GE Fused Disconnect Problems

[Jeff Wisnia]

We have a 60 amp 240 volt two pole GE fused disconnect switch in our attic supplying the air handler/auxillary electric heat for one of our heat pump HVAC systems. It is fed from a dedicated 60 amp breaker in the home's load center.

That switch has been a nuisance since the house was built about 19 years ago. Every year or so the switch contacts start heating up and will eventually heat the end carp on one of the fuses enough to melt the solder joining the fuse's link to the cap, shutting down things. The overheated fuse usually falls apart when I take it out, as its fiber tube is crisped.

I take the switch apart, clean up all the discolored switch parts with a fine file, paint some Kopper-Shield on them and put it all back together with a new fuse. My cleaning fixups last for another year or so and the same contact heating thing repeats.

About six years ago I gave up and figured maybe I just had a "bad" disconnect switch, so I bought an same model GE disconnect and just swapped in the guts to avoid having to mess around changing the housing and cable entries. The same switch contact heating problem happened again a year later.

The disconnect is in a dry area, and the switch is never thrown except when I have to fix it, so why does this happen? The current draw with the auxiliary heaters on is less than 40 amps, and as I'm using regular quick blow cartridge fuses, I doubt if there's much surge even when those heaters are cold, or the fuses would blow. After a cleanup I've let the auxiliary heaters run for ten minutes and then felt the disconnect switch parts (with the breaker off of course). They feel like they're only a few degrees above ambient then.

Is it just that GE fused disconnects are likely to be s**te, or am I possibly overlooking something?

Methinks I'll just pick up a non-fused disconnect and next weekend deep six that darned fused GE disconnect I spent an hour cleaning up this morning, when we woke up with no heat. I can't really understand why the installers used a fused disconnect there anyway. I appreciate the need for a disconnect in close proximity to the equipment, but having fuses in it when it's fed from a dedicated breaker of the same rating seems redundant. Am I right about that?

Comments?

Jeff

[Matt]

Im not a licsenced (See - I can't even spell it) electrician (personal attackes from those who are are welcome), but it seems to you have something out of balance with how the load is distributed from the fused disconnect. (Your comments about 1 fuse burning; at least one of the legs is drawing more than it should be).

Anyway - no, DO NOT replace it with a non fused disconnect, at least until you find the cause of the problem and have it corrected. Otherwise, all you are going to do is see or smell something else heating up somewhere, probably followed by your house becoming extremely warm.

[Matt]

A couple other thoughts..... I don't know if there is such a device, but it seems to me you need to measure the current draw over a sustained period of time. Something has got to be drawing more then 60A at some point, IMHO. I may have guessed that you had a loose connection or something like that.... but replacing the disconnect and having the same symptoms reappear is .... troubling.

Have you inspected the other end of the line at the heater? I'm thinking you might see some symptoms there as well, discolored/cracking insulation or something.

In any event, I say you should be thankfull that the fuse is there, instead of cursing it...!

[zxcvbob]

I think a loose or corroded connection to the wires is getting hot, and the brass contacts conduct the heat to the fuse cap. The fuse probably has nothing to do with it other than be the weakest link.

Is this aluminum wire? If so, you need to make sure the disconnect is listed for AL, and use some emory cloth on the ends of the wire, and apply de-oxide goop.

Bob

[Jeff Wisnia]

I see you missed the point completely Matt. One of the legs ISN'T drawing too much current, it's just resistive heating in the switch contacts which increases over time until it gets hot enough to make the fuse fail from heat, NOT current overload. FWIW my Amprobe says that what's flowing out of one leg is flowing into the other and verse-visa.

And, it hasn't always been the same leg which heats a fuse clip enough to make the fuse fail either.

If you can explain why Matt, I'll be all ears.

For those who don't know how a GE fused disconnect is constructed; One side of the each of the supply end fuse clips is flat, not curved, and is used as stationary contacts for the spring loaded switch arm to slide into engagement with. So, any resistive heating occuring at those contacts gets conducted directly to the fuse end cap.

Jeff

[Matt]

Why do you have to get all snooty?

I see you missed the point completely Matt.

[Fine, I won't reply to any of your questions in future, as I risk missing your point. Perhaps next time you would start out your post with: This is a question for ANYONE BUT MATT].

One of the legs ISN'T drawing too much current, it's just resistive heating in the switch contacts which increases over time until it gets hot enough to make the

fuse fail from heat, NOT current overload. FWIW my Amprobe says that what's flowing out of one leg is flowing into the other and verse-visa.

