White Rodgers 1F97 Thermostat Behavior

Captain Who

I've been curious as to what sort of algorithm this thermostat might be following so towards that end I did some checking of my thermostat today from 9:50 am to 2:45 pm and it ran a total of 8 cycles with the raw data as follows:

Gas Valve On (13 s after thermostat on)___Thermostat Off___Thermometer 1___Thermometer 2___Thermostat___Outside Temp

9:50:09_____9:54:55______66.4______64.4______61______27
10:21:08_____10:25:55_____65.8_____64.5_____61_____30
10:31:08_____10:34:55_____65.9_____64.5_____61_____30
11:42:08_____11:46:55_____65.2_____64.5_____61_____33
11:52:08_____11:56:55_____65.1_____64.4_____61_____33
12:52:08_____12:56:55_____64.9_____64.2_____61_____37
1:02:08_____1:06:55_____64.8_____64.2_____61_____37
2:35:08_____2:39:55_____64.9_____64.4_____61_____38
2:45:08_____N.A._____64.9_____64.4_____61_____38

It appears that what this thermostat is doing is it is running a 10m cycle (5m on, 5m off), followed by a variable cycle (5m on, "variable" off). It only does this when it has recovered from the overnight setback, ie. in the steady state condition. When it is recovering it will run as long as needed until the thermostat is satisfied.

Cycle #1 (4:46 on, 26:13 off, 31m total)
Cycle #2 (4:47 on, 5:13 off, 10m total)
Cycle #3 (3:47 on, 1:07:13 off, 71m total)
Cycle #4 (4:47 on, 5:13 off, 10m total)
Cycle #5 (4:47 on, 55:13 off, 60m total)
Cycle #6 (4:47 on, 5:13 off, 10 m total)
Cycle #7 (4:47 on, 1:28:13 off, 93m total)
Cycle #8 (4:47 on, 5:13 off, 10 m total)

Avg. Duty Cycle = 12.6%
Avg. Outside Temp = 35.3 F
Avg. Cycle Length = 36.9m

This is in contrast to the Honeywell thermostats where you can set the cycle length and it will vary the on time as needed.

Fred

On the Honeywell you can vary the cycle frequency/hr. from 1 to 6 (presented in the menu as "Steam", "Hot Water", "Heat Pump", "Gravity Hot Air", "Forced air/fan", and Forced air/Central Air conditioning" but not the length. I'm sure there is a fixed algorithm for each cycle frequency selection. "On Time" is, as you say, however long it takes to satisfy the Thermostat relative to temp setting.

Captain Who

Yeah I guess I meant that since the cycle length is the inverse of the cycle frequency, setting the latter also sets the former.

Fred

The "On" times for each of your cycles seems terribly short. It still seems like your thermostat is set for Forced air and the extended "Off" times are only because the thermostat isn't calling for heat during the next thermostat check. I don't know how you can get steam to the radiators in those short "On" periods???

Captain Who

Isn't steam supposed to be able to travel at 60 mph? The radiators were hot, but not all the way across. There is only two settings for this thermostat: 0.6 F and 1.2 F. I have it on the former, but the latter may give a longer "on time". I have just received my hourmeter and counter and will be setting those up before I do any adjustments of any kind, to establish 7 to 10 days of baseline data.

I believe the thermostat uses the fixed 10m cycles to calculate how long the "off time" should be in the next variable cycle.

Captain Who

The 2 backup thermometers showed a remarkably stable temperature though, don't you think? I mean it couldn't have done that over a 5 hour period without steam in the radiators.

Mark N

The thermostat is set to 61 and the other 2 thermometers are reading approx 65, why is the thermostat even calling for heat? A call for heat on a steam system should run around 20 minutes. I can't see steam even being able to distribute on a 4:47 run time. Steam only travels a 60 mph once the air is out of the pipes.

