Rule of thumb for resistance of heater - when is it dead? 1

[MetalworkerMike]

In the plastic industry we use a lot of heater bands and cartridge heaters and other simple resistance heaters. It's easy to do the numbers to figure out what the resistance of a heater should be, given voltage and wattage, but is there a rule of thumb for how far away from this value the heater can go before it's 'on the way out'? When several heaters are running in parallel on the same controller it gets tricky to efficiently figure out if they're all working properly.

Mike

 

[ScottyUK]

My theory on failure mode of heaters which use a coiled nichrome element within a metallic sheath: the wire thins to breaking point due to metal loss from localised overheating and as it breaks the slight tension in the coiled element is released causing the live broken element to make contact with the earthed metal sheath, thus causing the fuse to blow or breaker to trip. No way a Megger is going to see that before the event. You need a crystal ball.


----------------------------------

Sometimes I wake up Grumpy.
Other times I just let her sleep!

 

[MetalworkerMike]

Most of the heaters in question here are mica heater bands. They're semi-flexible, but the vast majority of them stay bolted in place permanently, so there is virtually no flexure in use. It's fair to say that these bands will see no flexure at all once they've been heated once. There are some exceptions but they are relatively rare. Mica band heaters are formed by a resistance ribbon sandwiched between two sheets of mica for insulation. I think maybe the megger will be able to detect cracks in the mica. The ribbon shouldn't be able to ground out unless the mica has been compromised, so I think this method has promise. Of course, I could be wrong...

 

[waross]

I would consider trending the currents of groups of heaters.
If one heater in a group is failing, it should impact the current of the group.
The control method for the heaters will influence the trend data.
ie: On-off, time proportional, PWM.
I would expect to see a slight drop in current before a failure. Failure may produce internal arcing, which may cause a fault to ground.
The normal failure mode will be a slight decrease in current followed by no current at failure.
In some heaters the physical layout internally makes it possible for the heater to short the incoming leads resulting in a low resistance and higher current and blown fuses at failure.
I like itsmoked's suggestion. Can you give us a link, itsmoked?
respectfully

 

[itsmoked]

Looking to predict a heater failure is like the holy grail. In the semiconductor manufacturing realm they have always wanted this. A heater blowing out during a process can cost them a million dollars. They are always looking for this. It's come down to they want to know within a couple of power line cycles so they can alter the process or process flow to minimize product loss. On three occasions I have had to deliver some solution specifically for this. Often they use SSRelays and want to know the instant the heater or the SSR fails.

Bill; These are what I am referring to. You stick one on each heater leg. If it ever glows then current is flowing thru the wire it is clipped around. They work great. You can get them with remote LEDs or built-in LEDs.

See CURRENT INDICATORS.

http://dkc3.digikey.com/PDF/T073/P2249.pdf

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[MetalworkerMike]

Itsmoked: I assume you would have to remove the shielding from the wire to use this current indicator device?

Mike

 

[MetalworkerMike]

Waross: You're right of course that watching the draw of the whole batch of heaters should give you a base-line that you can trend. The problem is that some of them have only a couple of amps of draw yet they all have been standardized with 50A gauges. Seeing a variation is more than I can expect of the operation staff. Still, I will start recording values after I check all of them individually with an amp meter (to make sure they are currently working before recording the 'good value').

Mike

 

[itsmoked]

Nay. You need one wire of a heater's power wiring run thru the center of the device. It can be anywhere in that circuit.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[MetalworkerMike]

If it needs to be one wire then I _will_ have to remove the shielding, since the pair of wires are run together in the BX shielding. It would be worth it, though.

Mike

 

[davidbeach]

BX has been illegal (against code anyway) for years and years. Perhaps you have AC cable? Certainly there is a junction box somewhere that the conductors are available. If not, you will need a junction box, you can't just break the sheath in an open run.

