HOME  |  NEWS  |  BLOGS  |  MESSAGES  |  FEATURES  |  VIDEOS  |  WEBINARS  |  INDUSTRIES  |  FOCUS ON FUNDAMENTALS
  |  REGISTER  |  LOGIN  |  HELP
Blogs
Made by Monkeys
Poor Lamp Socket Design Sparks Danger
4/10/2012

Return to Article

View Comments: Threaded|Newest First|Oldest First
Mydesign
User Rank
Platinum
Design Flaws
Mydesign   4/10/2012 6:06:06 AM
Jon, quality and finishing matters a lot, especially anything related to electricity. It's not safe to handle any devices made of poor quality materials, which can end up in electric shock. Proper design to handle the device is also important.

naperlou
User Rank
Blogger
Re: Design Flaws
naperlou   4/10/2012 10:08:18 AM
NO RATINGS
Jon is correct.  You really need to look for the UL, or similar, testing labels.  Cheap parts are nice in that they keep the price of products down.  On the other hand, as Mydesign points out it can be dangerous if you are dealing with electricity.

As for an new design for lightbulbs, the reason this has not been changed is that the manufacturers don't want to force the cost of an upgrade on the consumer.  This is nice, but it also perpetuates poor design choices of the past.

TJ McDermott
User Rank
Blogger
Re: Design Flaws
TJ McDermott   4/10/2012 10:17:23 AM
Naperlou, the independent party testing label is still not a guarantee of a good design, or even of a valid part.  UL is constantly watching out for parts with counterfeited labels.  The parts most commonly seen with such false labels come from overseas because it is difficult to prosecute.

Jon Titus
User Rank
Blogger
Re: Design Flaws
Jon Titus   4/10/2012 10:47:32 AM
Good point, TJ.  We have counterfeit semiconductors imported from Asia, so it's no big deal to put a UL or CSA label on a component such as a light socket, particularly if purchasing and manufacturing occur somewhere in Asia. And it's not just counterfeiting of electrical and electronic materials. The FDA has many cases of fake drugs--Avastin, for example--coming from overseas "suppliers."  In the case of Avastin, the source of the fake drug didn't even bother tp try to reproduce the type of label Genentech puts on the real drug and it still got through several distributors.

Ann R. Thryft
User Rank
Blogger
Re: Design Flaws
Ann R. Thryft   4/10/2012 1:08:23 PM

Jon, thanks for this article and even more for your comment about counterfeit UL labels. How scary is that?! I knew about fake chips and drugs, but fake electrical compliance labels is something we can't "watch out for" as consumers--how the heck are we supposed to know they are fake?


Rob Spiegel
User Rank
Blogger
Re: Design Flaws
Rob Spiegel   4/10/2012 4:02:39 PM

Not sure how drugs are examined to identify counterfeit, but the Independent Distributors of Electronics Association offers certification in identifying counterfeit electronic components. Apparently there are signs to help inspectors identify counterfeit components even though the counterfeiters are getting very clever (such as having legitimate components sitting on top of counterfeit components).

OLD_CURMUDGEON
User Rank
Platinum
LAMP SOCKETS, etal....
OLD_CURMUDGEON   4/10/2012 3:19:18 PM
There's probably ONLY two manufacturers of these devices left in the U.S. / Canada, and that's LEVITON & EAGLE.  You know that even IF they manufacture some items in China, they have rigid quslity control procedures in place to prevent such anomalies.

Jon touched on the very essence of the problem with this socket.  An aluminum shell with brass rivets.  That's the same lethal combination that was present in the late 1960s when copper was scarce due to the Viet Nam conflict, and houses were being built using aluminum wiring.  Even the revised copper-clad aluminum wiring still posed a potential threat.  IF you did any electrical work during that period, you learned very quickly to bathe the terminals in "NO-ALOX", or another anti-oxidizing compound to inhibit the formation of Al Oxide.  And, terminals on devices were relabeled w/ the CU/AL logo to identify them as safe for either copper wire, copper-clad wire, OR aluminum wire.

