Can loose connections overload a fuse?

I have an SCR sending 70 volts AC into a resistance heating wire embedded in my bathroom floor, and it has been working just fine for several months. The wire has a resistance of about 14 ohms, and that amounts to a load of around 5 amps at 70 volts. A few days ago, the 10-amp fuse at the SCR's output blew, and I am trying to figure out why. The wire in my floor is still open to ground and the only problem I could find anywhere is that the screws in the SCR's output lugs were barely finger-tight. I do not know how they came to be that way, but that is a different matter.

Before placing another 10-dollar fuse in the clamps:

Is it possible that fuse blew because of those loose connections?

Note: What you see in the picture is *not* a completed project!

Fuses are thermal devices, loose connections result in heat. A loose connection resulting in heat in close proximity to the fuse results in the fuse opening

BrianJohn said:

Fuses are thermal devices, loose connections result in heat. A loose connection resulting in heat in close proximity to the fuse results in the fuse opening

Click to expand...

I understand ... but would not proximity make a difference? My output terminals and the fuse are 5 inches apart, the conductors between them are mere half-inch-wide strips of film on an electronic circuit board and the unit has overheat protection built into it.

Can fuses blow more easily simply because they are fast-acting and/or old? The specific ones I have used so far are Shawmut AmpTraps, and I also wonder whether their fast-acting nature makes them actually physically fragile. Of the two I purchased online just last week, one of them arrived bad (open).

At the moment, I have a *variac* temporarily connected for testing the heating wire and everything is fine, but I have the variac set to only 50 volts since it is too small for the wattage of 70 volts at 5 amps. In a few minutes I will be heading out to get some kind of clamp-on amp meter if I can find one at one of the box stores.

Fuses have numerous classes, some are fast acting, some are time delay, depending on what the engineers dictated was necessary to minimize damage to equipment, but allowing safe operation of the equipment to operate.

As for proximity of loose connections this depends on the radiating properties of the conductors (heat sink), temperature obtained and load at time of the fuse operation.

Additional fuses and circuit breakers can deteriorate due to what is commonly referred to as metal fatigue. Numerous cycling of loading unloading, heating cooling.

If you can apply 50 VAC to the load, and the load ohms out (did you do a pre-installation ohm check?). It sounds like the load may be OK. Try different size fuses at your 50 VAC input, you should have an idea what it would draw at 50 VAC .

It is possible that the loose connection caused some kind of transient in the controller device that caused failure of a component, which caused a high load which caused the fuse to blow.

Transients related to opening circuits can cause high voltages due to interruption of inductive loads when the interrupted current tries to get where it is going. That is the principle that makes spark plugs work.

Opening the circuit could produce a high voltage that would damage a diode in the rectifier circuit, perhaps causing a short in the rectifier or DC-to-AC conversion.

I suspect that if you replace the fuse it will blow again; and that you will find that the controller is damaged.

BrianJohn said:

If you can apply 50 VAC to the load, and the load ohms out (did you do a pre-installation ohm check?). It sounds like the load may be OK. Try different size fuses at your 50 VAC input, you should have an idea what it would draw at 50 VAC .

Click to expand...

Yes, my heating wire reads around 14 ohms (as it always has) prior to any connection, and my new clamp-on meter shows about 6 amps when I temporary bump the variac up to 70 volts. I do not have any fuse in this testing setup with the variac, but everything seems quite okay.

At about 46 volts going in, I get the amperage readings you can see below in the picture ... and keep in mind that the wire reads completely open to ground when it is not energized.

At this point it appears you MAY have done all the preliminary testing and are ready to do a go no go test commonly referred to as a SMOKE TEST.

Bob NH said:

It is possible that the loose connection caused some kind of transient in the controller device that caused failure of a component ...

I suspect that if you replace the fuse it will blow again; and that you will find that the controller is damaged.

Click to expand...

We will find out in a few days when some more fuses arrive, and I again thank you for your help in my original engineering of this system as well as for all of that info about what might have happened to the SCR. The tech guy on the phone a few days ago insisted I must have (had) a short of some kind in my heating wire or its related wiring, but I know that is just not the case. Oh, and he liked the fixed resistor replacing the potentiometer.