 

[MetalworkerMike]

Dave,
When I refer to BX I'm talking about the armour on the leads to the heater band. Normally you can get the leads either with unshielded fibreglass wires, or those wires can be shielded with a stainless steel braided cover, or they can have an stainless 'BX-ish' armour on it. Maybe it's not correct to call it BX but that's what I've always called it for the sake of expedience since everybody knows what BX looks like and functions like, and this is just a version of that. I greatly prefer the BX armour rather than the stainless braid. The braid can kink at the drop of a hat... the armour is much, much more rugged which is quite important when you have the heaters mounted on a removable die, on an exposed flow tube, or on a screen changer plate where there is motion. For the sake of a couple of extra bucks per heater I always get the BX armour.

Mike

 

[davidbeach]

But it isn't BX and you wouldn't actually want BX even if it were available. There's a very good reason it isn't available any more.

 

[MetalworkerMike]

Dave, it may not be technically BX, but BX is a generic term for that kind of armoured cable, and has been for almost a century. It's not BX, it's really AC cable. If you're going to complain that my kleenex wasn't actually made by the Kleenex people but rather is a Scottie facial tissue, then that's your option, but you might want to look around to see if it's bothering anyone else.

Mike

 

[davidbeach]

I've no idea where this "Dave" business came from, but that's not me. BX and AC are (were) constructed differently. Call it what you want, but communication is more complete if it is more accurate. It isn't a matter of a Kleenex vs a Scottie, it would be a facial tissue vs a paper towel and you'd really confuse people if you mean a paper towel but talk about a kleenex.

 

[MetalworkerMike]

So someone with a screen name of 'davidbeach' who doesn't sign his posts otherwise, is at a loss as to where I got the idea that his name might be 'David Beach', and at the same time he is chiding me for my imprecise communication. Okay, I can handle the irony. I'll try not to refer to AC cable as BX cable in your presence again. I'll still call it that when I'm ordering cable, because if I don't then I'll get blank stares from people wondering what the difference is between cable for AC and DC. They shouldn't re-use active acronyms, in my opinion, but that is, of course, only my opinion.

Mike

 

[davidbeach]

My point was that I have never used "Dave" and never will, so I have no idea why anyone would think that "Dave" should be used in reference to me. There is no way of parsing my handle of davidbeach and coming up with "Dave".

 

[ScottyUK]

Mike,

David is too polite to say 'Please improve the tone of your posts to a professional level or stop posting'.

Thanks.


----------------------------------

Sometimes I wake up Grumpy.
Other times I just let her sleep!

 

[davidbeach]

Thanks, ScottyUK.

 

[stevenal]

You would not generally put a device back in service that had failed a meggar test. The test itself stresses the insulation, and I would expect a heater that had failed the test to fail to heat upon return to service. If the failure mode of the heater is an arcing ground fault insufficient to cause over current operation prior to actually open circuiting, this current has a distinct signature. Relays intended to detect high impedance faults come to mind.

 

[MetalworkerMike]

My understanding is that a megger test 'should' be non-destructive. At least, that's my understanding as it applies to motors using IEEE Standard 43-2000 (R2006), “The Recommended Practice for Testing Insulation Resistance of Rotating Machinery.” Now, if it fails the test, does the megger 'pave the way' for easier arcing in the future? Could be, I suppose. I use a megger on the motors regularly, and I suppose a non-infinity result would be considered a 'failure' in some respects... yet I wouldn't expect it to make them worse. I don't necessarily have facts to back that up, you understand, it's just a belief based on what I have seen and heard.

My use of the megger on these heaters would be to test insulation to ground, and specifically in the case of mica heater bands, looking for cracks in the mica. It may well be that the megger would be useless for this purpose... I've never heard of anyone doing this before. However, I don't mind taking the road less travelled (or even the road untravelled) if it might get me to a desirable destination.
I'll give it a try and see what happens. I'll compare the readings of new heaters to the readings of old 'condemned' heaters that are still working, but not working well, and see how it goes. If I fry a couple of heaters along the way, it's no great loss... they're as cheap as chips, relatively speaking, and the potential gain is considerable.

Mike

 

[OperaHouse]

Might be a better test if the heater was up to temperature when the megger test was done. Clearance would be quite different when cold. Or use a higher voltage, doesn't take much space for 500V.