Electric circuits in old houses--the Random Approach

my best friend almost lost his home to a K&T bad solder joint it was original to the home, had never been altered. fortunately it was in the basement cieling and not buried in a wall somewhere.

my friend happens to be a electrcal engineer, with a ton of letters after his name from Carnegie Mellon university in pittsburgh. he went down the basement smelled the tell tale odor, found the joint.

it got resoldered......

he told me solder can detoriate under time from thermal cycling........

as far as the maine challenge for practical purposes who wants to go head to head with a insurance company?

if your trying to sell a home, good luck K&T is the kiss of death, right along with flooded basements:(

stuff like you cantinsuklate a wall with K&T in it.....

Reply to
hallerb
Loading thread data ...

Folks:

Gosh, lotsa stuff. With a thread like this you just have to use the mulligan stew approach...throw anything in and see how it boils.

Zeroth...note that I am not a professional electrician, and do not have the depth of experience that this job would provide. I do have about 10 years of adding to, replacing, and upgrading electrical systems on various old houses I, my family, and friends have owned to refer to, and I do grok the relevant parts of the NEC, even if I can't cite article and section. I find it quite disturbing that on 90% of the occasions when I recommend to various DIYers I meet at the stores that they go down to the nice library and ask the nice blue-haired ladies to point them to the NEC residential manual the reply is a shrug and a return to picking out 2 1/4" deep old-work boxes. Sigh.

First...there is a 10th circle of Hell for insurance company policymakers and their one-size-fits-all Stat 101 modeling methods. Each K&T system must be considered individually by a competent electrician, like any electrical system. An intact, properly loaded K&T system, all splices properly twisted and soldered, rubber and friction tape, and properly added to is quite safe. A poorly hacked system like my house used to have, with old-school NM twisted or simply hooked on to stripped NM, 6 cables jammed into inaccessible 3 1/4" round j-boxes, hospital grade electrical tape :), all buried in cellulose insulation, is unsafe...but quite insurable, since all the *visible* wiring was old NM. Old NM, some of it beginning to pyrolize in the hot, overstuffed boxes.

Note that the K&T insulation, not so overheated, was still in good condition, though 30 years older.

Second...metal does not "crystallize", but it does fatigue, leaving a rough, "crystalline" appearance. A "cold joint", where the solder was melted and did not heat the wire enough to bond to it, has a rough, "crystalline" appearance. A properly made solder joint does not rely on the lead for conduction, but twists the wires intimately together. The lead protects and secures the joint.

Third...wire nuts can fail, yes, but in modern systems they are all stuffed in (hopefully accessible) boxes. The nut, if put on right, also twists the wires together internally, as well as clamping them with spring pressure. I also like to wrap some tape clockwise around the skirt of the nut, which a lot of people seem to hate, but I have a delusion that it adds a slight margin of safety in case the spring should weaken or the shell should become brittle or otherwise lose its grip on the spring. I have found older (bakelite I think) and ceramic wire nuts that were only held on by the tape wrapping some paranoid electrician gave them, pulling quite easily off the still well-conducting twisted-together wires. Of course those were pretwisted, but I find that Ideal wirenuts are quite capable of internally twisting 2 or 3 wires. With most other

brands I usually pre-twist.

Now for the original post. If you're not planning to do a from-scratch rewire, it would probably be best not to change much, as long as the old system is safe. There's really no way to recommend what you'd do unless I saw the system in person. Perhaps as you rewire parts of the house for convenience or remodeling it would be a good time to rationalize some of the circuits, if there's more than 2 represented in each room. Having 2 feeding each room (each circuit itself going to several rooms) can be quite nice when one of them fails, and you don't have to drag out the candles. :)

Cordially yours: G P

Reply to
pawlowsk002

Not home electrical wiring, but I spent a year or so doing stereo repair at a friend's shop. A very very common problem was failed solder connections in the stereo receivers due to the significant thermal cycling they saw. The solder would crystallize and develop a stress fracture ring around the component pin and become intermittent. This was particularly prevalent on units where the wave soldering deposited a fairly small amount of solder, resoldering manually with a more generous amount of fresh solder would fix the connections and make them more tolerant of the thermal cycling.

In an old home situation both the significantly greater age of the solder, combined with the thermal stress from applying modern heavy loads like space heaters, hair dryers, toaster ovens, microwaves and air conditioners to old 14ga wires on circuits that have been over fused because they kept blowing the proper 15A fuses has a high probability of causing those solder connections to fail if other parts of the circuit don't fail first.

I have no doubt that K&T wiring would have statistically more fires associated with it, but related to over fusing in the old edison base fuse panels and more modern "hack" add ons to the K&T wiring. A K&T installation that has not been modified and is fused with the proper fuses is in all probability just as safe as any modern house wiring. The K&T is just far more likely to have been hacked over the years.