Jon Titus
User Rank
Blogger
Re: LAMP SOCKETS, etal....
Jon Titus   4/10/2012 4:43:01 PM
NO RATINGS
I looked for some brass sockets with brass rivets with the idea that a bit of solder would provide a safer circuit. No such sockets exist, at least on the sites I reviewed. I also thought about running a 2-56 bolt through the brass tubing, tightening it and locking it in place. I might do that if I can't find an Eagle or Leviton replacement with a solid connection. We'll see...

OldGeek
User Rank
Iron
Re: LAMP SOCKETS, etal....
OldGeek   4/11/2012 11:42:06 AM
NO RATINGS
Jon, I think you're dead on about the need for an update in socket design. I'm also pretty sure it won't happen any time soon just because of the enormous number of installed sockets. If a good design comes forward that could be easily fitted to existing sockets by using a screw in adapter it would have a much better chance.

Your particular problem is primarily because of the 3-way design. In the standard base, the rivets don't have to remain tight because the spring on the center contact keeps them held against the socket shell. Because the third contact of your socket is solid, the only thing holding the rivet tight is the springiness of the socket shell and that doesn't have enough "give" to be much help. If you can't find a socket with a spring third contact the addition of a screw might be your best shot, especially if you add a spring lock washer.

While a brass socket shell SOUNDS like a good idea, it has one major flaw: all modern light bulbs have an aluminum base. Brass sockets went out of favor soon after bulb bases changed to aluminum, probably because electrolytic corrosion started cementing bulbs into sockets.

Best of luck in your quest for the perfect socket :-)

 

Jon Titus
User Rank
Blogger
Re: LAMP SOCKETS, etal....
Jon Titus   4/11/2012 11:53:56 AM
Even aluminum-on-aluminum sockets can have problems, so i always put a thin coat of high-temperature silicone vacuum grease on threads before I screw in a new bulb.  I use Dow-Corning "High Vacuum Grease." The price ranges from about $20 to $32 for a 150-gram tube, but it will last for several generations.  I also use this grease on outdoor electrical and RF connectors to provide additional water resistance.

Larry M
User Rank
Platinum
Re: LAMP SOCKETS, etal....
Larry M   4/23/2012 10:36:08 AM
NO RATINGS
Jon Titus wrote: " I also thought about running a 2-56 bolt through the brass tubing, tightening it and locking it in place. I might do that if I can't find an Eagle or Leviton replacement with a solid connection. We'll see..."

 

Jon, I've done this. I had to drill through the rivet, thinning it's walls, for the 2-56 machine screw to clear. The repair was easy and successful.

Jon Titus
User Rank
Blogger
Re: LAMP SOCKETS, etal....
Jon Titus   4/23/2012 6:54:46 PM
NO RATINGS
Good to know, Larry.  Thankfully I found a good quality socket, but I also have some 2-56 hardware handy, too.

oldguywithtoys
User Rank
Silver
It can happen with U.S. parts, too.
oldguywithtoys   4/10/2012 4:04:27 PM
NO RATINGS
I'm convinced it's the difference in the thermal coefficient of expansion between the brass rivet and the aluminum shell.  I've replaced several Edison sockets over the years. Though the ones in the little 150W PAR cans I use in a small stage-lighting rig are ceramic rather than phenolic, the construction is similar.  Over the years I've had a couple of the replacements (from a U.S. manufacturer) fail as well, always with the same problem.  But I'm old enough to remember Edison sockets with identical construction except that the shell was also a brass, bronze or copper alloy, not aluminum.  I don't recall seeing this particular failure mode back then.

William K.
User Rank
Platinum
Lamp socket construction sparks danger.
William K.   4/10/2012 5:25:18 PM
NO RATINGS
First a comment about aluminum wiring: it is not a potential problem, it is a real problem. It fails. good jobs last longer than poor jobs before they fail, but they still fail if they are used. Unused aluminum wiring is less of a problem.

If the brass-type alloy in Jon's lamp had been the slightly springy mix, and iff it had been correctly swedged, that would not have been the part that failed. Of course the problem with the 3-way sockets is that only the center contact has any spring, and so the other contact is challenged to stay connected. It was a poor design before it ever left the drawing board. Of course the Edison screw base is a challenge, but when they are made to the original standard they work quite well, and very few people cross-thread them. The main advantage, aside from the huge installed base, is that they don't depend on material springiness to hold the bulb in place. Most alternative bases demand much closer tolerances or much better materials in order to keep the contact presure high enough. And the wonderful sockets in your car would probably not pass the shock hazard and stupid user tests, or else they would be too expensive to compete.