Captain Who

You can't see it........you see it in the temperature data though. And I was sitting there the whole time and was comfortable despite sub freezing temperatures at the beginning and the radiators were hot after the burns. It's important in science and engineering to not argue against empirical evidence but to try to explain it. I guess that's why I posted it, because I knew it might be a bit controversial, since most people here seem to think a 1 hour cycle is the gold standard for steam. A 12.6% duty cycle ain't half bad though, to maintain a +30 deg temp differential to the outside.

Digital thermostats are notorious for mis-reporting the temp on their displays, I think. I should mention that this thermostat also has a remote sensor and the display is actually the average of the remote + internal, because I have the thermostat set to react to the equal weighted average of the two. I should have recorded the remote sensor temp data also but I thought that would be confusing because the 2 backup thermometers are right at the thermostat location. Having said that though, my previous White Rodgers 1F97 (which failed after 7 yrs) seemed to under-report the temp on its display also, by about 3 deg F. There is an adjustment to calibrate that up to 3 deg discrepancy but they swear that they accurately calibrate them at the factory. Hmmm. Thanks for commenting.

Mark N

You're right I can't see it. But what I do know is that it takes time to generate steam. Then that steam needs to travel down the mains to the vents. Then after that it takes more time for the steam to travel up the risers to the rads. Time how long these events take on your system and let us know the results..

Captain Who

How do you explain the empirical data if the steam was not getting to the radiators? I know I wasn't delusional that the radiators were hot to the touch, although not heated all the way across. Doesn't steam continue to flow into the radiators after the burner shuts off, to replace the steam which condenses there?

Mark N

How many sections of each radiator were hot? Were all rads equally heated? Were all the rooms at the same temp? How about some more info, 1 pipe or 2 pipe, and if 2 pipe what type. Is the boiler oversized? If so by how much? What is generating the steam that continues to flow into the rads after the burner shuts off?

Captain Who

This was not a post by me with the intention of saying everything must be perfect on my system......quite the contrary. I'm just starting to look for low hanging fruit to make improvements. It's a 1 pipe system with a 63-04L Peerless boiler. I have very large vents on the radiators downstairs (varivalves) and smaller ones upstairs (maid o mists) and only a Gorton #1 on the main, and the system is balanced (misbalanced) to produce even heat upstairs only after the setback really because after the morning, I spend 95% of my time downstairs. I haven't calculated the EDR of all my radiators yet, but my impression is that the boiler is a bit overpowered for the rads, especially since I know that two rads were removed from the original system.

So perhaps this helps to explain the discrepancy a bit for you, compared to what you would have expected. This is working very well for me and I'm tempted to let well enough alone. But I won't, not my style lol.

I was hoping also that other people with this thermostat would share their experiences, but maybe everyone uses the Honeywells.

As far as what is generating the steam after the burner shuts off, I'd have to say the steamchest, which would be the near boiler piping + the space in the boiler. Also the mains and the near boiler piping are very well insulated.

Mark N

The main venting should be many times greater than the venting on the rads. You want he steam to get to the end of the main before it goes to the rads. Is steam getting to the end of your main before the tstat turns off the boiler? It could be going right to the rad that satisfies the tstat because it has a varivalve on it.

Captain Who

Yes, exactly. I think that this venting configuration just happens to work great with this crummy(?) thermostat. Having said that though, I really don't know that this thermostat isn't just intelligent enough to have learned that it needs to do this extremely short "on time" in order to deal with that. It will be interesting to see what happens when I put 2 Gorton #2's on, and reduce the venting on rads. It might turn out that the thermostat won't be able to deal.

