has made a second video on the same subject but I have yet to watch it.

Having said that everyone seems to agree that buying quality fuses is the smart thing to do.


View: https://www.youtube.com/watch?v=mJ97tVmZizQ

 

[Skymaster]

The eFuses I'm using can operate forever at 5.0A, but once they hit 5.3A, they will trigger very very fast (milliseconds).

I'm genuinely curious to know how this pans out with dimming and PWM, given the LEDs are transitioning from 100% to 0% and only dimmed in perception, rather than actual lower instantaneous current draw.
Especially thinking around the folk that'll pack out a port, load it up to 5A total, but at 30% brightness.

Do you have any form of capacitance on the board to smooth out the apparent load, as not to avoid nuisance tripping in this scenario, or has it not been relevant in the testing you've performed?

 

[dkulp]

I'm genuinely curious to know how this pans out with dimming and PWM, given the LEDs are transitioning from 100% to 0% and only dimmed in perception, rather than actual lower instantaneous current draw.
Especially thinking around the folk that'll pack out a port, load it up to 5A total, but at 30% brightness.

Do you have any form of capacitance on the board to smooth out the apparent load, as not to avoid nuisance tripping in this scenario, or has it not been relevant in the testing you've performed?

The eFuses themselves allow for very short bursts up to 100% over (~10A). But that's "very short" bursts. For PWM cases, that should be OK. I've done some testing with them and have sold a bunch of the K8-PB's that have them onboard. The Falcon folks have done more testing with them as they are using the same part on the new Falcon v5 boards. So far, we're quite pleased with them. They certainly open up a slew of new possibilities. Being able to completely power down ports when the show is not running is a safety thing as well as money saving. The immediate alerts (and thus push notifications to phones/etc...) if a fuse triggers can be important. The current monitoring is interesting. Not sure what all that that will open up. The pixel counting is interesting, but I'm sure there is more we can do with it. For starters, I want an FPP Command that will do the Pixel Counting and compare to the configured outputs and alert if there is a miss-match. Could run it 15m before show or something to determine if a pixel is causing a large chunk of a string to be out.

 

[Skymaster]

Very interesting. I think it's a fantastic innovation brought to the table, and the possibilities in that space are game changers.