Jon Titus
User Rank
Blogger
Re: Lamp socket construction sparks danger.
Jon Titus   4/10/2012 5:39:12 PM
Yes, aluminum wiring is a problem. Copper has resiliance whereas aluminum doesn't. So aluminum simply compresses under a screw and offers little or no opposing force on the screw when you release torque. Thus thermal changes and even slight movement of wiring can cause an aluminum connection to fail. I know a family in Canada that has an electrician come in every few years to check its aluminum wiring for any problems. He usually findsseveral loose connections.

Instead of replacing aluminum wire you can use a special COPALUM crimp connector and an approved tool to splice a copper pigtail to the a aluminum wire and then connect the pigtail to your switch, outlet, light, fan, and so on. Many feedlines from a meter to a breaker box still use aluminum wire, but the connectors at each end handle the aluminum wire and some anti-oxidant "goo" helps protect the aluminum from oxidation.

Dave Palmer
User Rank
Platinum
Re: Lamp socket construction sparks danger.
Dave Palmer   4/10/2012 7:47:54 PM
NO RATINGS
@Jon: What you're referring to is not resillience, but creep resistance.  Aluminum alloy 1350, which is what was used for wiring in the 1960s, has poor creep resistance.  This means that, over time, it will permanently deform under load.

These days, 8000-series aluminum alloys (8030, 8176, etc.) are used for aluminum wiring.  Their creep resistance is much better than 1350, and is comparable to that of copper.

As you aluded to, aluminum has a higher coefficient of thermal expansion than copper, and this is true no matter what aluminum alloy is used.  This can cause thermal fatigue at connectors.  Oxidation is also potentially a problem regardless of the type of aluminum.  Galvanic corrosion between aluminum wiring and copper connectors is another potential issue.  So even with the improved 8000-series alloys, aluminum wiring is not without its problems.

You're absolutely right about using "pigtails" with older aluminum wire.

A good article about aluminum wire (from an insurance company, no less) can be found here.

Jon Titus
User Rank
Blogger
Re: Lamp socket construction sparks danger.
Jon Titus   4/10/2012 11:04:42 PM
NO RATINGS
Hi, Dave.  You are right about creep resistance as a factor, thanks for that information. I did mean resilience, though, as it pertains to copper's capability to "spring back" somewhat when deformed, so as you tighten a screw on copper wire, it offers resistance to its deformation that causes a tight bond. Aluminum doesn't, so creep resistance tends to loosen connections.

Manufacturers of copper wire use a "rebound tester" to measure wire resilience.

Dave Palmer
User Rank
Platinum
Re: Lamp socket construction sparks danger.
Dave Palmer   4/11/2012 10:11:38 AM
NO RATINGS
@Jon: Aluminum wire has a lower yield strength than copper wire, so you're right that aluminum can permanently deform under loads which copper would withstand.  But this, in itself, would not be enough to cause loosening. (If it were, then torque-to-yield fasteners wouldn't work).

Resilience is actually a measure of the elastic energy which can be stored in a material.  You can estimate it as the yield strength squared, divided by twice the elastic modulus.  Even though copper has a higher yield strength, aluminum has a lower modulus, so the resilience is actually pretty close -- although you're right that the resillience of copper is somewhat higher.

But when you're talking about connections loosening over time, what you're talking about is creep.

OLD_CURMUDGEON
User Rank
Platinum
Re: Lamp socket construction sparks danger.
OLD_CURMUDGEON   4/11/2012 7:57:36 AM
NO RATINGS
Jon:  Regarding the approved pigtails to convert aluminum branch circuit conductors to copper terminations ...... while in theory that seems to be the ideal solution, it's been my experience that in practice this is not always workable.  Many residential wiring plans resulted in using the smallest device boxes (least expensive!), and the MOST cables into & out of these boxes.  The result?  With so much room inside the box for grounding pigtails, circuit conductors, wire nuts, devices, etc., there isn't enough additional room to ADD even more wire PLUS "approved" crimp lugs.

