Electric Floor Heat

This project is beginning to make me feel like I was trying to sort out somebody's 12-volt conversion on a tractor. Bus Driver, your comments have really helped me out. I had forgotten the 125% rule for continuous loads. At first look, a 20 amp circuit seemed right and I never questioned it. The entire system is sized for 20 amps, with the single exception of the thermostat at 15 amps. Specifically, 20 amp breaker, 12 gauge wire etc. The 20 amp contactor is actually rated for 25 amp resistive load, but that is irrelevant. I don't know about the capacity of the power company's components for the curtailment system. But that makes no practical difference because it is a 12 gauge wire from it anyway.

Around noon today, I connected the control line (#3) directly to what had been connected to the contactor output (#4). This removed the contactor from the circuit. I then went to the box where #4 connects to the heating elements and disconnected two elements, leaving the one large element where the thermostat sensor is. This "simplified" configuration now matches the manufacturer's recommendations and puts the connected element in a circuit that is covered by the GFI. It has been running that one element for 9 or 10 hours now, no problems. Tomorrow I can hook in each of the other elements (one at a time) to see if it still behaves. At this point, I think it will.

The problem now is to decide on a permanent fix. While I was reconfiguring the elements, I took a resistance reading on each one individually, which is the only way I have to determine what is actually buried under the tile. The 2 large measured at 30 ohms, which matches the reading for the manufacturers longest heating wire (which they rate at 7 amps). The smaller measured 64 ohms, between 2 lengths (50 to 63 ohms and 4 amps) (67-83 ohms and 3 amps). The current draw calculation will depend on just what voltage you pick. My earlier quoted current draw came up a little higher than their numbers. Whatever the exact load, running all 3 elements properly will require either adding a second complete circuit for part of the load or upgrading the existing circuit. Upgrading the existing circuit essentially requires a complete replacement. Either option would require pulling the power company in to some extent or another.

The other option is to reduce the load down to what the existing circuit can support. The two larger elements together heat a roughly rectangular area in a room. The smaller heats a hallway. If I leave the smaller one disconnected, the two larger will draw 14 to 16 amps. At 125%, that is in bounds for the existing circuit. The manufacturer says the 15 amp thermostat can handle the two larger elements without added relays etc. The boss has OKed killing the heat in the hallway. In fact, as I spelled out the situation, she suggested it before I got that far.

Unless I discover something else (like a surprise as I test with both large elements connected), I am going to reconfigure it to only use two elements. This requires:
1) Remove the contactor and contactor box.
2) Place an electrical box where the contactor was, connect the original input (cable 1) to the cable feeding the thermostat (cable 2). Run a new cable from the output of the thermostat to the box where cable 4 is connected to the elements. Connect elements 1 and 2 to the new cable.
3) Cables 3 and 4 and element 3 will all be abandoned in place ( and tagged for future reference).

With these changes, it will be wired according to the manufacturers instructions and the elements that are in use will be protected by the GFI. (Yes, I understand the protection is actually for the people standing on/near the heated floor.) In some ways, I'd rather have a GFI in the breaker box. But with this configuration I should be up to code.

The only question I see right now is whether the thermostat/GFI will continue to function properly. Since it will be installed per the manufacturer, if it doesn't, it needs replacing anyway.
 
And all these years I thought figuring out amperage, circuits and timers for all these Christmas lights was an adventure! Wow, you guys have me confused.
 
Jim Becker":1icclj9s said:
........The problem now is to decide on a permanent fix. .........
Click to expand...
I'm thinking wood stove at this point.

Jim Becker":1icclj9s said:
.......Unless I discover something else (like a surprise as I test with both large elements connected), I am going to reconfigure it to only use two elements. This requires:
1) Remove the contactor and contactor box.
2) Place an electrical box where the contactor was, connect the original input (cable 1) to the cable feeding the thermostat (cable 2). Run a new cable from the output of the thermostat to the box where cable 4 is connected to the elements. Connect elements 1 and 2 to the new cable.
3) Cables 3 and 4 and element 3 will all be abandoned in place ( and tagged for future reference)......
Click to expand...
Are you SURE you didn't accidentally copy this from some instructions for a toy you assembled Christmas Eve?

The only thing this thread is lacking is pictures. I'm thinking a giant bottle of aspirin would be appropriate.
 
There is another possibility to operate all three on a 20 ampere circuit. The hallway would have it's own thermostat as an addition. That thermostat would receive it's power supply from a replacement contactor. The new contactor would be double-throw in that it has one set of points normally closed (NC)and a set normally open (NO). The hallway supply would be through the NC contacts and could operate as that (new) thermostat operates UNLESS the thermostat for the larger area closes the NO contacts to supply power to the larger area heating elements. Thus they are sequenced and the load at any moment is less than 16 amperes.
The elements could thus be said to be interlocked.
It would require a GFCI breaker in the panel.
 
Jim Becker":2phiijrl said:
I plowed my drive today. When I got done, the Cub still had half a tank of amps.
Click to expand...
Jim must be the guy behind that electric Cub we read about.
 
Bus Driver":ul1hmt7j said:
There is another possibility to operate all three . . .
Click to expand...
I had to mull that over a bit. Essentially the current thermostat and 2 large elements get priority and the second thermostat only gets power when the large elements are turned off. I can think of a couple ways to do that. Some would require changing the breaker to a GFI to avoid replicating the non-protected situation I started with.

I'm going to stick with my earlier plan to leave element 3 disconnected. I will keep this option in mind in case we decide to start using element 3 later on.
 
I switched to the two large elements and ran it quite a while last night. After running it several hours today, I measured the two together at 15.6 ohms. Who knows if the last digit means anything on my meter, but it is about 1/2 ohm higher than they were cold. The temperature was probably up 10 to 15 degrees from cold. Yes temperature makes a difference, but heating wires buried in mortar won't have a temperature swing like a light bulb. I did all my initial observations and calculations based on cold elements and a full 240 volts on the lime. That way I was generally looking at worst case.

It has been running with the two larger elements and no contactor for long enough now that I think it is OK with that configuration. I am going to permanently wire it up that way. I just need to run a new wire from the thermostat to the junction with the elements (couple feet away). Then change out the contactor enclosure for a box with a splice to run power to the thermostat.
 
Those elements typically will not exceed 110 degrees and the flooring material is wicking away heat from the elements all the time. The empirical evidence indicates that the contactor coil has some slight leakage (fault) to the non-current carrying metal parts of the contactor. No doubt the contactor was mounted in a metal enclosure that was grounded, as it is required to be. The slight current to ground actuated the GFCI.
Have you noticed how sharply the horizontal heat migration stops at the edge of the heating elements? Two inches away the tile will be cold.
 
The metal box for the contactor was grounded through the ground wire in the power cable. One of my early tests was to remove the ground from the box and put a milliamp meter between the box and the ground wire. The meter never showed anything (at least while I watched). With the ground disconnected, the GFI went longer without tripping but I don't know if that was meaningful or just a random variation. Leakage between the control coil and one of the main poles could trip the GFI without any leakage to ground too.

Based on the ohm readings (implied element length) and the area covered, it appears that the cable was set at a pretty close spacing. I haven't been able to use heat to trace the actual path (might be able to with am IR thermometer). But it is easy to tell the difference near the edges where they didn't run the cable.
 
You must log in or register to reply here.
Back
Top