My furnace's unreliable-start problem turned out to be the electrode

Old Crow

Jan 10, 2015
South Central PA
The Atwood 8516-IV furnace in my 01 Hawk has been failing to start combustion on the great majority of attempts. After each set of three attempts, it was going into the lockout condition (indicated by the LED code of 3 flashes, a 3 second pause, 3 more flashes, etc). I could do another set of attempts by setting the thermostat to minimum and waiting 60 seconds. After a few of these 3-attempt cycles, I'd finally hear it start combustion but then it would only run for six to seven seconds. And when it stopped, it did so abruptly and I thought I could hear the click of the gas-valve closing.

I called Atwood and talked briefly with a tech support guy. He suggested the most likely culprits were the circuit board (also known as the DSI or Direct Spark Ignition module) and the electrode/flame sensor assembly. He also told me many RV shops will test the board for free. And he noted the problem could also be caused by a bad connections, bad ground, low gas pressure (and others) but those two were perhaps more common.

I called up my local RV shop and they did indeed have the tester and would be happy to test my board for free.

When I took the board in for the test, the counter guy just pulled out the tester and ran the test in front of me there at the counter. He used one of the testers from Dinosaur Electronics ( ). When it passed, he said “Congratulations, you just saved $130.” Since the next thing on my list was the electrode/flame sensor and it was only $10, I bought one there.

When I pulled the old electrode assembly, I checked for the standard 1/8” gap and noticed the main electrode wire was loose in the insulator. The gap would vary from almost-none to nearly 1/4” just by touching the side of the terminal connector. The new electrode, on the other hand, didn’t move and had the standard 1/8” gap.

The new one is also different in that instead of a (male) slip-on terminal on the connection end of the electrode wire, it has a wire coming out of the insulator with a female slip-on connector on the end of the wire. The package also has what I thought at first was a jumper wire but it’s actually a replacement for the wire from the electrode to the coil on the circuit board. The older style part # 37057 was superseded by the newer-style kit (part # 34570).

With the new electrode in place I turned up the thermostat and this time combustion started on first attempt. And it has been starting on first try ever since.

Also-- Several months ago I had ordered a Frost Sentry non-programmable garage thermostat to replace my mechanical one, largely because of this thread: so I also installed that this week.

The install was very easy and the furnace appears to work properly with it but I do have one minor unresolved issue... the Fan Only button does not turn the fan on. The troubleshooting guide says that could be the gas/electric jumper and I realized I had not changed it from the default of ‘electric’ to ‘gas’. But changing it made no difference.

Also- The exploded-parts diagram for my 8516-IV says the board is part # 36716. The board on mine, though, is part # 37875 (the part number is on a sticker on the coil). I was considering ordering a Dinosaur Electronics aftermarket board to replace mine (before I had mine tested) and if I had gone by the diagram I would have ordered a UIB-S to replace part # 36716 when I really needed the Fan 50 Plus Pins model to replace part #37875. (note: it’s easy to miss the cross-reference chart-- it’s on the Ignitor Boards page and is marked “*** Ignitor Boards Usage Chart Link ***” ).

Also- The Fan 50 Plus Pins board has additional troubleshooting indications. My Atwood board doesn’t have a green indication at all but if the Fan 50 Plus Pins board doesn’t go green, that indicates a high-limit or sail-switch problem. See for more detail.

Also- to remove the board for testing or contact cleaning-- Carefully remove the cable connector and disconnect the wire between the coil on the board and the electrode behind the gas valve. Remove the wing-nut on one side of the board and the screw on the other side. Slide board out in its carrier. Remove board from its carrier by removing the screws in opposite corners.

Also - At first glance the electrode appears to be impossible to remove past the gas valve. It will indeed come out past the valve (and without removing the wires to the valve). It is a bit tricky to get the hold-down screws on the electrode back in place after you put the new one in.


PS- I have to write this stuff down cause I won’t remember what I did! Sorry for the wall of text.
Thanks for the detailed post! Figured I'd add my similar experience, as I've actually been in the midst of many of these details for the last few days.

Looks like I had the same ordeal initially with my circa 2010 Atwood 8012-ii in my FWC Finch. After FWC service stated they didn't think it would be the circuit board b/c I was getting flashing with the lockout code, they directed me to Atwood tech service. Called Atwood for diagnostic steps, who really directed me more toward checking voltage on circuit board connections - particularly at the valve connection terminal. Not sure if I actually shorted the board or something with my multimeter, but the board stopped starting the fan or showing any voltage at any connection to the board, so all my diagnostics culminated in ordering the Dinosaur UIB S board ($84) in error from Amazon with overnight shipping to arrive yesterday. Identified the wrong valve lead/electrode connection when opening the package, but also noticed that the raised grommets in the corners of the board (which appear to be the same on Fan 50 Plus Pins model) preclude the use of existing screws to connect the replacement boards to the carrier. A quick call to Dinosaur service directed me to a really nice dealer, Roy at R&G for the right circuit board. Increased expense of Fan 50 Plus Pins board and inability to return had me hesitate on purchase, so he directed me back to Chris at Dinosaur tech service, who I missed and heard from the main customer service rep that he was really good to troubleshoot with before ordering. Will try him on Mon. morning. Will focus on faulty electrode possibility from Old Crow's experience before taking the plunge, as I think I could actually return my Amazon-purchased board if it turns out to be the electrode. Similar experience with complex disassembly of electrode component when I had examined it before (didn't measure the gaps, though, so maybe back into it later today). Ended up dropping one of the screws in the furnace fan box on attempted re-assembly and was glad I had a magnet to retrieve it (watch for that).