And, rewiring an entire house to eliminate ALL the aluminum cable is for MOST installations an impractical solution since the original cables ("ROMEX", Type NM) is stapled at critical locations along the path.

About the best one can hope for is to periodically check heavy-use receptacles, especially those in the kitchen circuits, A/C circuits, etc. for secure connections.

How do I know this?  Because I've been through it personally AND with relatives who have owned houses with aluminum wiring.

 

3drob
User Rank
Platinum
Re: Lamp socket construction sparks danger.
3drob   4/11/2012 9:41:00 AM
NO RATINGS
When I first met my wife, her house (built in the early 70's) had aluminum wires.  My first sign of trouble was a burning smell from a little used room.  I found an outlet with a charred cover (nothing plugged into it to suggest an overloading problem, but hot enough to burn my fingers doing a touch test).  The load was somewhere else daisy chained thru that box.  It was "only" a loose connection, so tightening the screws on the outlet fixed it.  I tightened ALL the outlets/switches in the house soon after. 

The second sign of trouble was an outlet that was intermittent.  Touching a cord plugged into it could make the connection come or go.  The screws were all tight, and a new outlet didn't fix the problem, so what could be the cause?  Turns out the wiring in the wall was intermittent!  Aluminum is brittle, so a wire that was probably mishandled during installation eventually broke inside the insulation.  So, not all the fire hazards are at the junctions.  I had to replace that entire section of wiring with copper.  If I remember correctly, they have special wire nuts to join copper to aluminum and special switches and outlets to use aluminum wires.

As for pigtails, from my experience any time you handle the wire (and I'm extra cautious, not bending it more than necessary) you run the risk of breaking it.  Best to re-wire in sections as needed, or (smarter) just sell the house and leave the problems to someone else.

RNDDUDE
User Rank
Platinum
Re: Lamp socket construction sparks danger.
RNDDUDE   4/11/2012 10:12:44 AM
NO RATINGS
The usual problem with aluminum wires is due to galvanic corrosion where the wire meets the brass screws in the outlet, especially if the wiring is daisy chained. The corrosion builds an oxide layer and the resistance slowly rises at that point, causing heat, and ultimately a fire. Pig-tailing is a common cheap solution, although it only delays reoccurance. It is done usually because there is not enough excess aluminum wire to trim back to a virgin section and re-attach to the outlet, which is just as good of a solution. I suppose creep could be a contributing factor, and pig-tailing with a spring-loaded connector would probably overcome that problem somewhat. I just ended up replacing all of my aluminum wiring bit by bit as I remodeled my house.

Dave Palmer
User Rank
Platinum
Re: Lamp socket construction sparks danger.
Dave Palmer   4/11/2012 10:32:27 AM
NO RATINGS
@RNDDUDE: I agree with you; the root cause of the problem in the lamp story seems likely to be galvanic corrosion between the aluminum sleeve and the copper tube.  It doesn't help that copper is at the top of the galvanic series and aluminum is at the bottom.  They just don't play well together.

William K.
User Rank
Platinum
Aluminum wire.
William K.   4/10/2012 9:45:12 PM
NO RATINGS
REplacing aluminum wire is a profitable project when it can be done. But in many installations it does not pull through that easily. Those are the ones where I save the boxes and simply install new copper wire. Aluminum wire is for clotheslines and such, not for inside the house wiring. REally. Aluminum wire has no place in inside wiring, since copper is so very much safer. Just burning one outlet box out of a wall undoes all of the aluminum wire savings. Als, buring up electrical wiring smells bad, and it is bad for your health.

bob from maine
User Rank
Platinum
Lamp Socket Sparks Danger
bob from maine   4/11/2012 10:23:08 AM
The current path on those sockets is thru the rivet to the aluminum outer sleeve which makes contact with the threads on the bulb. The rivet to wire type of connection is also used in UL approved mechanism's such as electric motor starter circuits that use a printed circuit board. Once any resistance develops, heat follows, with hundred or thousands of cycles, eventually the resistance gets high enough to cause enough heat to carbonize the supporting plastic and restistance increases, etc. etc.. General Electric made a series of hot water circulating pumps used on furnaces from the 50's to at least the 90's and they had a brass rivet holding a crimped wire terminal against the printed circuit board. Once a little oxidation appeared the resistance went up and each time the motor started, a small arc would occur. Eventually this arc would cause the epoxy board to decompose and the contact would become intermittent. It was a real difficult problem to locate and the solution required replacing a $115 motor because the $1.50 circuit board wasn't available. Using this type of termination for components is not all that uncommon, even today.

