Refer: Operation issue using LP…

When I bought this RV, I was a bit worried about the refer being a Norcold, but research on the net showed my model, a N842IM, was made after the thousands of defectively designed refers were sold and installed in 10’s of thousands of RVs. That maybe with this model Norcold had gone back to an older, but more reliable design. It was the right age for that to have happened. But I didn’t know for sure, as Norcold has been less than helpful or forthcoming with info.

While I was busy spending time investigating that, I failed to test the refer on LP while it was on the consignment dealers lot.

And sure enough, a few weeks later, I found a problem with it.

The refer appeared as though it hadn’t been used much at all. Like the OP was afraid of it and left it off the whole time they owned the RV. This is plausible because the Norcold brand of refers were causing fires and deaths, and that fact was spread around by RVers and on forums. If you’ve ever lived in apartments, you know it’s nearly impossible to get the inside looking clean after someone moves, but this one was sparkling, looking brand new. I checked the coils on the back and they looked pretty good too. So I wasn’t all that worried about it.

After I drove the RV off the lot, and spent a month parked and plugged into shore power, the refer always cooled. A few weeks on, after leaving the Mesa area where I bought the RV, I only took short trips so I didn’t really notice that there was no LP cooling while I drove from place to place. My food never really got warmed up or spoiled during those short trips.

Eventually, one time before I plugged into shore power, I did see that there was an error message on the two character refer display that said ‘no FL’ (no flame). OK, I thought back then, I’ll check that out next time I’m parked for any length of time.

Well, a year passed without it ever being a big issue since I seldom boondock, but today in March ’17 seemed like a good time do something about it. I’m planning on a multi-day drive up north and would like my food to be OK when I arrive. I’ve tested enough since I first noticed the problem to know that it would always fault when on gas. It never had the symptom of ‘work one time and not the next’, it would just never work for very long on LP. This actually makes it easier to track down a problem.

Opened the outside cover to take a peek around…set the refer on LP, run back outside, and find there’s no flame. In the couple minutes before the fault display occurs, I was hearing the propane valve operate, and saw the electronic spark fire several times. But the flame looked terrible. Hmm. Removed the heater tray cover and in the little metal tray assembly under the chimney and heater assembly, I could see a large accumulation of soot. It almost filled the tray. The kind of soot that comes from too little air. Odd. Not seeing much rust, which is a good thing…rust is a typical problem with refers that sit over the winter and many owners need to clean it off the burner regularly.

The soot needed to be removed so I grabbed my trusty mini shop vac and blew all the soot out of that tray along with a few leaves and some dust and dirt from the rest of the compartment. It’s the sticky kind of carbon black soot and wouldn’t be good to try to vacuum it into the vacuum’s filter bag.

What I thought was that the refer had been turned on when the gas was turned off or the tank was empty, and as it ran out of the fuel remaining in the pipes, that poor mixture of gas to air caused the soot. Than after I bought the RV and on the trips I took I always left the refer on, and with the gas back on, it would try lighting, perhaps even run for a while, but all that soot blocked the flame, and created more soot.

Well, whatever the cause, after a few seconds of cleaning, the soot is all gone. Here’s a look at the burner assembly (the burner orifice is behind that black tube) and spark assembly (device with the red wire going to it) and the tray where the soot had previously accumulated. Very difficult to work on things with this design. That black pipe containing refrigerant is in the way, and there’s really no easy way to take out the orifice or spark/probe (serves dual function…arcs to light the gas, than the heat from the flame tells the circuit that the flame exists) without removing the entire refer from the RV. Over on the other side of the compartment is the propane valve, that small device in line with the brass pipe. On top of it is an electrical solenoid that shuts off the gas flow, along with a manual shutoff incorporated in the same housing. Wires from that solenoid connect to the black box on the wall, which houses the electronics.