(Should have said so before you got all snooty)

And, it hasn't always been the same leg which heats a fuse clip enough to make the fuse fail either.

[Good. Then they both do it. That's MUCH better.]

If you can explain why Matt, I'll be all ears.

[Something about how there are 5 billion fused disconnects in the world, but ONLY YOURS is overheating. Dunno though, I guess you just got the same bad model, not once, BUT TWICE.]

For those who don't know how a GE fused disconnect is constructed; One side of the each of the supply end fuse clips is flat, not curved, and is used as stationary contacts for the spring loaded switch arm to slide into engagement with. So, any resistive heating occuring at those contacts gets conducted directly to the fuse end cap.

[Then why the f*ck are you even asking for opinions? Sounds like you are a master electrician, and an expert at how disconnects are manufactered. Congratulations! You just answered ALL your own questions. Next time, save us the trouble and just stay silent. [On second thought - tear out that fused disconnect and REPLACE IT NOW with a non fused one.]

[Matt]

Also, another tip, much easier, and much less expensive. Just replace the fuses with copper pipe.

[HorneTD]

Jeff Check the label on the unit to see what Over Current Protective Device is called for by the manufacturer. When the electrician installs a fused disconnect it is often because the manufacturer specified fuses rather than allowing the branch circuit breaker to serve as the air handler / heater protection. If the label does allow a breaker for protection and you can still buy breakers that fit your panel then you can install an ACHR rated breaker and change the disconnect to an unfused one. One common reason for switch terminals to fail is that the circuit uses Aluminum wire and the terminations were over or under torqued. Terminations for aluminum conductors need to be tightened with a torque wrench or torque screw driver. As others have already pointed out the stripped end of Aluminum wire should be cleaned with a stiff wire brush or with emery cloth and then immediately coated with anti oxidation paste such as Noalox®.

-- Tom H

[HaHaHa]

Jeff, any particular reason the disco is fused?

Usually, fused disconnects are redundant. Having a disco within sight of the equipment is required, but to include overcurrent protection twice (once at the main panel, and again at the disco) seems silly.

Or replace with a simple, 2-pole circuit-breaker disconnect.

[e-mail]

[Jeff Wisnia]

Thanks Bob. And no,it isn't the wire connections, they are all copper and tighter than a tick and don't show any signs of overheating. All the discoloration and oxidation is on the moving switch arms and the stationary contacts they rest against when the switch is closed.

My best guess now is that the leaf springs which create the contact force may take a set with time and not press the contacts together hard enough to prevent heating. That would be analogous to your reference to the wire connections not being tight.

What I was really fishing for was someone with wider experience to confirm my suspicion that those GE switched disconnects have a proclivity for failing that way after long periods of use without having been switched off and back on. Switching them occasionally might even help prevent the problem by mechanically wiping over the contact areas and scrubbing off the oxidation. I was also looking for confirmation that a 60 amp non-fused disconnect would be an applicable replacement since the circuit is already protected by a 60 amp breaker.

I will let the group judge Matt and his responsess for themselves, as he's convinced me of his inability to follow and understand a technical description of something as simple as overheated switch contacts caused by contact resistance. I made plain that was what was happening in the second paragraph of my OP. Further responses to Matt from me would be useless.

As for Matt's suggestion that the fuses be replaced by copper pipe shunts, that wouldn't do squat to eliminate the contact heating, it would just let that heating continue and get worse until the nonmetalic switch parts charred enough so that the switch contacts fell apart and opened the circuit.

Jeff

[Dave Morrison]

If this is one of those pull out the block disconnects, it's junk. Replace it with a real fused disconnect, one that operates from a side mounted handle. When you see the price on a real disconnect you will understand why lowest bidder contractors use the pull out style. Dave

[John Grabowski]

Jeff if there is no heat as a result of a loose connection or bad contact, I am wondering if the attic ambient temperature is a factor. Check the disconnect's label to see if there is a mention of allowable ambient temperature.

Another thing would be to check the label to see if the disconnect is rated for the type of load that you have. Also check the nameplate on the heating unit to see what it requires for a disconnect. It should say fuse or HACR breaker and give an amp rating.

One other thing would be to check the connections inside of the heating unit to make sure that they are all tight.

Have you checked starting current? Your 40 amp draw sounds like running current. The unit probably has a much higher starting load. Also, check the voltage when starting and running at the disconnect. If the voltage is too low the current draw can go up.