Anyways - thought I'd add some more detail to the post in case someone else is looking for thoughts later. I know Old Crow's post helped me!
OK, more two cents. I have put several posts on the site due to numerous Atwood issues. I also have the 8516-IV in my Eagle. So far I have replaced the circuit board, electrode, and the fan motor. My furnace STILL has intermittent issues. Yesterday I noticed that those automotive-type wire connectors (especially ones that connect 3 wires) are not reliable. If you wiggle them you can get various parts to stop working. I have had so many issues with this darn thing that yesterday I decided I must have a poor ground, or connection and in desperation began soldering the connections together (nothing to loose at this point). If there is any one thing about these furnaces that I know for sure is that if you have an "intermittent" problem, it could be anything and that you may think you have fixed something, but it won't be long before issues re-appear.

Good luck guys. I hope your luck is better than mine. If I didn't think Atwood products were so poorly designed (both for serviceability and reliability) I would just buy a new one. But that is probably throwing good money after bad.

Specjoe- I don't quite get the logic of the board being okay if it flashes the lockout code. To me, lockout just means "I didn't get the right signals so I closed the gas valve". Then again, I could be mis-understanding the point. Also- I took a quick look for "Roy at R&G" and that appears to be the guy at and I see his email address is on there. Thanks for that resource. Boards are also available on Amazon but it would make sense to order from someone with recognized expertise on the product (assuming roughly similar pricing and return policies). Also- if you have any details on what Atwood advised about voltage testing of the board, I'd appreciate seeing it.

Chris- I don't have enough experience with furnaces to have an opinion on the reliability of Atwood furnaces. I've used a Buddy heater in our van since they first came out but this furnace in the Hawk is my first and I have very little experience with it. I would have been fine with my Hawk not having a furnace at all but have to say I'm kind of enjoying learning about it. Having said that, I've also had my share of frustrations with electro-mechanical equipment and plan to add a Wave 6 to the Hawk (and will be taking the Buddy along until it proves itself!). I'm not a fan of the furnace's noise or the nearly-five-amp electrical draw.


PS- Youtube clip of the tester in action:
Thanks for the info OC. This is very helpful as you have given me another couple of alternatives if I give up on the Atwood. I getting ready to take off on a 2-plus week camping trip so I will soon know.

OC -
So I think what the advice from FWC about the board being okay referred to was that the board seemed to have power, as it was flashing the lockout code. Therefore, it was probably an issue with another component, such as your electrode issue. The Atwood tech support asked me to check the voltage going into the board and from the lead going to the valve (third from the left of the multi-lead board pin connector that connects to sail switch, thermostat, valve, and electrode (in that order L-R, if I remember correctly). I had 13+ volts going into the board, consistently, until I tugged a little on the lines coming out of the furnace cabinet and pulled out the positive lead going to the off/reset switch (will address in a sec with description of my resolution). The valve lead wasn't getting any voltage, so that's when I ordered the wrong igniter board (Dinosaur UIB S instead of correct Fan 50 Plus Pins model) from Amazon for $4 overnight shipping. That's when I realized the board was incorrect and had the great experience with Roy and Dinosaur. I finally got a hold of Chris (tech support) from Dinosaur yesterday and he had me check the thermostat lead on the pin connector on the board. No voltage, so realized I had an issue with the thermostat line - allowing me to troubleshoot the thermostat line and finding a loose connection at the electronic thermostat I had installed years ago. Fixed that and everything came on and worked like a charm! Getting the old board picked up by UPS today for a return, as Amazon appears to be honoring a return, despite everything I heard about Dinosaur not allowing retailers to accept returns.

So, moral of the story is: either I should have checked the thermostat connections before disassembling most of the components on the outside of the firebox ~or~ my old approach of disassembling everything and putting it back together seemed to fix the issue ;-) (at least for now - not really confident with Atwood furnace's reliability at this point, as new, inexplicable issues seemed to pop up throughout my troubleshooting process).

Also, some things to note from my experience. First, the positive power lead connecting to the main reset/off switch and then to the ignitor board came out with a slight tug of the bundle heading back into the furnace cabinet. This could be the cause of the issues I've been reading about with intermittent connections (and some of the strange issues that were popping up during my troubleshooting). I replaced the crimp connector and it all seems solid and reliable now. Also, I asked Chris at Dinosaur tech support about whether all the boards had the grommets at the corner board attachment points and he replied that they all did and it could be removed carefully with some different tools I didn't recognize... I think a replacement with a Dinosaur board, had I needed it after all, would have necessitated getting some longer screws to attach to the Atwood ignitor board cover that then attaches to the furnace.

Hope this answers your questions and others' that make come upon this thread when they get into the kind of mess we did!

-joe D.
Top Bottom