Re: Bad design?

Ross Herbert wrote:

On Wed, 19 Dec 2007 18:56:32 -0800, "Paul Hovnanian P.E."
<paul@xxxxxxxxxxxxx> wrote:

I recently extracted a floor heating thermostat that had probably
been
mis-installed and (as far as we can tell) never worked. The story
about
the electrician who did the job and left before testing the system is
another sad and long one for some other thread.

The unit is (as far as I can tell) a triac that controls 120 VAC to
an
electric mat installed under a bathroom floor. It incorporates a
ground
fault sensing circuit so as not to electrocute one stepping out of
the
shower. Because of this, the unit has two line input wires (black and
white) and two output wires to the mat (also black and white). See
crummy ASCII art below. We suspect that the electrician initially
mixed
up the line and load connections (due to the way he left quickly
without
testing the system). I'm trying to figure out how one could damage a
triac by swapping the Line/Load wiring. Due to the crappy color
coding,
this seems to be the most probable wiring error.

System (properly wired) as I envision the
thermostat innards:

triac
black |\| black
----o-------+-----|/|---MMM-o---------+
| /|/| | |
| / |GF CT |
| | | |
Line +------+ | Load
|brains|------+ |
+------+ | |
| | |
----o----------+--------WWW-o---------+
white white


Most of the likely errors I can envision would simply prevent the
unit
from operating. Am I
missing something?

Your diagram appears to show the GF circuit breaker AFTER the
thermostat/controller.

Yes. This unit incorporates its own internal ground fault sensing for a
fault in the mat. Based upon the observation that the unit has a GF
fault LED that illuminates following a fault or test (test button not
shown here), I figured it just shuts off the triac and the brains remain
powered.

There's a mechanical on/off switch (also not shown) that is (probably)
upstream of the triac in the hot leg to comply with UL requirements for
a disconnect. This switch is not automatically opened following a GF,
like a GFI receptacle reset is. It must be cycled manually.

My main concern is, given the color coding scheme shown above, could an
electrician swap the two white wires or two black wires and cause a
triac to fail? I don't know what is inside the 'brains'. I didn't open
this thing up due to the fact that 1) it is expensive 2) in isn't mine
and 3) we traded it in for a new one. All I know is that the electrician
left the job site quickly, without informing anyone that the unit didn't
work. Its possible that something else failed internally (maybe a
fusible link).

The one error that I can envision is to connect source hot (black) to
t'stst line black and source neutral (white) to t'stat load black. But I
have yet to meet an electrician dim enough to connect a black to a white
wire.

Since he didn't buy it (my friend did), he could have very easily said,
'Hey buddy, this mat you bought is junk.' We only discovered the problem
later and, prior to isolating the problem to the t'stat, the fault could
very well have been in the parts now embedded under an expensive tile
floor. Fortunately, that was not the case and his failure to alert the
customer is a subject for another thread on professional ethics.

--
Paul Hovnanian paul@xxxxxxxxxxxxx
-----------------------------------------------------------------------
Procrastinators: The leaders for tomorrow.
.