That's all I can think of.

John Grabowski

[Pop]

Just read thru all the posts so far and -think- I understand your = situation. It's pretty hard to tell, but one thing no one mentioned is = the "area" of the switch contacts. The smaller the contact area, the = faster it can heat up, so if there's any hot weather relationship, it's = possible the "set" you mentioned in one post was enough to actually = "offset" it a tad, and thus drop the sq. in/sq. cm area of the = connection. 40A running current is enough to cause some heat buildup = during normal operation; so, on a hot day with a start-surge, in a = super-heated attic, the buildup could take a run-away attitude. The = hotter the metal gets, the more impedance it presents via its own = properties, plus that faster oxidation can form, and off it goes. The = heating process is often a logarithmic as opposed to linear pattern, so = what takes ten seconds at first, may only take 0.1 seconds in a minute, = and so on, meaning the heat increases exponentially also. Is there a = heat-rise spec on the plate? =20

Something tells me that 40A ac might be at or beyond the "running" = (constant current) limit of the contacts and that the 60A spec is a = timed spec: eg, 60 A for x minutes or hours vs. constant current. No, = I'm not talking slo-blo type of stuff. If so, then 40A would be too = much for it. This would be worth taking up with the manufacturer or = better yet, their eng dept if you can find an "in" to them. Some = places, Square D for instance, make it easy to talk to an engineer, = others won't let you in a million years, but it's worth asking for = someone with design experience, or the actual technical design specs of = the switch. =20

Since this has happened so often, relatively speaking, it just about has = to be a design vs. use issue. Perhaps if you posted the specs from the = switch plate, someone could make a better estimate. =20 Any chance you can parallel a fan to cool the contacts when they = close, and see if that extends the time it takes, or even stops the = burning? Probably not; sounds like an awful long time between problems. = Nah, guess it wouldn't work; too close to the fuses themselves to = separate the air flow. =20

Anyway, them's my inejukatid thoughts. A little far out, but still = within the realm of possibility. =20

Pop

Jeff Wisnia wrote:

--=20

--=20 One should not be so p-h-i-l-o-p-o-L-e-m-i-c lest they be seen as disputatious.

[Jeff Wisnia]

I've been up in the attic in the heat of summer and it really isn't unbearable. Since the fused disconnect in question is rated for outside service, I'd expect its inside would get hotter that the attic if it was used outside exposed to direct sunlight, so I doubt that's a factor.

I didn't check the voltage, but as the bulk of the load is just resistance heaters, the current would have to be lower, not higher, if the voltage was low, wouldn't it?

The nameplate on the air handler says 35.5 amps for the auxillary heaters and 2.5 amps for the blower motor. that should be trivial for a

60 amp disconnect. And, if there was a huge starting surge I'd expect the 60 amp standard fuses (they're not "lag" types) would blow. They don't.

The fuse links in the two charred fuses I replaced today were intact, they just became unsoldered from the end caps which got heated by normal running current levels passing through the poor switch contacts. If the disconnect is rated for 60 amps and the 60 amp fuses don't blow, but the switch contacts fry, there pretty much has to be something wrong with those switch contacts, eh?

I checked with a electrical contractors and an HVAC guy in my Rotary Club this evening and they both said they've seen the same sort of failures in GE fused disconnects, so I guess I have my answer, and it's what I surmised. I'm convinced here's nothing inherently wrong with anything else, things work just as they should for the better part of a year and then the switch contacts start to oxidize and the problems start. When I clean up the contacts things go back to being good until about a year later.

I think I've about "saucered and blowed" this one now.

Jeff

[Jeff Wisnia]

pretty hard to tell, but one thing no one mentioned is the "area" of the switch contacts. The smaller the contact area, the faster it can heat up, so if there's any hot weather relationship, it's possible the "set" you mentioned in one post was enough to actually "offset" it a tad, and thus drop the sq. in/sq. cm area of the connection. 40A running current is enough to cause some heat buildup during normal operation; so, on a hot day with a start-surge, in a super-heated attic, the buildup could take a run-away attitude. The hotter the metal gets, the more impedance it presents via its own properties, plus that faster oxidation can form, and off it goes. The heating process is often a logarithmic as opposed to linear pattern, so what takes ten seconds at first, may only take 0.1 seconds in a minute, and so on, meaning the heat increases exponentially also. Is there a heat-rise spec on the plate?

current) limit of the contacts and that the 60A spec is a timed spec: eg, 60 A for x minutes or hours vs. constant current. No, I'm not talking slo-blo type of stuff. If so, then 40A would be too much for it. This would be worth taking up with the manufacturer or better yet, their eng dept if you can find an "in" to them. Some places, Square D for instance, make it easy to talk to an engineer, others won't let you in a million years, but it's worth asking for someone with design experience, or the actual technical design specs of the switch.

design vs. use issue. Perhaps if you posted the specs from the switch plate, someone could make a better estimate.

if that extends the time it takes, or even stops the burning? Probably not; sounds like an awful long time between problems. Nah, guess it wouldn't work; too close to the fuses themselves to separate the air flow.

realm of possibility.