Pete C.

Reply to
Pete C.

PC:

In this case, the solder *is* the conductor. This has never been proper practice for home wiring, where you start with a sound mechanical connection, and use the solder to seal and secure it. There should be no stress on the solder to cause fatigue, and if the joint is grossly overloaded to the point that the wires become hot enough to melt lead, the lead could melt completely without letting the joint open. You'd have other problems by this time, of course.

One problem is that the inaccessible splices may not be properly done, and another is that somebody might have hacked something on without regard to proper workmanship. But fatigue of the solder is not an issue with a proper Western Union, tap, or pigtail soldered splice. (What is often called 'crystallization' in metal due to cyclic loading is actually evidence of fatigue).

Cordially yours: G P

Reply to
pawlowsk002

The connection does not have to get anywhere near hot enough to melt the lead to cause a problem, just moderate thermal cycling can cause the solder to begin to fail.

The biggest risk is simply that with the age of K&T the probability is very high that in those 100 years or so someone *has* hacked around with the wiring and/or over fused it.

Pete C.

Reply to
Pete C.

PC:

Right, because the thermal cycling causes expansion & contraction & stresses the joint. If it relies on solder, the solder will be stressed. My point is that house wiring isn't like a circuit board, where the solder is both the mechanical and electrical connection, but is simply a mechanical, twisted joint, covered in solder.

Agreed. Unfortunately, this is equally risky with old BX, old NM, or just about anything old. It's too bad that the insurers single out K&T, in particular, by virtue of visibility, exposed K&T, which is the safest kind, since it's well-cooled and easily viewed. Models and statistics are fine to protect the profits but only an individual consideration can protect the personal safety of the homeowner and his family.

Cordially yours: Gerard P.

Reply to
pawlowsk002

...

Have you actual data to substantiate that. Having known working statisticians for severl companies whose work is such, that doesn't corroborate with what I know...

...

I think this is mostly opinion with which, for the most part I disagree...

Only a few dozens, granted... :) We averaged 4-6/year for 12-13 years. I'm sure it will depend on locale and underwriters...as noted, we did a significant number of very old dwelling renewals in a revitalization area and of all the issues there were, wiring _type_ was never an issue w/ insurance underwriters. Condition, otoh, is/was something else but in almost every instance I can recall specifically, the underwriter followed the lender's lead. Probably 80% of these were with an area/region-wide concern, so perhaps a comparison to some of the "name" guys might not be so favorable. It was also in a smaller/medium-sized market altho I don't know that had a tremendous amount to do with it, but perhaps???

Again, was never an issue that came up except for natural disasters or (later on) roofing materials. Probably, again, more to do with specific carriers and areas than anything else.

Reply to
dpb

...

Yes, the comparison of soldered wiring joints to electronics isn't very apt.

In a fairly quick search for actual data, I found none that statistically related the prevelance of fire to be any higher in K&T as compared to other forms--all data were simply reported as "fixed wiring" in the causation summaries and not broken out further.

Once analysis/review of the literature on causative factors in wiring-cause fires indicated the only real effect found with respect to K&T was owing to blanketing and the resulting induced temperature rise when the design limits of the insulation were consequently exceeded. Someone else noted about not being able to insulate walls w/ K&T and that is a valid limitation substantiated by that report. Oh, dang...I lost the link I intended to post--it was a pretty informative review. :( I'll try to find it again but don't have the time to do so right now...

Reply to
dpb

..

Unless it's already a defective join, why should there be thermal cycling? How much heat does a 14GA copper twisted wire pair at 15A generate, anyway?

Reply to
Goedjn

Doug:

200 volts...hmm...are both these 'circuits' in the same overhead box? It sounds like these were actually one multiwire circuit...and this is just what would happen with a bad neutral.

I wonder why the neutrals would fail first, though, all other things being equal. If somebody down the road had mistakenly connected both hot conductors to the same service bus, I could definitely understand it (the neutral would then be overloaded, being trying to simultaneously carry current from both circuits) but then the 200v / 40v issue would not be possible...an open neutral would simply cut out both circuits. Furthermore, with a multiwire circuit the neutral only has to carry the current that's not balanced by the 2 hot conductors, so it should be loaded more lightly.

Wire nuts or no wire nuts, I think I'd be worried about what's putting so much load on that neutral.

Did this 200v/40v issue and the neutral-overload issue happen in the same boxes?

Cordially yours: G P

Reply to
pawlowsk002

Yeah, I just add new circuits as I need'em, or shift some of the heavier stuff on to new circuits. The old wire, altho in excellent shape, is 14 ga.

