blowing fuses

Do you use a toroidal transformer ?

Do you have large storage capacitors in the power supply ?

If the mains source impedance at your premises is high (long and/or thin cabling from your distribution transformer to your socket) and at the customer's premises, the mains source impedance is low (short an/or thick wires), much higher peak currents will flow in the latter case.

In order to charge a big capacitor from a high impedance source, the charging time is long, but the peak current is mediocre and the heat

Now assume the source impedance is halved, double peak current would flow, but the capacitor would be charged in half the time, thus the

the first case. Thus no wonder that the fuse blows.

A soft starter, such as a VDR might help in places with low mains source impedance.

You need them to put a power monitor on their line or take a picture of the AC via a scope.

There are a few things I can think of that can cause that, most of which is at the customer's end. Non Sinewave power, Severe harmonics, saturated line. etc..

Also, have you tested your unit under worse conditions of use ? They could be miss using it, causing some unexpected over load on your part of the circuit.

Jamie

That's 90+25= 115W, which at 50% efficiency would make 230W for nearly 2A *in-phase* line current. I'm pretty sure the Littlefuse recommendation would be a minimum of 2x 2A= 4A slo-blo. The 2A is a bad specification.

bloggs.fred...

** The efficiency at full load is gonna be well more than 50% - more like 70 to 80%.

However, supply current is the RMS value of the peaky waveform DC supplies draw from AC which makes for a poor power factor, typically about 0.65

So the calc is: 115 divided 0.7 then divided again by 0.65

Gives 252VA draw = 2.1 A at 120VAC.

The 2AT fuse rating is correct.

.... Phil

A common misconception is that the purpose of the fuse is to protect the network _after_ the fuse.

In practice, the purpose of the fuse is to protect the _feeding_ network, not the load !!

A device using a razor cord should be ne able to handle the short circiuit current by the fuse protection.

If it has an iron power xformer the reason is clear. It has to do with core saturation. If the unit by chance is turned OFF when the AC Cycle is nearpeak, the flux will be retained in the core. If at the next turn ON the AC wave is again near peak, The core can go into deep saturation. This will cause a very large turn on current surge. The usual soft start strategies can help solve this. Brute force method is to use more iron which is prob not An option for you. M

I have known the inrush to crude rectifier and capacitor combinations to be huge, where NTC thermisters have been fitted, but admittedly not generally on 100W linears.

I'd still think its likely a fast blow fuse has been fitted in error.

--
Mike Perkins 
Video Solutions Ltd 
www.videosolutions.ltd.uk

The fuse is not selected to indicate device out of estimated operating area, it is selected to clear a fault. If you compute up to 2A worst case in normal operation then 2x or 4A is a more appropriate value for fault protection.

bloggs.fred

** The fuse is likely the only protection the AC tranny has - so you are wrong in this case. ** That is not the case here, the unit is a PSU and the max operating current is 100% known. A 2AT fuse should not fail in normal operation, ever. So if it does fail - there is a fault.

.... Phil

The OC devices in your house are intended to protect the wiring

*after* the OC device. Fuses in appliance are intended to protect the appliance (keep smoke in). The wiring on the appliance is sized to its OC device (if there is one).

I have no idea what the above means.

Obviously it can be either.

NT

** That's no misconception.

Fuses in distro panels are "cable protective" so rated to suit the particular cable used AFTER them.

** That's kinda automatic.

.... Phil

Examples please.

George, What you are talking about above happens when you have what is called a "multi-wire branch circuit" with a bad neutral. (A multi-wire branch circuit is also called a "shared neutral branch circuit".)

A schematic is worth a thousand words:

------- ------- | CT |-Hot "A"-| Bldg |-Hot "A"--[LOAD A]--+ HV===| Pole | | Entry | | |xformer|-Neutral-| Panel |---bad neutral-- X -+ | street| | | | | |-Hot "B"-| |-Hot "B"--[LOAD B---+ ------- -------

Under normal circumstances (ie the neutral is not defective) current in either load returns to the entry panel via the neutral, and each load "sees" 120V. But if the neutral is defective, both loads are in series with 240V. Each load will "see" 240V - Vdrop in the other load.

Say Load A draws 1 Amp normally, and load B draws 10 amps normally. That means A impedance = 120 ohms, and B impedance = 10 ohms. If the neutral opens, about 1.8 amps will flow through each load, way over the normal current for load A, and way under for load B. The neutral can go high resistance at a loose connection instead of open which changes the computation, but you still get the situation where one load gets too much current and the other gets too little.

Ed

That's not how fuses are selected. Fuses are relatively crude and selected to protect against a hard fault. Here's one method of selecting slo-blo:

All of this is derated with increasing ambient temperatures, meaning you could end up with a surprisingly large fuse in some applications. Now if that is not enough protection for the transformer then you have to supplement with a thermal or something, but messing with the element fuse is not the way to go.

Lots of PSU iron transformers are protected only by a mains fuse with or without a secondary fuse. Ideal or not it's common enough.

NT

Yup, I'm certainly part of the "human error" equation.

Yeah I offered to send the right fuses, next time I'll be more forceful.

George H.

2A is what is recommended on the big power supply. (We've sold ~200 of these, and this is the first time I've had a fuse problem.)

George H.

I'll try again

see the section on inrush current

m