1.25*FL for standard fast blows. 1.10*FL for slow blow fuses.
Problems exist with switching supplies or those with large caps. They have lots of inrush currents with the large caps and many designs do not have a delayed start circuit, something you find in properly designed power management circuit.
Jamie
OK. Lots of good pointers. Just to clear a few points up. I use the word "friend" to indicate someone whom I know extremely well, and whose qualifications I know intimately. This particular guy is a long-time designer of electronic equipment, but his speciality is logic and microcontrollers, so he is not well versed in the black art of power supplies and fusing them. This is part of the reason that he chose to use a pre-fabricated brick for the PSU, the other part being that such an item already has all of the safety and EMC issues taken care of. For those that were interested in the brick type, its basic specs can be found here :
along with a link to a detailed pdf. To address a couple of questions that Jeff had, The device across the front end is definitely not an NTC inrush thermistor - which would, of course, be in series with the mains, not in shunt as the device actually is. Be assured that it *is* an MOV. Find it here :
As for any problems being caused by whiskers, there aren't any. I know this, because this particular one was built by me ... I'm pretty sure that I had mentioned the fuse type and rating elsewhere in the thread, but if anyone missed it, the originally specced fuse was a 20 mm glass cartridge type with a rating of 200 mA and an "F" fusing characteristic. He arrived at this value by measuring inrush and running current. I don't know what the actual figures were, but knowing the type of person he is, they will have been measured very carefully, and the fuse value will have been derived by looked-up theory. However, from my personal long experience at the sharp end of mains-powered equipment servicing, no matter what the theory might indicate to the contrary, in practice, 200 mA is never going to be enough to guarantee reliable operation, nor is making it an F type. I have now fitted a T400 mA, which seems an altogether more practical value.
Having now discussed this whole issue at length with him, he is in agreement with what I have said, based on my own experience, and the input from the good folk on here. He says that he is inclined to say that he was erring on the side of possibly excessive safety, with not being well versed with this aspect of design. I have put it to him that if any items fail in the front end of the brick i.e. the bridge, main filter cap, or switching element, or the MOV fails as I have seen on occasion, the fuse is gonna blow in pretty much any value short of a 6" nail ...
Arfa
Ahah! Numbers finally. Model 47122 is listed on the data sheet at:
5v DC out 550ma DC out Pout 2.5 watts Efficiency 65% Temp 70C So, if the device runs at full load (a bad assumption but lacking anything better, I'll use it) the input power is: 2.5 watts / 0.65 = 3.9 watts For a nominal 230VAC input, that's a current draw of: 3.9 watts / 230V = 17 ma Worst case is at the low end (-15%) of the voltage input: 3.9 watts / 196V = 20 maA 200 or 400ma fuse should work just fine for that as long as there's no inrush current. Again, in the interest of blowing it up now, instead of in the customers hands, I suggest you short circuit the output of this 5V supply and see how much current it sucks. If it's any good, it will shut down. If it's a piece of junk, it will suck many amps. Note that if you use commodity electrolytics for power supply bypass caps, you'll probably have at least a few power supply shorts.
On to the MOV:
Rated at 275VAC. MOV's conduct on the voltage peaks. Peak voltage for 230VAC is: 230V * 1.414 = 325V peak. It's going to conduct even with normal AC power. That's why the fuse blew (and will blow again no matter what fuse value is used).
So, let's do the math. Your power line voltage is likely to vary at least +/-10%. +/-15 is unlikely, but possible. I suggest a maximum voltage of: 230V * 1.15 = 265V Peak voltage is: 265V * 1.414 = 375V The nearest value above 375V on the data sheet is 385V.
To calculate the required power handling capability, I would need to know the total resistance from the power plug, up the cord, through the switch, and through the fuse. The inverter module is not involved. It's probably easier to measure it. Typically, you'll get about 0.3 ohms but it can vary, especially if you use a slow-blow fuse. The MOV would be expected to handle: 230V / 0.3 ohms = 766 Amps Therefore, the 1200A device (S07K385) would be the minimum.