All good stuff thanks!

The only offsetting thing I can say is that the auxillary heaters typically only get switched on when it gets too cold outside for the heat pump to take care of the heating requirements, and that's the time of year when the attic is cold, not hot. The rest of the year the only load is just the fan motor, a pissant 2.5 amps. So, I doubt that high ambient temperature is the troublemaker.

Rather than turn this "one off" situation into a major research project, I think I'll go with replacing the GE fused disconnect with an unfused disconnect and see what happens in a year.

Thanks again.

Jeff

[bumtracks]

I'm just curious .. if you know. what make and part number fuse have you been using

[Jeff Wisnia]

OK, I got your question in time to pull the remains of the fuses out of the workshop trash can.

The fuse which opened this time, because the solder holding the fuse link to the end cap next to the switch contacts melted, has GE's name on it and is marked 60 AMP, One-Time Fuse, 250 VAC or less, 10,000 Amps Internal Rating, Class H.

The other fuse in the disconnect, which fell apart when I pulled it out because the fibre had toasted but the fuse link was still intact and soldered to both caps, Was marked "Ace Hardware", Type NON, One-Time Fuse, 250 Volts or less, 50,000 Amps internal rating, Class K5.

Gee, maybe now that Matt guy will come back in to tell me the problems I've had with this disconnect were likely caused either by using an Ace Hardware fuse in a GE fused disconnect or the fact that the pair of fuses was "unbalanced" by 40,000 Amps.

I am always suprised to see just how skinny fuse links actually are in relation to the gage of the wires in the circuit they are protecting. I just miked the link from one of those charred fuses and its only .008" thick by .065" wide at the narrow spots. Obviously it has to be that size to perform its function, but that small a cross section sure doesn't "look" like it'd carry 60 amps continuously.

Sorry to waste so much bandwidth on this thread guys, it's my curious mind always wanting to find the "real reason" why things happen.

Jeff

[bumtracks]

[Jeff Wisnia]

Yep, I remember those.

But, I'm darn cerain I've got this one diagnosed corectly, as the other ends of the moving switch arms, the ones which mate with the contacts at the input terminals, also oxidize and show heat discoloration. And before somebody asks me again, the input wire terminal screws ARE tight, the wires in them are copper, not aluminum, and there's no signs of heating up there.

Also, the output ends of the fuses have never had any problems like that, the clips on those ends are still in their original (tin plated?) condition, with no signs of overheating there.

I also remember cartridge fuses with replaceable links, even in smaller sizes like 20 amps. I don't know if they are still made now that we've become used to living in a "throw away" economy.

Being a depression era baby, I really miss frugal stuff like that.

Jeff

[TURTLE]

This is Turtle

Jeff , These fuse disconnects are stated as rated at 60 amps but if you read the amp draw rating in the data plate with it. You will see it is suppose to be used on amp draws of less than 40 amps.

It states that it will handle short time useage of near 60 amps but when put in a contenous amp draw the 40 amps your drawing on it become the max it will handle. I have the Connecuit Electric type I use on condenser outside and they are rated at 60 amps but if you read closely you will see it says contenous use

40 amps is the MAX. it will handle. You just don't run anything at Max. By running this disconnect at 40 amps is like it says on the Lexus LS-400 Maxium operating speed is 161 M.P.H. and then you can run it at that speed and be safe. So if you don't run your car at Max. Speed 160 M.P.H , Don't run your 60 Amp disconnect Rated at Max. Amps 40 Amps.

I have a rule I use and that is I will rate the load at 1/2 the rated load on the disconnect like it rated at 60 amps on the cover. I will not put more that

30 amps on it contenously.

And yes my Lexus LS-400 will go over 161 M.P.H. at 165 M.P.H. and is the model before the Kill switches put on them. And yes I drive the speed limit most of the time and also drive my disconnects below the amp limit.

TURTLE