Atwood furnace shuts down after first failed ignition

[obscenic]

Furnace: Atwood hydroflame 7912. Probably a 1983. The board has been replaced with a dinosaur board with fan control.

I have a very bizzare problem, that I'm hoping someone out there might have some ideas about. I've spoken at length to technicians from dinosaur boards about it with no outcomes. Here is what happens.

My furnace will run fine for many days, through many cycles. Then, with no changes to anything else (usually in the middle of the night or early morning) the following happens:
1. Thermostat turns on
2. Fan turns on + sail switch engages
3. Propane valve opens and igniter strikes
4. Furnace immediately shuts down and turns fan off
5. My programmable thermostat loses all of it's programming, time, date etc.

...There are no re-try attempts (there should be 3)

6. The furnace will remain off until the thermostat is turned off altogether for at least 60 seconds. Then there is a barely audible click and I can try again to turn the thermostat on.
7. Turning the thermostat on causes the exact same thing to happen.

To workaround this, each time this happens I remove the ignitor, blow air into the combustion chamber to clear out the propane build-up, reinstall the ignitor, turn the thermostat on, and it ignites just fine.

It seems at first glance as if there's two issues at play here (1. failure to ignite and 2. immediate shutdown) but I'm not totally sure it is two separate issues. I think if the blower wheel was allowed to run (as it is supposed to after a failed ignition) and clear out the propane it would probably ignite the second time.

[obscenic]

So... the saga continues.

I replaced the furnace. I thought with the frequency with which I use it that I could afford reliability. I put in a new 7916-II from atwood, and it works much better, but:

Once in about 3 times trying to start, the t-stat STILL shuts down during the ignition attempt, causing a failed ignition (because it opens the switch which closes the gas valve etc) and starting the whole startup cycle all over again 2 seconds later. This means the furnace will sometimes go through 3, 4, or 5 startup cycles before it actually starts burning gas and producing heat.

I'm at a loss now. The wiring is brand new, the t-stat is brand new (with fresh batteries), and the furnace is brand new. Are programmable thermostats just incompatible with RV furnaces?

[obscenic]

dougrainer wrote:
From your post, it seems you have 2 controls now for the fan motor----The Dino board and the original fan relay which should have been removed when you install a upgraded Fan control module board.

Out of curiosity I bypassed the relay, moved the jumper on the board to the position for no-relay, and now the fan won't even start.

[obscenic]

Ok, I was wrong about the t-stat. Jump starting it by directly connecting the t-stat lines also causes the board to shut down the fan on a failed ignition. No purge cycle, no power to the board.

I'm out of ideas. The relay is working perfectly, the board is new enough that unless it's a dud, it's not a consideration, the t-stat has been ruled out, the ignition gap is fine...

[obscenic]

gotsmart wrote:
If everything else is good and the consensus is the t-stat, I'll throw this into the mix:

Could it be that the make and model of the t-stat you are using does not perform well in an RV environment?

In a sense, I think you're right, because it's clearly some bad discharge going down the line and momentarily breaking the t-stat connection right during ignition...

With the remote sense line cut nothing has changed.

[obscenic]

dougrainer wrote:
From your post, it seems you have 2 controls now for the fan motor----The Dino board and the original fan relay which should have been removed when you install a upgraded Fan control module board. I don't understand WHY you left the Fan relay in place when you went to the Dino board.

I *think* I may have figured it out. I *think* the spark is jumping to the remote sense terminal which isn't even used by the dino board. This is getting back to the t-stat, which is momentarily flaking out on making the connection (I can hear it happening). This shuts down power to the relay, which (once the t-stat is back on) is still letting power through to let the fan purge. But it can't because the board isn't letting power through...

I've just chopped the remote sense line, I'll let you know how it goes.

[obscenic]

Old-Biscuit wrote:

You stated "The wiring from the t-stat to the unit is brand new"

You pulled new wire?

Then you have a scape on the insulation that intermittently touches and grounds/shorts.
Blower shutting down at same time burner shuts off is caused by bad/incorrect wiring or faulty relay.

Other than that....I have no idea.

It makes sense, but I know it's not scraped, it was a straight free run with well insulated wire.

dougrainer wrote:
This is a GOOD example of what happens when someone CHANGES a LOT of parts with NON OEM parts.

Can't get fenwall DSI boards anymore, that's the only thing that's not OEM.

dougrainer wrote:

That said, the FIRST place to look is to replace the Time delay (fan relay) on the Furnace. THAT is the gateway to the fan motor. From your post, it seems you have 2 controls now for the fan motor----The Dino board and the original fan relay which should have been removed when you install a upgraded Fan control module board. I don't understand WHY you left the Fan relay in place when you went to the Dino board.

Dinosaur doesn't suggest taking out the time delay relay (http://www.dinosaurelectronics.com/Instructions/Fan50PP_inst.pdf)
Should they? It seems to me that you would still want the relay there so that the power going to the blower (3 amps) isn't running through the thermostat.

dougrainer wrote:

Also, there is no reason to have to blow out any perceived residual LP in the chamber. The purpose of a sail switch and the delay of 20 to 30 seconds before ignition is to purge the chamber. Doug

Yeah, I realize it makes no sense, it just seems that whenever I do that (I initially did it to check the spark gap), it works immediately afterwards.