Now do you see why I like to see numbers?
-- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
"Arfa Daily"
** See:Note the line:
" Input protection by integrated fusible resistor "
This will likely be the only component that limits the inrush surge - but what value is it ??
Going on many small SMPS and CFLs, the value used is likely around 22 ohms.
Now, here is a crucial bit - what resistance is a F200mA fuse ?
Answer: 10 ohms or more ( I measured a few).
Wanna guess the value of the filter electro inside that 5V brick ?
It works OK at 47 Hz and down to 80 volts AC so how about 22uF ??
Now some very simple math: the worst case surge is when the switch closes at peak AC voltage or 340 volts in the UK.
10 plus 22 = 32 so Ipk is just over 10A.The surge duration is a couple of time constants or about 1mS.
So, that poor little F200mA fuse has to dissipate circa 1000 watts for about
1mS.Or about 1 joule.
That is RIGHT on the edge of what it can take, as 2 watts causes failure in
0.5 seconds.... Phil
You see Philip ? You really *can* talk sense, when you put (what's left of) your mind to it ...
Arfa
If you read the data sheet for the item, it would suggest that an output short forces a shutdown of the supply, from which it "automatically restarts when fault condition is removed". This would be in keeping with pretty much all modern switchers that I see, where it is easy for the designer to use a current monitoring pin on the control IC.
I'm not sure that I am following your reasoning there. If the device is rated at 275V AC, that equates to 388 V peak, doesn't it ? So that would seem to be rated exactly correctly to do the job it was intended to ? Clearly, the fuse is not blowing all the time "even with normal AC power", otherwise I would not be discussing the one failure so far, on here. Am I somehow misinterpreting what you're saying ?
See above
Well, yes, but I'm not sure that I am quite understanding what you are doing with them ...
3
ut
es
ef)
Phil talks sense pretty much all of the time, but he does intersperse the sense with a great deal of unnecessary personal abuse.
This is not uncommon on this user group.
Our most prolific poster - John Larkin - is a little more prone to post nonsense than Phil, in part because he also posts about non- electronic matters, and spends less of his time on the unnecessary personal abuse, but he still goes in for it.
The next most prolific poster - Jim Thompson - is pretty reliable on integrated circuit design (which is where he makes his money) but reliably off the wall on all other subject, which is why I call him Jim-out-of-touch-with-reality-Thompson. He's not all that enthusiastic about personal abuse, but he claims to have reported me to the US FBI for my "dangerously anti-American attitudes" which is probably more worrying. Since I once had a security clearance that got me into US Army ECOM at Fort Monouth, NJ, in 1970 he probably destroyed his credibility with the FBI in the process.
The third most prolific poster has stopped posting here, and will shortly be overtaken by fourth most prolific poster - me - unless I suddenly find something more useful to do with my - excessive - spare time.
-- Bill Sloman, Nijmegen
Me and Phil go back a looooong way. We have had many spats over the years over on S.E.R. I don't take that much notice these days
It's a shame really, as most of the time, he talks a lot of sense, but occasionally, just 'goes off on one' for no apparent reason...
Arfa
"Bill Sloman" "Arfa Daily"
** My god, spelt my whole name - correctly.Will wonders never cease .....
Phil talks sense pretty much all of the time,
** And the rest of the time, it is merely because Buffalo Bill misses the point.Bovine creatures have that propensity ....
Our most prolific poster - John Larkin - is a little more prone to post nonsense than Phil, in part because he also posts about non- electronic matters,
** GIANT huh ????Farkin Larkin posts so many absurdities about " lectonics " - the posturing fool could make a damn good living as a stand up comic in Silicon Valley.
Have the all drooling code scribblers and manic wafer bakers in permanent stiches !!