Here’s another shot from a slightly different angle that shows the black control circuit box on the wall. [That light blue device in the middle of the picture is the water valve for the ice maker. It’s interesting to note that the pipe going up to the ice maker has a heater wire. They are wrapped together in that dark blue foil wrap. Must have been an expensive option. I was working in there once and failed to rewrap the white heater wire around the valve itself. If I ever need it, I’ll get it back in position. That’s the white heater wire dangling down in the lower section of the photo]. And a look up inside that compartment towards the roof top vent with a view of the refer coils. There isn’t much rust on the coils compared to other refers I’ve worked on…which is a good thing.  Here’s a close up of the gas burner orifice assembly…everything to the right of that brass compression fitting. It doesn’t look all that bad in there, now that the soot is gone. The spark/probe device is just above the orifice and burner housing. I blew this area out with the vacuum and brushed off the area around the spark rod to remove any stubborn soot. It took a while (seemed as though it took some time for the gas to work it’s way up to the orifice), and several refer restarts, but eventually, it flamed. Notice how nice and blue the flame is? If it were mostly or all yellow, that would be a symptom of a problem, but it’s perfect. Hmm. Maybe too much soot had gotten into the air supply causing the failure? It’s all cleaned out now. The funny thing is that soot wouldn’t be created by such a lovely gas flame, and this design uses a fixed setup for introducing air to the flame…no adjustment possible so it can’t ever go out of position and close somehow either. So I’m at a loss as to how the soot was produced.This is what the flame should closely resemble (figure B) and happily, mine does. My propane tank gauge registers around 1/3rd tank, and air temperature is in the 70F range.

I let it run for several minutes, but the wind was gusting, so eventually, I put the cover back on. Plus it looked so good in there, I didn’t really think that there was anything wrong with that part of the system.

So I came inside and carefully watched the display on the refer, and 1 hour and 10 minutes later, it gives the ‘no FL’ error code. DOH! Well, I know that the gas orifice is as clean as a good strong air stream could make it when I blew it out, the blue flame told me that it’s getting gas and is set up right with the gas/air mixture, and that the orifice is in good condition, I saw numerous sparks that looked healthy, so that leaves the control board, or the gas valve as a potential culprit. I think what’s happening is that occasionally, it’s calling for cooling, and tries to restart the flame, but either the spark is missing, or the gas valve doesn’t operate and it throws the error code. I’ve never smelled gas in that area like the valve had been turned on with a missing spark, so I think that the valve or circuit that is suppose to turn it on is bad. Or a connection, or the valve is sticking because it’s been used so seldom over the last 12 years.

Running another test and prior to starting the refer again, I wiggled and tugged the wires going to the valve and on the circuit board so if there’s a poor connection there, it might be OK now. Time will tell. And after 1 & 1/2 hours, it faulted again. So this time I tapped the solenoid a few times with the plastic end of a screw driver, removed and replaced all the terminal connectors between the solenoid and the control board…and turned it back on. It once again lit easily.

It’s now the next day, and it’s run without faulting for 17 hours so whatever the problem was seems to have fixed itself for the moment. Perhaps that’s evidence that the solenoid was sticking and repeated operation has loosened it up a bit. I’m in Mexico and can’t easily get any parts down here, so I’ll just limp along with it for now.

On Edit: It only operated 22 hours and then faulted again with the same ‘no FL’ error so I gave up for now and switched over to AC.

The next day, I went and checked outside again. And…the soot is back. This time big chunks. Last time in this spot was a pile of small lightweight flakes.

The chunks were so big, I went ahead and vacuumed them out this time. Then spent 5 minutes again blowing out the rest of the cabinet paying extra time and attention to the chimney. Lots of little chunks came out…again. Banged on the chimney several more times trying to dislodge any more chunks. Some more stuff did fall out. Still not seeing flakes of rust so that’s good. I’m not recognizing the mechanism that might be causing new soot to form, or am I looking at old soot? I think the latter, but I’m running another test to try to find out. Also measured the solenoid and the attached wires this time…86 ohms. One ohm over the recommended max. Not too bad though. Cleaned the solenoid terminals and tightened the wire terminals too.

