"Oh its not the t-stat now is it?" I don't know Jeff. ".. sounds like a lockout to me." A lockout needs a hard reset. Once a primary trips - it won't reset simply by cycling the breaker. You have to re-arm the primary manually. "or a number of things that can't be checked over the internet" True, Jeff but diagnosing a lockout over the internet is no more sure than diagnosing a t-stat over the internet. Jeff - no slams here; just diagnosing what we've been told. Cheese- All the stat does is make and break a low-voltage connection. Essentially it's a part of a relay circuit. It opens and closes what is usually a 24v circuit, that's all it does. The stat gets it's working power from the batteries. If the batts are weak; the stat won't reliably close the connection. IOW it isn't always telling the furnace "it's cold, make heat." Jeff wasn't wrong-I just don't believe that a diagnosing a bad t-stat is over your head, 24 volts will not hurt you like 120v or 240v. If you lose power in your display - it's not the 24v from the furnace. It's the stats internal battery power failing - or some other element of the stat failing ie. bad component on the board, or bad contact on the board etc. Are your batteries new? -br

Thanks for the responses! It turns out that Mike was close to the answer. My aging ears were playing tricks on me. A hand on a register confirmed that the blower does NOT start immediately when the thermostat calls for heat. A little cheating of the access panel interlock shows that when the furnace starts, the ignitor immediately begins to heat. This, with the draft fan, accounts for the initial 6+ amps. After some seconds (10-20?) the ignitor reaches ignition temperature, the gas valve opens, and poof, the burner ignites. Then the ignitor shuts off. Now only the draft fan and gas valve are drawing current, hence the drop to 1.6 amps. Several more seconds pass and then the heat exchanger temperature rises enough for the blower to start. Then the operating current again rises, to around 5.7 amps. I still don't know why the current gradually falls from there to around 5 amps as the furnace operates, but at least the first part of the cycle makes sense. As an aside, I wondered whether the AFUE rating of modern furnaces takes into account the electrical energy used to burn gas more efficiently, and how this would shake out in the efficiency game. If the gas valve and draft blower together draw 1.6 amps, that's about 192 watts, equivalent to about 655 BTU/hr if it were passed through a resistive heating element. That's not a huge percentage of the furnace's heat output, if my math is right - for a 40k BTU, maybe ~1.5%, if my math is right. So if a traditional draft hood furnace was, say, 60% efficient (I'm making that number up; how [in]efficient *were* those old furnaces?) obviously we're ahead with the modern furnace. The next question is, does a 96% AFUE unit use more electricity than an 80% AFUE? How much more efficient is it really, after considering the additional electrical energy input (if any)? There I have no data - anyone?

It could be just about anything. I was woken up by voices in the middle of the night, once. It took me forever to figure out where they were coming from, but it turned out to be my computer speakers picking up someone's HAM radio conversation. I had to unplug them to make it stop. I plugged them back in the next morning, and it has never happened again. Oh, just thought of another one! I have a Curious George doll for the Grandbabies. He giggles when you squeeze his feet/hands. He sits in a basket under an end table, nice and quiet until the kids are visiting. I switched the wooden table out to one that is glass, and George started giggling on his own at random times, but I couldn't figure out what was causing it. Turns out the igniters on my gas cooktop were doing it. Apparently he was shielded under the wooden table, but not the glass.

Update and solution for those having a similar problem. I pulled the wires from the thermistor sensor on the body of the heater and put a 10k resistor across the ends. In theory, the control board should have read something like 77 degrees but the temp display exhibited the same erratic behavior. To eliminate the wires as a source of error, I then hardwired the 10k resistor across the two thermistor pins on the control board. Alas, same problem. That meant control board, unfortunately. I bought a new Pentair board on-line for about $310 plus tax. Took me all of about five minutes to install once the section of the heater with the board and display membrane is able to be removed and turned upside down. Guess what? Everything works perfect now. Only downside is that the temp readout on the Pentair board is less bright than what I had on there before, but it's not a big deal. End of problem. I love those sorts of outcomes! Nick in Palm Springs