The steam is evenly distributing to all the rads (not just the thermostat rooms.....dual sensors) on the 1st floor (5 in total). I made sure of that years ago. I also have two ceiling fans blowing upwards (but not in thermostat rooms) The theory says that I should have lots of banging and boiling in the rads and so on with such large vents on them, but I don't. Also, the water in the sight glass is steady as a rock (no surging up and down at all). The water line does tilt about 1" towards the end away from the riser when the steam starts to generate however, which indicates the near boiler piping is inadequate and/or need more total venting.

nicholas bonham-carter

When the burner cuts off as the thermostat is satisfied, the steam production comes to an end. Steam remaining in the pipes condenses quickly, and air rushes into the vacuum in the system. With your tiny main vent, you can probably hear the radiator vents whistling, or wheezing as air flows into the rads.
Residual heat in the system comes from all the metal which has been heated up. If you had a two-pipe vacuum system, then steam production would continue, as the vacuum forms, lowering the boiling point of water in the system.
Try increasing your main venting, and decreasing your radiator vents.--NBC

Captain Who

Yeah I agree, with the exception that I think the steam in the radiators condenses quickly, but the steam in the insulated mains and near boiler piping and steamchest condenses more slowly and provides a reservoir of steam for at least some amount of time to flow into the vacuum that would be otherwise created in the radiators.

Fred

Per my initial comment above, water can't come to a boil and generate steam, heat and get through the mains, heat and move through the radiator runs and heat radiators in those short run times and that White Rodgers simply doesn't have the super intelligence to make that happen. Something else is going on here that we are missing.
Perhaps the radiators have some residual heat from the longer run time (to recover from your set-back, which you aren't showing that run time) that causes you to believe they are actually heating during the short runs?
In any case, if steam doesn't get to the end of the mains first, the radiators can't heat at anywhere close to around the same time and those towards the end of the Main probably won't heat at all (except perhaps during those longer runs).
Science has proven what can and can't happen with steam and your proposed senario falls into the "Can't" catagory. Just not likely and we are missing something.

Captain Who

The 5 1st floor radiators were quite cold before each of the pairs of short burns, rest assured. They were definitely hot after completing each of the 2 short 5 min burns, which was separated by a mere 5m off time, believe me (or don't). As I said, the 3 upstairs radiators don't really receive much steam with short run times, with the exception of the tiny bathroom radiator with a varivalve, due to their relatively small maid o mist vents installed. The setback recovery was programmed to complete by 7:45 am, and if you add another hour or so to allow for the walls and furniture to heat up, it is safe to assume its effects were done by 9 am. I didn't start recording cycles until 9:50 am.

It never really crossed my mind that people would be skeptical of whether my radiators were actually receiving steam. If I didn't know better I'd think you were pulling my leg. This is starting to remind me of a scenario where Newton yells at the apple and says you couldn't possibly have done that. It's all good for discussion though I guess. It was my concern that the thermostat seemed to be short cycling that caused me to sit down for 5 hours and closely observe it to see what was going on. If it had run a much longer "on time" it would have undoubtedly overshot the setpoint.

Mark N

You say you can't argue about empirical data, the stable temp. I don't know where you're located, so I don't know if was sunny where you live. It was sunny here in New Jersey yesterday and I checked how much my system ran yesterday. Between 8:30a and 4:30p, my boiler ran one time at around 10:40am for 18 minutes and maintained 70 degrees all day. You ran 8 times. Post some info, how long to make steam, how long to steam gets to the end of the main, how many sections of each rad heat on these short runs. It has been my experience that a 5 minute run time is not long enough to even get steam to the end of the main.

Captain Who

That wasn't what this post was about (at least for me). It was about the pair of short "on times" of fixed duration that this thermostat's algorithm employs, followed by variable "off times". Trust me it was not all due to solar radiation because it was quite frigid outside, although that did play a beneficial part in the heat loss on this day as it often does. The windspeed averaged 17.8mph with gusts ranging from 22 - 30 mph, which I didn't even mention!! So you still question whether there WAS steam in my radiators and that it is possible for my radiator vents (adjustable varivalves) to evenly distribute the steam to my 1st floor radiators, even with a small main vent? If a horse starts a race a bit later than the other horses, but it has a big heart and stronger lungs, can't it possibly win the race or maybe tie?

I am also in NJ (northern) but what were your degree hours between 8:30 am and 4:30 pm? When was the next time your burner ran after that 18 min run at 10:40 am? You're saying that it didn't even run again ALL DAY??? What was your average windspeed? Do you have a passive solar house with super insulation?