HiTech wrote:
What's the 12v system voltage when this is happening? Maybe low? What is the 12v system voltage when it is working fine?

Good all around

beergardens wrote:

This is my thought too. Suppose your ignitor tip is burned down somewhat and the gap is now too large for the spark to jump across consistently.

The gap is perfectly within spec.

[gotsmart]

If everything else is good and the consensus is the t-stat, I'll throw this into the mix:

Could it be that the make and model of the t-stat you are using does not perform well in an RV environment?

Why do I ask that? I owned a S&B house in NH with FHA heat. Without a humidifier on the furnace, static electricity was a problem throughout the house in the winter. When I went to replace the old round dial t-stat with a setback t-stat I bought a Hunter (brand) t-stat for something like $39. I don't recall the model #. The tiniest static discharge from a finger touching its control surface would erase all of the programming that I did. I returned it and bought a Honeywell CT3400 setback t-stat for $99. At the time I thought $100 was too much for a t-stat. IIRC, it was advertised as resistent to static electricity. It is. I would drag my feet on the carpet, touch it, and generate huge - mildly painful - shocks and nothing hapened to the t-stat.

Where am I going with this? If you replace it with an 'old fashioned' t-stat, similar to one of these, and the furnance works fine - then research the available t-stats. Test 2 or 3 of them. Look for a ruggedized unit designed for hostile environments. It just may be that the one you bought sucks, or is not well suited for the application that you are using it for.

[beergardens]

obscenic wrote:

Old-Biscuit wrote:

Then why do you loose programming on t-stat?

At best guess I would say it's getting a shock down the line. .

This is my thought too. Suppose your ignitor tip is burned down somewhat and the gap is now too large for the spark to jump across consistently. That spark needs to go somewhere when the field collapses, and if the gap between the ignitor tip and ground isn't the easiest path, that voltage will dissipate somewhere else, possibly somewhere that the board doesn't like.

Granted, I am not familiar with your furnace model specifically, and am assuming it uses a spark for ignition. If so, I would have a good look at the ignitor and verify the gap is within spec.

[HiTech]

What's the 12v system voltage when this is happening? Maybe low? What is the 12v system voltage when it is working fine?

Jim

[dougrainer]

This is a GOOD example of what happens when someone CHANGES a LOT of parts with NON OEM parts.
It is VERY difficult to determine cause when you have replaced with non standard parts. That said, the FIRST place to look is to replace the Time delay (fan relay) on the Furnace. THAT is the gateway to the fan motor. From your post, it seems you have 2 controls now for the fan motor----The Dino board and the original fan relay which should have been removed when you install a upgraded Fan control module board. I don't understand WHY you left the Fan relay in place when you went to the Dino board. Also, there is no reason to have to blow out any perceived residual LP in the chamber. The purpose of a sail switch and the delay of 20 to 30 seconds before ignition is to purge the chamber. Doug

[Old-Biscuit]

obscenic wrote:

Old-Biscuit wrote:

Then why do you loose programming on t-stat?

At best guess I would say it's getting a shock down the line. The programmable thermostat is removable, it has pin connections to the backing plate which is where the t-stat wires connect. I can take the t-stat off it's backing plate, and leave it sitting on the counter for 3 months and it still displays the time and maintains it's settings.

You stated "The wiring from the t-stat to the unit is brand new"

You pulled new wire?

Then you have a scape on the insulation that intermittently touches and grounds/shorts.
Blower shutting down at same time burner shuts off is caused by bad/incorrect wiring or faulty relay.

Other than that....I have no idea.

[obscenic]

Old-Biscuit wrote:

Then why do you loose programming on t-stat?

At best guess I would say it's getting a shock down the line. The programmable thermostat is removable, it has pin connections to the backing plate which is where the t-stat wires connect. I can take the t-stat off it's backing plate, and leave it sitting on the counter for 3 months and it still displays the time and maintains it's settings.

[Old-Biscuit]

obscenic wrote:

Old-Biscuit wrote:

Old problem....new board didn't correct.

Yup, exactly.

Old-Biscuit wrote:

OK but is the t-stat any good

Brand new

Old-Biscuit wrote:

You are losing power at/to t-stat (loosing program/time/date etc.) so either new wiring has a scape on insulation and is grounding out, a loose wire nut, poor connection OR old relay is faulty.

T-stat doesn't get power from wires, it has batteries. There's only two wires going to t-stat; t-stat just opens or closes them to connect 12v back to the furnace.

Then why do you loose programming on t-stat?
5. My programmable thermostat loses all of it's programming, time, date etc.

[obscenic]

Old-Biscuit wrote:

Old problem....new board didn't correct.

Yup, exactly.

Old-Biscuit wrote:

OK but is the t-stat any good

Brand new

Old-Biscuit wrote:

You are losing power at/to t-stat (loosing program/time/date etc.) so either new wiring has a scape on insulation and is grounding out, a loose wire nut, poor connection OR old relay is faulty.

T-stat doesn't get power from wires, it has batteries. There's only two wires going to t-stat; t-stat just opens or closes them to connect 12v back to the furnace.