| General > General Technical Chat |
| Just Replace the Fuse and You're Good!! |
| << < (5/5) |
| Alti:
--- Quote from: engrguy42 on May 22, 2020, 10:30:01 am ---Alti, you're applying a DC to the capacitive dropper. Why? We're talking about AC. --- End quote --- To be precise, I apply a step function. This is 0V in the past and 110VDC*sqrt2 afterwards. This is no different than sticking your meter at peak of sin(x) to the mains receptacle. If you want to, you can use sin(x) 60Hz, starting at peak but you won't see any perceivable difference of inrush during those 1.5ms |
| engrguy42:
--- Quote from: Alti on May 22, 2020, 12:23:02 pm --- --- Quote from: engrguy42 on May 22, 2020, 10:30:01 am ---Alti, you're applying a DC to the capacitive dropper. Why? We're talking about AC. --- End quote --- To be precise, I apply a step function. This is 0V in the past and 110VDC*sqrt2 afterwards. This is no different than sticking your meter at peak of sin(x) to the mains receptacle. If you want to, you can use sin(x) 60Hz, starting at peak but you won't see any perceivable difference of inrush during those 1.5ms --- End quote --- Alti, yeah, you're right. This freakin' LTSpice was assuming the capacitor starts off fully charged |O So I added a shunt resistor to discharge it before applying the AC, and yeah, there's a super short current pulse (< 1mSec) to charge the capacitor. Maybe that would be enough? Though it seems in this case it was more of a mechanical issue. But that's interesting, I never realized it might be an issue. Thanks. EDIT: BTW, the correct way to set initial conditions on the capacitor is to CTRL-right click on the capacitor and in the "SpiceLine" cell set the value to "IC=0". |
| Alti:
--- Quote from: engrguy42 on May 22, 2020, 12:41:43 pm ---Maybe that would be enough? Though it seems in this case it was more of a mechanical issue. --- End quote --- No, it is not enough because even super-fast LRC Littelfuse would have survived 50x higher abuse (once). Whatever 200mA fuses you are going to get, these can only have higher i2t rating. As for mechanical failure - I really doubt it. It might be this was just damaged during production although it is hard to believe they do not test fuses for continuity. If you are interested to investigate further then I can recommend non-destructive fuse testing . You cannot check i2t this way but if they messed up and it was rebadged 32mA then this should be easy to show. Just gently crush the glass, get the wire out and test it. Just a side-note: a well designed gear should also consider a scenario when you unplug device near peak (+156V) and, with cap fully charged, plug it back 8.33ms/2 later at negative peak -156V. Then the current shuld double and i2t should quadruple, reaching 1.6e-3A2s which is still much below 2e-2A2s. |
| engrguy42:
Yeah, I agree it strange that they didn't seem to even plug the thing in prior to shipping to see if the LCD comes up. But anyway, it is what it is. I'll just install a bayonnet removable fuse so if the fuse blows in the future I'll just replace it. Regarding whether the fuse might blow in the future due to this inrush thing I think is almost irrelevant now since I can easily replace it. And these things have been on the market for years so I'd assume if it was happening a lot everyone would be complaining and they'd just re-engineer with a different fuse. I'll just mark this one as a bad mechanical connection in the fuse upon receipt and that's about it. Heck, it's only a $20 device and I have 2 of them, so no big deal if one fails. We can hypothetical-ize for the next 2 months about what could happen I suppose. And, of course, I can't do non-destructive testing on the fuse since it's broken/open inside. :D But thanks, I learned some stuff today. |
| Navigation |
| Message Index |
| Previous page |