Mark N

Here is the actual data.
Boiler on: 8:25:14am
Boiler off: 8:48:27am

Boiler on: 10:36:04am
Boiler off: 10:51:27am

Boiler on: 4:10:40pm
Boiler off: 4:39:19pm

Boiler on: 6:20:59pm
Boiler off: 6:41:39pm

So actually threes 3 cycles in that time period. According to the nearest weather station the temps were:

8:27am 21.4 degrees
10:29am 31.5 degrees
4:34pm 38.3 degrees

wind between 5 to 15 mph all day.
Only insulation is in the attic.

Does your boiler maintain temp for a tankless coil?

Captain Who

So your Duty Cycle for those 3 cycles was 11.3% vs 12.6% for my diseased system over my 8 cycles. Not a heck of a lot of difference and you had that one aberrant cycle that was 5:34:36 long vs 2:10:50 and 2:10:19 for the previous and next cycles. Are you sure you are not missing a cycle there? Your on time was only 15:23 in that aberrant cycle vs 23:13 and 28:39 for the previous and next ones.

No I don't have a tankless coil. I have a separate gas DWH.

Mark N

Your boiler is on pretty consistently for 4'47". How much steam is produced in that amount of time when you factor out how long it actually takes to start producing steam. You might only produce maybe 2 minutes worth of steam. I would set your swing to the 1.2 degree setting and see how your system operates or buy a thermostat that has a cycle setting for steam. Add up your total EDR or all your rads, your boiler might be very oversized.

Captain Who

I can pretty much tell by its behavior that it is DEFINITELY oversized. It only took about 40m (approx!) to recover from a 8 deg setback this morning. I'll get around to buying Dan's book and calculating the EDRs at some point. I'll probably put back the missing 2nd floor radiator also at some point, as well as adjust the upstairs rads so they get the steam equally to the 1st floor. I forgot to mention that I have the upstairs guest bedroom radiator shut off at the valve also. All that will give me more even heating throughout the house and more boiler load, but will COST ME MONEY.

What everyone seems to be neglecting is that it is really 2 x 4:47 (the thermostat thinks its 2 x 5:00) because it does the double short cycles separated by a 5:13, which is practically no time at all for cooling to take place. The rads do not get much steam after the first one, but after the 2nd one they are hot. I'm not sure, but it seems to be doing triple short cycles today......not really paying close attention. I don't have a datalogger.

Captain Who

PS: I will do the 1.2 F setting after I install my hourmeter and counter and collect some baseline data. Probably the first thing I'll do.

Mark N

OK 4'47" on and 5'13" of adds up to 10 minutes that is a cycle rate of 6 CPH. When you're recovering from set back is it one 40 minutes long run or are you cycling on pressure? Are the vents hissing while recovering? The input to your boiler is fixed it costs you the same per minute of run whether a rad is turned on or off. If your boiler is oversized, you make it more oversized by turning off rads and then you're producing higher pressure and that costs money.

Captain Who

No, it was 4:47 on, 5:13 off, 4:47 on, variable time off.........lather rinse repeat. At least that is what it was doing yesterday. Today it may be doing triples.

If you increase the EDR and vent appropriately for more even distribution, such that you are now heating the upstairs where you weren't previously (or minimally except for recovery), you will have to run a longer amount of time to satisfy the thermostat which is downstairs. If you spend 95% of your time downstairs you will be spending a lot more money for heating with minimal benefit.

Yes the vents do make more noise during recovery than they would if I had generous main venting and they may wear out sooner due to particles spewing forth. Yes it does cycle on pressure x number of times towards the end depending upon how cold it got that night and therefore how cold the walls and windows are. I will be installing a dwyer-stat at some point in time when my 1823-20s arrive and I purchase a relay (thanks MarkS) and that should save a little more duty cycle wise although it may cycle more times on pressure on recovery.