And whew, those fast-acting fuses specified by Payne are both difficult to find and expensive! A big supply house nearby that has been here for ages does not have anything even close and Payne only sells them by the box (10). After a couple of hours of searching, I finally found a Bussman FWH-10 online somewhere for a little over $25, but then I found some of the Shawmuts at my favorite auction site for a price that ends up being about $13 each, including shipping.

BrianJohn said:

At this point it appears you MAY have done all the preliminary testing and are ready to do a go no go test commonly referred to as a SMOKE TEST.

Click to expand...

Yep, and I thank you very much for the help setting it up!

leejosepho said:

We will find out in a few days when some more fuses arrive ...

Click to expand...

... but first another fuse question if I can figure out how to ask it ...

Can all fuses be checked for continuity prior to installation?

That question sounds so silly to me, but I now have three more fuses and I am getting all kinds of readings while trying to check them. After taking them out of a USPS box that had been sitting outside in the cold all day, my meter said each of the three was "open". After they warmed up a bit, I began to *occasionally* get some readings, but the readings I did get were all over the place between 2.4 ohms and 3600 ohms ... and they changed or even went back to "open" whenever I finger-tapped on the fuses just a bit.

At the moment, only one fuse will give any readings at all, and those range from 1.1 to 4300 ohms.

I cut one of these fuses open, and they have some kind of laminated (riveted) strip inside that looks like two layers of fiberglass with something in-between. Also, these fuses are filled with some kind of white granules.

Am I ignorant of something about certain types of fuses reading "open" until they have more current going through them, or am I correct in believing all of these I have are bad? They all came from a salvage dealer at my favorite auction site.

Thank you, gentlemen!

All sorts of fuses out there, and depending on their size, rating, and type, plus the type of meter you have, you could get different ratings...they'll rarely read zero ohms if you have a meter that will read small values. Fast blow, slow blow, high voltage, etc., types all are built a little different. I'm not aware of any real fuse that changes a huge amount if you shake or tap it, though, unless there is a poor connection which might inidicate it is bad. I don't think it should ever read an open!

jadnashua said:

I'm not aware of any real fuse that changes a huge amount if you shake or tap it, though, unless there is a poor connection which might inidicate it is bad. I don't think it should ever read an open!

Click to expand...

That is what I thought, and I thank you!

I took the one that gave the best reading most often and gave it a try in my variac-powered testing setup, and it only lasted about 5 seconds at 50 volts (about 4 amps). Since then, I have discovered those fuses have not been available since sometime in 2002. Being that old or older and coming to me through a salvage dealer, maybe they had once been wet or something. So, still looking for a fuse ...

Are you sure they are really fuses and not a self resetting device? On the one you took apart, does it look like there are movable contacts somewhere on that bimetallic strip? Have you looked at the manufacturers specs of the device?

I have never really run in to "shelf life" problems with fuses. Shipping could be another issue. I don't suppose the guy selling the pulls checked them before sale. A warranty on a fuse would be real close to silly.

It really does feel like something has failed in the controller. Do you have a resistive load that you can use that would not require removing the floor if it breaks?

Attach a load and see what the voltage is. Start low and see if you have any control. Don't put a fuse in the output leg. Is the primary side of the controller fused or on a breaker? If you try this be careful you don't make a big mess in the controller. If the SCRs failed (or the control of the SCRs failed) so they are full on you would get 120V across the load and this would be consistent in terms of blowing 10 A fuses.

Have you measured the output voltage of the controller? You could use a regular incandescent light bulb (hundred watts) as a test load. The bulb will keep the output current from exceeding rated value. You will probably be able to adjust the voltage; if not and you just get 120V it is the controller.

alternety said:

Are you sure they are really fuses and not a self resetting device? On the one you took apart, does it look like there are movable contacts somewhere on that bimetallic strip? Have you looked at the manufacturers specs of the device?

Click to expand...

Yes, no and yes. They are fast-acting fuses specifically intended for SCRs and the like.

alternety said:

I have never really run in to "shelf life" problems with fuses. Shipping could be another issue. I don't suppose the guy selling the pulls checked them before sale. A warranty on a fuse would be real close to silly.

