3
\$\begingroup\$

Electronics have a habit of breaking in my house. In the last 2 years or so, I had 3 desktops, 1 monitor, 4 cellphones going dead. They are not completely dead like they won't turn on. But in the case of cellphones, one of them end up showing only white screen at max brightness, and cannot be turned off. While another's capacitive touch screen stopped working. Just recently, my 8 months old Macbook Air is playing up and unable to charge every now and then. I haven't found a definite fix, but shutting it down and restarting, or unplugging the power cord a few times seem to do the trick... sometimes....

Being a student in Computer Systems Engineering, I am curious and suspect if it might be something to do with the nature of my house that's affecting these electronics. The first thing I thought of was surges or spike from the main AC power supply. As far as I understand, these, unless large enough, don't usually immediately kill electronics. Instead, they put strains on the system and results in significant shortening of life span. And shortening of life span is exactly what I have witnessed.

So I was wondering, are there relatively cheap and convinient ways for me to test my theory about spikes through the main? Or any other possible problem that might be taking place?

Obviously I can just be extremely unlucky, but I'd prefer to do some research and testing first, you know, engineer style.

\$\endgroup\$
  • \$\begingroup\$ Have you tried measuring the voltage in the outlets? \$\endgroup\$ – sharptooth Jul 9 '12 at 10:17
  • 2
    \$\begingroup\$ This doesn't sound like bad power because several of your device are isolated thru charger power supplies. You don't need electronics advice, you need a exorcism. \$\endgroup\$ – Olin Lathrop Jul 9 '12 at 12:28
2
\$\begingroup\$

Your appliance failures may well not be maians spike related.

Fitting surge suppressors to your outlets where expensive equipment is used sounds like an extremely good idea. These can be relatively cheap and a DIY solution is to use MOVs which are as good as many cheaper commercial offerings.


Below is a circuit and description from patent GB2231672

This is "a bit basic" but could potentially do an OK job of doing what you want.

  • A detector for detecting spikes on a mains electricity supply and for counting the number of spikes occurring during a period of time includes a zener diode (Z2, Z3) connected in series with a resistor (R4) and a counter (IC1) having an input connected across the resistor.

    The zener diode and resistor are connected in series to the mains supply and the zener diode has a breakdown voltage of a magnitude such that under normal conditions it does not breakdown but upon occurrence of a spike in excess of a predetermined magnitude the zener diode breaks down and provides an input to the counter.

enter image description here

http://worldwide.espacenet.com/espacenetImage.jpg?KC=A&CC=GB&FT=D&NR=2231672A&date=19901121&locale=en_EP&flavour=firstPageClipping


This interesting combination circuit from here detects spikes in a similar manner to above but also adds "flutter" and noise". They drive LEDs but you could add a counter or other monitor. Note that unisolated connections to mains can kill with ease.

enter image description here

\$\endgroup\$
-2
\$\begingroup\$

Any cause is possible, but I think different component failures occurred in each case due to different stressers or root causes.

  1. Cell phones can be sensitive to rain or >90% RH at high temp if no watertight seal on critical connectors. I also depends on amount of residue in solder flux & dust ingress also if they use "no clean process" or "with clean." Some gaps under chip are hard to clean. The internal ribbon connectors can also become displaced when the phone is dropped. Also weak solder joints or heavy parts on board such as display can break joints if cell phone is dropped. It is hard say what stress phone has seen, just maybe.

  2. Mac Air magnetic DC plug is excellent but is hard plastic and the DC wire is soft flexible wire and jacket. Sharp bends on wire at plug and frequent pulls are most likely stresser. I happen to own an 1 yr old Mac Air and have 2 different DC chargers and both are weak in this strain relief design. They opted to make plug too easy to connect with magnet and disconnect by yanking and inadequate strain relief on hard plug to soft jacket interface. I wrapped both of my plugs with stiff tape ( not duct tape) and even with that one is failing at the wire plug interface. ( I am hard on mine )

Graduated stiffness so there is no sharp bend in wire. Sharp bend is anything with bend radius < 20x wire jacket radius is OK but with frequent flexes. premature failure will occur. They use many awg 48 stranded wires thinner than human hair to make very flexible cable, But it can break without obvious physical symptoms.

It may be too late to fix. I just noticed same Mac Air DC charger failure symptoms on one of my charger as yours and added a few wraps of red Tuck packing tape around wire next to DC plug. So far so good, but next time I have to be more gentle. No more accidental plug yanks. ( which disconnect so easily with magnetic plug but overstress wires.) It is deceptively smooth, a good design but with a fault for heavy users who are mobile with charger. Stationary chargers would be even better, if you could.

The same is true for headphone, earbud and microphone cable and cell phone charger cable users.

If they stayed on back of desk, between uses,they might last for ever. But if stuffed, moved back and forth daily with mobile useage, maybe last < 6 months. It depends on strain relief near both ends of wire and frequent flexes < 20x thickness.

The MTBF stress factors for cable depend on bend radius ratio and frequency of radial and axial stress per day.

Cell phones are design to withstand the international standards for ESD > 15kV. If you get a lot of lightning in your area, get a few 25 cent coin size MOV's for your extension cords and make a plug with one. Your power meter in the home will suppress transients> 6kV, but yet the industry only certifies consumer equipment to 1500V or maybe 3kV since 6kV costs more with line filters. But cheap protection is an MOV rated to clamp just above your max line voltage. It is good for 10 or 20 large surges.

**You do not need a fancy spike meter unless you are really keen. But you may want fancy line filter or just a cheap one.

A cheap spike meter is ac powered piezo alarm with the switch turned on but the wire broken and splice with exactly a 1mm air gap and sealed with tape. THis will yield a 1kV arc gap +/-25% 1kV/mm for smooth gap. So when line transients occur > 1kV you will know you need an expensive $25~$50 line filter with 1500V protection using CM choke, diff choke, MOV, TVS and gas tube with current limiter.**

MOV's are cheap 25 cents typ. http://www.digikey.ca/product-detail/en/MOV-10D431KTR/MOV-10D431KTRCT-ND/2408233

DC chargers may be prone to breakdown with spikes burning enamel insulation but more often fail due to self-heating or lack of free air flow (stuck under blankets) and quality of vendor with poor vacuum processing of injection molding.

\$\endgroup\$
  • \$\begingroup\$ The problem with surge protected power bars is they charge twice as much as unprotected bars yet only add a non-replaceable MOV which costs them at most few pennies. There is no simple reliable way to verify how many zaps have conducted and how many more it can handle, but they are cheap to buy and essential in areas with exposed line voltage distribution outside in high lightning prone areas. So pls reply if I identified any root causes that fit you. \$\endgroup\$ – Sunnyskyguy EE75 Jul 9 '12 at 13:55
  • \$\begingroup\$ Another problem is that MOV's require some resistance, usually specified as some minimum length of wire. They probably won't work if the outlet only has 10 feet of 12 guage romex. \$\endgroup\$ – Erik Friesen Jun 18 '13 at 23:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.