Flame Sensor/ignitor Takagi

[adobejoe]

I have a Takagi JR on demand water heater/boiler about 10 years old. Have been reasonably satisfied but I have been getting error code 12 recently as  it tries to sustain a burn cycling twice. Not sure if this is appropriate here but Takagi tech support is stumped. And I am quite frustrated. I have good gas pressure (LP), and have cleaned both air intake and flue stack. The flame spark-rod assembly connects on two separate wires, one for creating spark, one to give feedback.   The controls start up and flame ignites and goes maybe a second or two goes out and relights and then goes out and the display shows error code 21. Yes, I have cleaned the rod? What to try....or do I buy a Rinnia?

[OperaHouse]

I've had that happen twice on my natural gas furnace.  I've taken a paint brush and cleaned the flame detector probe and insulator and it has always worked for a half dozen years after that.  Might be leakage from dust on the board or wiring.  That sensor has a very high input impedance and would look for something shunting it.

[adobejoe]

Cleaned it with what? They suggest 100 grit sandpaper, but I also seen using a non-abrasive dish pad, or cleanser like simply green. I am considering buying some other brand of flames sensor rod and trying it out, rather then plunk down $2k or more for another model. My plumbing lines for the radiant heat are all configured for the takagi so It will likely not be the same with some other model like Weil McClain, Rinnai, or Bosch. I have been poorly opinionated about "modern" appliances. My wife was using our Jen Aire range at 400 DEG F the other day and the inner door window just shattered. We have a GE Arctic Air frig and it freezes vegetable tray and the water dispenser like freezes. The stuff just seems to self destruct at 10 yrs +/-

[TomT]

If it has a blower and diaphragm switch to make sure the exhaust blower is running.
Then try cleaning the tube port that goes to the blower housing.
I had a furnace that did the same thing and that was all that was wrong.
It had built up a layer of dirt and covered the port.

[XeonPony]

it is known as a flame rectifier, clean it with isopropyl alcohol at 99.8% purity, then polish with a firm cloth. That should solve the issue, if not then use a fine sand paper and polish both the rod and the section of the burner near it. All so check all your groundings, they are very sensitive to bad grounds.

if the unit is forced draft then as TomT stated check the fan proofing circuite.

[OperaHouse]

This likely has multiple burners with the sensor furthest from the ignitor.  Check to see if the burners are even.  An orifice may have a slight blockage giving an uneven flame.  Sensor shouldn't need excessive cleaning.  Alcohol has a lot of water and sanding can cause water content to be maintained for a period of time.

[oztules]

Single wire sensors tend to use the resistance of the flames plasma to get a signal to ground.... this means if there is any unusual flame shape that would allow the flame to not touch the sensor, then it will not see a path to ground... ie it uses the flame as a piece of high impedance wire.

I have seen an entire burner run perfectly,,,, but for the row under the sensor... it has a tiny corrosion growth around that one  burner jet (out of 30 or so)... just that one row stopped the thing in it's tracks. Any other row going out would not have affected it at all.


................oztules

[bart]

   All I can add is to look at the ceramic that surrounds the igniter / sensor. Must not be cracked, if it is, needs replaced.

[adobejoe]

Thanks for all these suggestions. Certainly worth a shot and each not too difficult to address. Will see how they work and report back in a few days.

[OperaHouse]

I am beginning to think the flame sensor is a red herring.  Flame detectors tend to be erratic weeks before they actually stop working.  It looked like that error code was a grab bag of problems burner related.  It could be a vacuum sensor that insures the draft motor is running.  All this technology is great till it stops working.

[TomT]

If the flame sensor uses tempered glass.
It can get dirty and cause problems.
I had that problem on a stove and dryer a few times.
Clean the glass and it worked fine after that.

Also the draft motor vac/switch problem was on a furnace.
1. blower motor on.
2. Igniter heated up.
Then there was a click and the gas turned on and started a flame. Then another click and it went out.
Then it just started to cycle and do the same thing over and over.

[adobejoe]

Thanks again for input. Update. Cleaned flame sensor rod with alcohol, same symptoms. The takagi is 10 + years old and in our laundry room. Opened up and removed burner assembly. Lots of dryer lint and dust evident. Also, quite a bit of dust on small fan. Flushing out with Simple Green cleaner Cleaning and will blow out and let dry. Also will clean gas orifice jets, about 15 of them. Then I will reassemble and see what happens.

[southpaw]

The diaphragm switch that is detecting the vent fan operation is very sensitive and if the burners don't light quickly and smoothly the switch may open briefly from the "whoosh" of the burners lighting causing the control to reset and starting the cycle over again and again.
 Make sure the hoses are clean as they sometimes get condensation in them causing the switch to sense a lower pressure differential.

[adobejoe]

Good news!! Totally removed the burner assembly and cleaned with simple Green. flushed with water, then compressed air and let dry. Remove and cleaned 15 brass orifice jets with SG and pin gauge. Removed fan assembly and totally cleaned. I found dryer lint and dust in burner assembly especially. Cleaned flame rod sensor with isopropyl alcohol and put it back together...works fine now! I will try again tonight. I have this system in my radiant heat system with four zones. I also plumbed in a standard electric hot water heater in series. This gives me the option of using electricity or propane to heat my floors. With a 10 KW grid connect wind turbine, and 3.3 KW PV we have an excess generation this time of year of 3,000+ KWH. Because of the way the utility companies in Wyoming interpret and apply the net metering statute, they will not pay but half the retail price. And, yes rates are going up in Wyoming, about 12 c/KWH where I live. Some areas higher.

Thanks for your ideas and input, think I have this licked!

[OperaHouse]

Couple days without hot water. I was thinking your wife is more tolerant than mine.  Good to have a backup.  I have a add on heat pump water heater that feeds a standard electric water heater.  That allows me either or both.  Electric was a replacement for gas water heater.  I did a quick emergency install and ran it on 120V for quite a while.   At 120V, a 240V element will be 1/4 power.  Even 36VDC will heat water. That made it a 700W water heater.  Earlier this year I hooked up a single 100W solar panel to it with PWM to keep it at power point.  It was only an experiment, I have a camp system that diverts unused solar PV to water heating.  It is easy to get 3KWH of heating a day.  You could easily divert 800WH to your heater and have a decent preheat to 80F.

[SparWeb]

Operahouse...I was thinking your wife is more tolerant than mine.  Good to have a backup.

Great idea, OH.  Next time the hot water breaks down in my house, instead of putting up with my wife's nagging, I'll go live with my girlfriend. 

Joe,
Glad to hear you got it worked out.
The compulsive trouble-shooters here can't resist a good tech problem.

[OperaHouse]

These days when a woman invites me into her bedroom it is to look at something that is broken.
I habe enough things in my life that need to be fixed without adding to it.