Jon Titus
User Rank
Blogger
How many wires in a box?
Jon Titus   4/11/2012 11:38:36 AM
NO RATINGS
Hi, OLD_CURMUDGEON.  The electrical code specifies a certain size box given the number and gauges of wires in the box. So if you overstuff a box, your work won't pass inspection. Boxes should have their volume clearly marked and a calculation yields the volume needed for the wires.  Fine Homebuilding magazine ran an article that includes an example: http://www.finehomebuilding.com/how-to/articles/more-wires-need-bigger-boxes.aspx.

Bernie
User Rank
Iron
Poor lamp socket design
Bernie   4/11/2012 3:28:48 PM
NO RATINGS
Toward the end of the 1960's, aluminum was used for house wiring.  There were fires, initially in travel homes, but later, regular homes also experienced fires.  The cause was aluminum wires.  Aluminum has two properties that caused this problem.  First, it is a soft metal, and second, its oxide is a tough film, and an insulator.  In the case of the home wiring, vibration in mobile homes during delivery or driving, loosened the screws in the receptacle and socket contacts.  This was followed by oxidation and a reduction in the contact area because some of it was oxidized and became an insulator.  The result was heating at the contacts.

A similar problem arose with stationary homes.  Here, since the aluminum wires were softer than the brass contacts, if the wires at the contacts heated enough, the aluminum would deform because of the restriction of the contact screws.  When the contact cooled, the aluminum would contract and now there would be space where the aluminum could oxidize.

It is possible that this mechanism was the cause of the failed socket.

Tool_maker
User Rank
Platinum
new vs old
Tool_maker   4/12/2012 4:56:02 PM
NO RATINGS
I read so many posts in this publication about the failures of new elecric devices that I have to wonder if any manufacturer knows how to make reliable electrics today. I am a fan of estate sales and have a great deal of luck with refurbishing old stuff.

On the aluminum wiring, I heard years ago that it has a higher resistance than copper, so powering the exact same devices in a house with aluminum wire will cost more than if the wire is copper. Does anyone know if that is correct?

I also replace all "bulb down sockets" with pocelain. Ceiling fans for example and put a drop of WD 40 on aluminum light bulb bases. It seems to make bulbs easier to change.

Jon Titus
User Rank
Blogger
Re: new vs old
Jon Titus   4/12/2012 5:29:17 PM
Yes, aluminum has a higher resistivity than copper, but the resistance overall is so low compared to the load resistance that you lose an insignificant amount of energy in either type of wire. Instead of WD-40, try the silicone grease I recommended in an earlier blog post.  WD-40 is a water displacement agent used to help prevent corrosion.

Partner Zone
More Blogs from Made by Monkeys
Made by Monkeys highlights products that somehow slipped by the QC cops.
Made by Monkeys highlights products that somehow slipped by the QC cops.
Made by Monkeys highlights products that somehow slipped by the QC cops.
Made by Monkeys highlights products that somehow slipped by the QC cops.
Made By Monkeys highlights products that somehow slipped by the QC cops.
Design News Webinar Series
12/11/2014 8:00 a.m. California / 11:00 a.m. New York
12/10/2014 8:00 a.m. California / 11:00 a.m. New York
11/19/2014 11:00 a.m. California / 2:00 p.m. New York
11/6/2014 11:00 a.m. California / 2:00 p.m. New York
Quick Poll
The Continuing Education Center offers engineers an entirely new way to get the education they need to formulate next-generation solutions.
Dec 15 - 19, An Introduction to Web Application Security
SEMESTERS: 1  |  2  |  3  |  4  |  5  |  67


Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.
Learn More   |   Login   |   Archived Classes
Twitter Feed
Design News Twitter Feed
Like Us on Facebook

Sponsored Content

Technology Marketplace

Copyright © 2014 UBM Canon, A UBM company, All rights reserved. Privacy Policy | Terms of Service