Click to expand...

I believe they are NOS rather than pulls, and the guy sent a message today saying the rest of them had been checked and about half were definitely bad. He is willing to give me a complete refund, but I told him to go ahead and send four more since they have now at least been checked.

alternety said:

It really does feel like something has failed in the controller. Do you have a resistive load that you can use that would not require removing the floor if it breaks?

Click to expand...

Yes, I have another piece of heating wire I can use, and I thank you for that suggestion since I had not thought of that.

alternety said:

Attach a load and see what the voltage is. Start low and see if you have any control. Don't put a fuse in the output leg.

Click to expand...

Yes, I plan to remove the fixed resistor and use the original potentiometer to begin at zero and go up, but I will have to have a fuse in the SCR output to get anything at all.

alternety said:

Is the primary side of the controller fused or on a breaker?

Click to expand...

It gets its power from an electronic thermostat that gets its power from a breaker. The thermostat "pulses" or whatever to conserve power when it believes it can, but the tech at Payne told me that is no problem for the SCR.

alternety said:

If you try this be careful you don't make a big mess in the controller. If the SCRs failed (or the control of the SCRs failed) so they are full on you would get 120V across the load and this would be consistent in terms of blowing 10 A fuses.

Click to expand...

"Full on" would be too much for my 14-ohm heating wire, but the total load at 120V would still be less than 10 amps.

alternety said:

Have you measured the output voltage of the controller? You could use a regular incandescent light bulb (hundred watts) as a test load. The bulb will keep the output current from exceeding rated value. You will probably be able to adjust the voltage; if not and you just get 120V it is the controller.

Click to expand...

The SCR needs a minimum resistance load to work at all, and I do not know that number at the moment. Maybe a couple of bulbs would be sufficient, or maybe not. The last (and only) time I have tried the SCR after its OEM fuse blew, the output was the usual 70 volts as established by the fixed resistor. The NOS-salvage fuse being used at that time blew in about 4 or 5 seconds, but that was before I had discovered the loose output connectors.

I will have some more fuses late next week, and a distributor in Louisville has told me where I can get (or probably order) some truly new ones locally. For now, I will check prices at those places while hoping at least one or two of the NOS replacements being sent are actually good!

You don't need to put the variable control back. Use it as it is with a load small enough to not stress the limit of the supply at 120V. Then measure voltage. If it is 70V that tells you that a lot of the controller is working properly.

If you have enough resistance wire that is probably better than a bulb if you have to have a fuse in place. Resistance goes up as the load heats. In a light bulb it is a fairly large change because of the temperature change range and may cause short term surge current that makes something unhappy. That was why I suggested a smaller bulb.

If the controller is old (and the fuses) clean the contacts on both. As said before in this thread, heat generated where the fuse plugs in is more likely to impact the fuse.

What model of controller do you have?

Lee do you still need fuses if so IM me I have a local connection.

FICCORP.COM

alternety said:

You don't need to put the variable control back. Use it as it is with a load small enough to not stress the limit of the supply at 120V. Then measure voltage. If it is 70V that tells you that a lot of the controller is working properly.

Click to expand...

I think I understand you there, and I did get a 70-volt output reading for the 2 or 3 seconds the one fuse I had lasted.

alternety said:

If you have enough resistance wire that is probably better than a bulb if you have to have a fuse in place.

Click to expand...

I have a 26-ohm piece of heating wire that can take 120 volts without getting too hot, and that is the one I will use for testing once I have another fuse or two. Also, you had previously asked this and I now believe I had misunderstood:

alternety said:

Is the primary side of the controller fused or on a breaker?

Click to expand...

I do not know. The power comes in at terminal "1" and goes to one end of the fuse holder, then the other end of the fuse holder goes to one of the output terminals ("4"). I am a long way from being any kind of electronics technician, but it looks to me like the fuse is common to both controller and load while the controller decides how much current to let pass between input "2" and output "3". The one thing I do know here is that the fuse is as large as the overall load the controller is capable of handling.

alternety said:

What model of controller do you have?