Mark N

If your running a setting of 6CPH, that would be a 10 minute cycle. It runs 4'47" sits off for 5'13", you didn't reach the set point so you run again for 4'47". This time after the 5'13" you make the set point, so now the thermostat is satisfied and remains off until the temp falls to the set point minus the differential. Then repeats the same cycle.

Captain Who

Yeah except that's not what it has ever done YET, at least not while I've been watching. The next cycle after the fixed 10 m cycle is a "variable" one. If you check my data again in the original post, you'll see that the cycles averaged about 37 mins.

I think you're mixing up the thermostat with the pressuretrol, but I think I get the gist.

Mark N

Of course its variable, nothing is being timed. The thermostat doesn't call for heat again until the temp falls to the set point minus differential. This will take however long it takes.

Captain Who

I suspect that it is timed or calculated actually. It must be making a heat loss calculation and determines the variable "off time" from that. That's not to say that the calculation may not be overridden by an actual sensor reading that is not as predicted. When it is maintaining such a tight range of temperature control it pretty much has to behave this way. Check my data where you can see how little change there was in the readings on the two digital thermometers, especially the 2nd one.

IMHO

Fred

Captain Who said:

I suspect that it is timed or calculated actually. It must be making a heat loss calculation and determines the variable "off time" from that. That's not to say that the calculation may not be overridden by an actual sensor reading that is not as predicted. When it is maintaining such a tight range of temperature control it pretty much has to behave this way. Check my data where you can see how little change there was in the readings on the two digital thermometers, especially the 2nd one.

IMHO

All I can say is that must be some thremostat!

Captain Who

Yep Fred. I say that because the "off times" were at values as follows: 26:13, 5:13, 1:07:13, 5:13, 55:13, 5:13, 1:28:13, 5:13.

Do you see a pattern there that indicates it may not have been entirely random?

Mark N

A thermostat is just a temperature activated switch. There is no pattern. The 5'13" is when the set point is not met and it runs again. The others are when it is met and how long it takes to fall to the set point minus differential.

Captain Who

Okay if you say so

Fred

Mark N said:

A thermostat is just a temperature activated switch. There is no pattern. The 5'13" is when the set point is not met and it runs again. the others are when it is met and how long it takes to fall to the set point minus differential.

I think this is falling on deaf ears (or rather blind eyes).

Captain Who

There is no pattern? Okay guys. If the thermostat is operating ENTIRELY on when the temperature drops to the setpoint minus the differential, HOW do you explain that the variable "off times" were in EXACT to the second one minute increments? If you won't explain that I see no need to continue this because you are the ones with deaf ears and blind eyes.

Why do you think that the thermostat couldn't calculate a rate of change of heat loss, ie. Delta Temp vs Delta Time and therefore calculate the "off time"? It is basically a simple computer and associated programming combined with a timepiece and a temperature sensor.

I'm not saying that it would not have adapted if something drastic had happened like if I had opened the doors and windows. That would've disrupted the calculated "off time" and overridden it.

Mark N

Don't you think that if your thermostat could do what your saying, that White Rogers would be using it as a selling point.

Captain Who

I don't think it is anything extraordinary, actually.

Fred

It's an older model Thermostat that has some type of spring loaded anticipator that lets you choose between .6F and 1.2F. There is no extraordinary intelligence built into that unit, like a dynamic heat_loss calculator.
I can't explain your numbers because I wasn't there to see what was happening. All I am suggesting is that no one else on this site boils water, heats a header, mains, raditor runs and radiators in 4 minutes, on finely tuned systems and you aren't doing it either on a system that you say has inadequate venting on the mains, may or may not have an over-sized boiler and short cycles on pressure.
Is it possible that your system mysteriously fell into some unnatural state of optimum performance? Yes that's a possibility but just not likely.

Captain Who

Okay Fred. It isn't an older model. It's a new revision of an older model. I just bought it. But you seem to be unable to answer my question of why the "off times" are in exact one minute increments to the exact second. I figured you were too pig headed to address that