Also silver coated/tinned!!! Throughout!!! This may interest you: In the 1996 NEC, under conductors, nickel-plated (iirc) wire has like FOUR times the ampacity as regular copper wire!!! Why would that be?? I know that *bare* wire is rated at a much higher ampacity than insulated wire, and I believe this was all insulated. But the nickel plating just ramps the ampacity way up! Weird? So mebbe my 14 ga is more like 12--or 10!!

If you need the exact page number, holler--it'll proly take a while for eyes to adjust to the small print. :)

Reply to
Proctologically Violated©®

I'm just recalling something I've seen mentioned before, so take this with a grain of salt...

The majority of the power transferred over a wire is carried on the surface of the wire, not the core.

Perhaps the nickel/copper combination is lower in resistance?

Reply to
Noozer

certinally 100 year old insulation can be detoriated.

its odd people who think nothing of buying new car after new car get upset that wiring wears out.

in a 100 years how many furnaces has the home had?

how many kitchen remodels?

how many new hot water tanks?

how many times has the carpet been replaced?

how many new roofs?

replaced your original water lines in a 100 years?

now why should wiring ne any different?

nearly all the items above cost many more times what new current code wiring costs....

yet people complain insurance doesnt like it.

Reply to
hallerb

Yeah, you probably heard it from me. :) But, for a while, it was looking like I was wrong. :( Lotta people on the physics, engineering ng disagree w/ surface conduction of a wire. BUT, if this is in fact the reason for surface plating wire, mebbe I was right!! And if this is what the NEC is citing, then mebbe it is the case!

Who knows.....

-- Mr. P.V.'d (formerly Droll Troll), Yonkers, NY Stop Corruption in Congress & Send the Ultimate Message: Absolutely Vote, for *Anyone BUT* a Democrat or a Republican Ending Corruption in Congress is the Single Best Way to Materially Improve Your Life entropic3.14decay at optonline2.718 dot net; remove pi and e to reply--ie, all d'numbuhs

Reply to
Proctologically Violated©®

I think you should go read 310.16 again. The difference in ampacity depends on the isnulation temperature rating (except aluminum which is generally one size bigger for a given ampacity) The "tinned" wire is lead coated and that does not change the ampacity

That is all academic in sizes 14-10 anyway since 240.4(D) rules

10ga -= 30a 12ga = 20a 14ga = 15a

There are very few exceptions to this rule. None affect a circuit with receptacles on it.

Reply to
gfretwell

Sounds unlikely.

There is a "skin effect" for AC signals but it doesn't become significant until you're into Megahertz if memory serves. At 60Hz it's probably nearly unmeasurable.

Reply to
Greg Guarino

OK, you forced me get my NEC--and my magnifying glass.

I did overstate my case, but look at Table 310.19:

For #4 wire, for example, the four ampacities are 190, 220, 150, 278, where

278 is nickel coated copper. And the 150 is bare wire?? Strangely, for #12, the ampacities are 60, 68, 35, 78.... really high! Special insulation?? For the nickel-plated wire, the ampacity is higher, but the temp is also notably higher, and the insulation is different in all cases, so it sorta seems like the plating is helping, but too much other stuff is happening.

Actually not sure if the coating on my wire is nickel or lead, but it seems brighter/smoother than what I would imagine lead would provide.

Oh well, not as dramatic as I thought... :(

Reply to
Proctologically Violated©®

My splices are twisted like I've never seen wire twisted, soldered, *wire nutted* AND taped!! Think there was a paranoid electrician at work?? Oh yeah, on plated wire! Cloth covering is *still* supple--thank gawd...

Reply to
Proctologically Violated©®

...

Certainly insulation much less than 100 years old can be deteriorated, yes.

But, you seem to have misinterpreted my post. I'm not saying it shouldn't ever be replaced and I can't (and don't) really fault an insurance company for requiring it necessarily. OTOH, if an inspection shows it in good shape and undisturbed, there's nothing _inherently_ unsafe about it which seems to be the point others always want to make irregardless of any other factor. That's all...

Reply to
dpb

You should have looked at 90.1(C) Intention. This Code is not intended as a design specification or an instruction manual for untrained persons.

The table you are looking has nothing to do with residential wiring. That nickle wire is either a high temperature fixture wire or part of MI cable (something most electricians have never seen). Niether of these are used in a home. Look at the temperature ratings of those wires. 302 degrees F up to 482F. If you have old Romex it is 60C max (140F), for that matter that is all you can use new Romex for in basic ampacity. (the first column in

310.16) The 90c column is only valid for derating purposes. As I said upthread, the over riding rule is 240.4(D) and that is the old 10ga=30a, 12ga=20a, 14ga=15a. The only exception that affects homeowners is usually the dedicated motor circuits like the condenser on a split system AC unit. In that case you can use the label on the A/C unit to select the wire and breaker combination. That was an engineered circuit dedicated for a motor.
Reply to
gfretwell

HomeOwnersHub 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.