Click to expand...

I have an 18D-1-10i, and I bought it here less than a year ago:

http://www.thermalinc.com/power/18D.htm

BrianJohn said:

Lee do you still need fuses if so IM me I have a local connection.

FICCORP.COM

Click to expand...

Thank you, maybe, and I will.

I poked around and have the feeling these may be a pretty old design. That fuse costs more than a pair of SCRs would now. The specs say that fuse acts in 2 milliseconds. That just does not feel like the way a even reasonably current design would approach things. The people you bought it from seem to have a full range of product so it is probably currently manufactured. I am just sitting here trying to think of the possibilities for faults. You of course don't want to break anything.

The fuse is in series with the load but it is only for protecting the SCRs. To test with an external load that will limit current at 120V to an amp or two (a 25 W light bulb), I would probably just bypass the fuse for testing. Unless you are unlucky enough to have the bulb decide to burn out during test, with that load the fuse is doing nothing. Two bulbs in series would make the possibility pretty remote. You should not need very much current at all a a minimum load. Probably in the milliamperes. You just need enough to latch the SCRs on during each half cycle.

There was an option for that controller to provide current limiting. If yours has that try adjusting that down and see if it does any good.

If it is not easy to get it to stop blowing fuses you might buy another controller that does not use a fuse. Baldor (and others) make similar devices. I use a Baldor BC140 for my water pressure pump. It takes either 120 or 240 in and provides variable voltage DC up to 11 amps with the optional heat sink for controlling a motor. More complicated than you need; just an example. Look on an auction site.

You might try to get some more info on just what that thermostat does. You said it is powering the controller directly and that it "pulses". It would be useful to know what the pulsing is. If it is a simple mechanical on/off based on temperature that is OK. If pulses mean a short interval pulse train to modulate the load, that could be a problem. If the thermostat is pretty smart it might be the latter.

Try it for a while with an external load. It is possible that there is arcing in your underfloor wiring when it has higher voltage applied to it. With the speed of that fuse it could blow it.

Edited to lower light bulb wattage to provide provide a better limit for inrush current on a cold bulb.

The fuse is in series with the load but it is only for protecting the SCRs. To test with an external load that will limit current at 120V to a amp or two 50 or 100W light bulb),

Click to expand...

That is one good idea, wish I had thought of it...

Try it for a while with an external load. It is possible that there is arcing in your underfloor wiring when it has higher voltage applied to it. With the speed of that fuse it could blow it.

Click to expand...

This is possible if you know any electricians, see if you can borrow a megger to test the underfloor cabling.

alternety said:

You might try to get some more info on just what that thermostat does. You said it is powering the controller directly and that it "pulses". It would be useful to know what the pulsing is. If it is a simple mechanical on/off based on temperature that is OK. If pulses mean a short interval pulse train to modulate the load, that could be a problem. If the thermostat is pretty smart it might be the latter.

Click to expand...

Here is my thermostat and some info from its PDF:
http://www.aubetech.com/products/produitsDetails.php?noProduit=30&noLangue=2

---
Temperature Control
The TH115 thermostat works differently than conventional electromechanical thermostats. It is equipped with a proportional integral adaptive (P.I.A.) controller which determines heating cycles by analyzing the temperature behavior history within the room. The P.I.A. controller reduces temperature variations providing an accurate temperature control while increasing user comfort. The controller determines the amount of power required by the heating system to maintain the setpoint temperature.
---

The tech I talked with at Payne told me that stat would not cause any problems for the SCR. Based on some kind of 15-minute cycle, it puts out a graduated 20/40/60/80/100% (at 120 volts) as necessary for maintaining a steady temperature. After initially warming the floor at 100%, the stat usually sends out either 40% or 60% for a few minutes at a time whenever needed.

BrianJohn said:

... if you know any electricians, see if you can borrow a megger to test the underfloor cabling.

Click to expand...

I think I know where I could do that.

Since I only need 6 amps to put 70 volts into my heating wire, what would either of you think about using a lower fuse that is not fast-acting, or even some kind of circuit breaker?