.... Phil
Wow, you're about to pass Rich Grise? I hereby christen thee Bill the Philosophiser. ;)
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 845-480-2058 hobbs at electrooptical dot net http://electrooptical.net
t of)
the
Just this once you don't know what you are talking about. Rich Grise hasn't posted enough to qualify as a prolific poster - number three was Eeyore/Graham from the UK.
And what I posted wasn't philosophy, but observation - perhaps verbose observation from a particular point of view - and none of the philosophers I've known would have said any different. If you want to argue, identify the philosophical proposition being advanced ...
-- Bill Sloman, Nijmegen
The 275VAC is an AC rating. MOVs are rated differently from Zener diodes, but similar to TVS suppressors. They can be used at the AC voltage they are rated for. The clamping voltage with significant current is going to be way higher.
Take, for example, a V275LA40AP 20mm 275VAC varistor- the voltage at
1mA current is between 389 and 473V, and it will clamp to maximum of 710V at 100A.
BTW, one thing that has not been mentioned is the interrupting capacity of the fuse. The glass 5x20mm fuses have pitifully low interrupting capacity (like 35A) and are not really safe for anything but something at the end of a long cheap line cord and 120VAC. Any kind of 240VAC industrial situation where thousands or tens of thousands of amps might be available may need a better fuse. With enough fault current, they arc across the ends and the glass literally explodes.
Yep. That's what the data sheet says. All you have to do is verify that they're telling the truth. I've been surprised when the protection circuitry goes into oscillation.
Oops. Y'er right. The data sheet at:
has two columns. One for AC the other for DC. I was looking at the wrong column. 275VAC * 1.414 = 325V Peak, which is close enough to the 350VDC rating.
However it's still too low. The highest power line voltages (+15%) it might normally see is still: 230VAC * 1.15 = 265VAC which is still uncomfortably close to the 275VAC MOV rating.
However, that still might be too low:
indicates that 117VAC US surge protectors use 330VAC rated MOV protectors. For UK 230VAC, that would be double or a 660VAC MOV rating. (I would check some power supplies and power strips, but I'm at home and won't be in the office until tomorrow).
Try the varistor selection guide:
The maximum current calculation is easy enough. When the MOV conducts, the fuse should blow, not the MOV. By knowing the DC resistance of the power wiring, switch, and fuse, you can calculate the maximum current through the MOV. The MOV should be sized to handle this current without exploding. You can also calculate the amount of energy you would expect the MOV to handle, but you would need to know the voltage and duration of expected glitches.
-- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
Yep. However, I seem to be missing something here. I've always assumed (the mother of all screwups) that the rated voltage printed on the MOV is the DC voltage. However, looking at the data sheets and your explanation, it seems to be the AC rated voltage. Yet, when I go to a page on calculating the required voltage rating:
I find that they're multiplying the applied RMS AC voltage by 1.414 to get the peak voltage which then becomes the "required varistor voltage". If that calculation is correct, it should be the DC value that's printed on the MOV. Something is wrong here.
Ok, a rather soft knee. However, when conducting, it only has to conduct enough current to blow the fuse. Looking at Fig 6 for the
320VAC device, the 400ma fuse will blow at 650V-Pk or 460VAC-RMS. That's twice the nominal 230VAC which should work (ignoring the up to 25% voltage tolerance which I haven't tossed into the calcs).More later. I'm late...
-- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
Now why would you say that ? I ALWAYS spell your name correctly when I use the whole thing ... To do otherwise would be insulting, now wouldn't it ? d :-)
Arfa
Yay ! Whatever happened to Graham ? Is he still about ? Just a lurker now, maybe ? He just disappeared like a puff of smoke a couple of years back, after being, as you say, a prolific poster on several 'lectronixy groups for many years.
Arfa
This item is consumer-related. It will be permanently wired in (via a short flexible cord) to a standard UK ring-main circuit with 30 or 32 A MCB protection back at the board
Arfa
I thought most of Detroit was already lost? ;-)
Not to mention DNA..
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.