A gas detector is a good idea, and the page
liked below has the details of our original plan (and
implementation.) But from early-on there were issues with
the LP Gas detector.
Problem Brewing
Initially it worked as designed. It was
transparent and didn't seem to be there. An important
point here is that this detector controlled a valve to cut off
the gas supply if it detected a gas leak. After a few
months of regular operation, the detector started generating
"false" alarms. I replaced the unit and all was well...
for a few/several months.
This continued. We went through a laundry list of
things that might cause alarms, but my ultimate conclusion is
that the unit could not last in the very humid environment of
the grooming vehicle. The manufacturer was grudgingly
helpful, repairing/replacing defective units (after we had
purchased about six.) That cooperation ended around the
end of 2009, when the company apparently disappeared.
Do Something
OK, so we gave up on the detector, but were left with the
problem of the cutoff valve. As a "safety" feature, the
valve is normally-closed, meaning that it cuts off gas flow
unless power is applied. Removing the detector meant
shutting off the gas. (By the way, removing the valve from
the gas pipeline is a huge project.) You would think, I did,
that I could just apply 12 volts to the gas valve to hold it
open. But NO, the valve operates on 9 VDC. The
detector provided the voltage conversion. I had
dealt with this issue, earlier in the saga of working through
failed detectors, by using a small 110 VAC transformer-type
power supply, to provide 9 volts DC, by plugging it into an
outlet powered by the inverter. No big deal.
Plan B
However... That won't do for a long-term solution
because of the interaction with the inverter's shutdown process.
The valve requires very little power to hold it open. The
inverter monitors its load and shuts off when there is no load.
The power drawn by the power supply/valve is below the level
that keeps the inverter on, so it shuts off. Valve closes,
CLICK. The amount of power to reactivate the relay is
enough to cause the inverter to turn on again, CLICK. This
cycle continues through the entire period the truck systems are
in idle state. Overnight, etc...
The Solution
At least at this moment, the solution is a voltage converter
configured as a "canary" to detect fire. Some Internet
research came up with Murata Power Solutions NMG Series device,
NMG1209SC. This component accepts 12 VDC input and
produces 9 VDC output. Here is
a link to the data sheet for the NMG series. Connection is
straight forward, two wires in for 12 VDC and two wires out, to
the valve, with 9 VDC. Normally I would enclose the
converter in a protective enclosure, use heavier than required
wires to connect it and protect the wires along their routes.
In this case. I mounted the converter high in the equipment
compartment, used light gauge wiring and only protected them
from accidental physical contact, leaving them exposed.
This is my pseudo fire detector. If there were a fire
these components would fail, and the gas would cut off.
Not the goal of the original gas detector, but until something
better comes along... By the way, the same company
made the carbon monoxide detector used in the grooming area of
the vehicle. That was less of an issue since it was only
an audible alarm.
|