After all the cleanup, I switched it back on and watched and listened as the system went through it’s procedure, which all looked and sounded good. Now, after 27 hours of fault free operation, I believe that the problem is solved.

I’m going to guess the problem was caused by the propane burner system being run in a way that caused excess soot to be created somehow and then blown up into the chimney. Lack of air is the most common cause of soot in a propane system, and that turns the flame mostly yellow, and lowers its temperature. That lower temp causes the system to shut down, but not before it’s already created some soot and the rising hot air has carried it up into the chimney.

After I bought the rig and drove around the country with the refer turned on, because it hadn’t been properly cleaned out, it would create more soot somehow, or the residual soot from the original event would drop down on the burner causing the next spark event to misfire, prolonging the problem. Now that I’ve seemingly gotten it all cleaned out, it hopefully won’t be a problem again. But if more soot drops down during my next trip, I’ll go ahead and make a small chimney brush and go at it from the top of the RV. I probably could have hurried this fix up a bit if I’d just trusted the other RV’ers who posted that they needed to clean their chimney occasionally and followed their advice to begin with. Well, live and learn.

Had to cut this latest test short because I’ve gotten to below 1/3rd tank of propane and I want to wait until I’m back in the US to fill up. So I’ll be back to edit this post if anything new happens.

On Edit: March 28th, 2017 – Travel day from San Felipe, Mexico to Gila Bend, Arizona

Apparently those last large chunks of accumulated soot shown in the above picture that dropped out of the chimney were the last of them. When I left Mexico early AM and drove to Gila Bend, Arizona, I had the opportunity to check the refer many times along the route since it was necessary to stop multiple times. And every time I checked, the refer was working fine on gas. Now that I’m settled for a few days, just checked the tray out in the refer compartment and it’s free of soot. [Apparently, not a long enough test. See below].

On Edit: June 2017 – Final component checks & adjustments

After heading north from Gila Bend up to Oregon, I did check the flame several times along the travel route and found that yes, the gas flame was still dropping out occasionally. Not nearly as often as before, but still dropping out. And big chunks of carbon would still fall in the tray, though not as many and over the miles that decreased to zero.

So in June ’17 while at a RV park, I removed and inspected the spark probe because I’d read that it’s a maintenance item and needs cleaning occasionally. Using the instructions in the Users Manual, and instructions I found on the RV forums I frequent, I felt that the ‘Electrode-Spark/Sense’ (ESS) metal probe looked fine, but I cleaned it anyway. Also adjusted it’s gap from the burner orifice per specs. Finally, tightened the terminal attached to the ESS wire where it attaches to the control board. Then over the next few trips watched it carefully, and the refer stayed on gas, never gave the ‘no FL’ code.

And that seems to have done the trick. Fixed.

On Edit: Oct. ’18

Have verified that the refer continues to operate on gas whenever I travel since June ’17. It hasn’t ever failed again. And I’ve never spotted any more soot in the tray so that soot from some long ago event seems to have worked itself out of the chimney.

I now consider the intermittent operation on gas issue to be resolved.

Back in Feb. ’16 when I bought this rig, I thought that the refer looked new. Like it hadn’t been used at all by the previous owner. And perhaps it hadn’t. Whatever caused that ‘soot event’ in the past may have scared them into thinking this Norcold was one of those that might burst into flame as it must have smoked a lot, so they just left the refer off and never used it. Perhaps it happened within a few days of driving it off the dealers lot and they were already in Arizona (they bought it in Michigan and I bought it in Mesa, AZ). Finding that the ESS just needed cleaning (that’s all I think was wrong, but I still haven’t a clue what caused the soot long ago) saved a considerable amount of money and hassle. I’m pleased with the operation of the refer now, especially since I added an inside fan, and expect it to last at least the 10 years I intend to own it. As long as I remember to do the regular maintenance as required.