Finally some Colder Weather Data

PMJ

Finally got somewhat colder temps (24 hours at 20F+/-) to see more data from my system that adjusts cycle burn times and wait times based on demand using one remote temperature sensor and a system pressure sensor.

To review, during calls for heat all burns run a set timed amount after two conditions are met - steam arrival at the feed pipe of my most remote radiator in the garage, and, that all the vacuum from the previous cycle is gone. When both of those conditions are present a timer starts and ends that burn 210 seconds later. If the call is still in progress the next burn starts after a 240 second timer which is started by that same remote temp sensor when it opens.

It turns out that in mild conditions wait periods are long and vacuum is almost always gone before steam arrival and the sensor due to longer reheats. As it gets colder, however, steam arrives before vacuum is gone and so the vacuum amount serves to extend the burn time that is actually filling radiators. Today, steam arrived 4 minutes after fire and vacuum was gone about 1.5 minutes after that. It seems that vacuum is a real time indicator of the demand, and so can be used directly to adjust burn times to the conditions with no special equipment beyond knowing when the system has reached atmospheric each burn which can be done with a very inexpensive sensor.

Anyway, I reset the data collection 24 hours ago and got my wife to leave the thermostat alone at 70 for one day. On the time to steam chart you can see the new calls for heat where time to steam was a little longer. Consecutive burns all have arrival times of 4 minutes +/-. This time to steam is from the moment the vent damper starts to open to when steam arrives at the most remote location in the system. For reference the first time to steam after a 4 hour setback is about 14 minutes.

The system is running along just barely satisfying the stat a few times and never for long because there is no overshoot. It is calling for heat 71% of the time/ 10 calls in 24 hours. The call just prior to the one showing on the main screen was over 3 hours. The burn time percent is around 26 as opposed to 20 or so at 35F. Burns get longer and waits shorter as it gets colder with the bigger change in the wait times.

This is really even heat and could be done without a PLC. I am using one because it is easy to make changes and is required to collect data. Cycles per hour is easily changed with those two timers for those who want fewer cycles and don't mind more swing. There is nothing magic about my two numbers. But I have left them alone for two seasons now while I study how the system responds to conditions. This can also work without vacuum quite well, it just isn't quite as responsive to the demand and needs to be set for longer fixed burns with the timer.

Attached are the screens.

EBEBRATT-Ed

So when I look at burn time versus hours/day it looks like in 25 hours you had 43 burn times, so that's 1.72 burner starts /hour on average.

Am I looking at this right?

Your system looks pretty impressive

PMJ

EBEBRATT-Ed said:

So when I look at burn time versus hours/day it looks like in 25 hours you had 43 burn times, so that's 1.72 burner starts /hour on average.

Am I looking at this right?

Your system looks pretty impressive

Yes, you are seeing it correctly.

STEVEusaPA

Cool stuff.
You're second screen shot looks like your control is giving me the finger (lots of them).

AdmiralYoda

@PMJ good work! Out of curiosity what pressure switch are you using?

My system is as hot as it can get and full of steam at 4oz of pressure. I've been thinking of putting a variable pressure switch at the boiler to trigger a time delay of say 15 minutes or so when it hits 4oz. This will give the radiators time to give up some heat and if after the delay time if the T-stat is still calling for heat the cycle will start all over.

My boiler is oversized and cycles on pressure when its 20F and colder. This would tame it a bit.

PMJ

@AdmiralYoda ,

Its been years since I actually looked at it but I will get you the make and model. Pretty sure it is an ultra low pressure HVAC Dwyer diaphragm switch that is adjustable to fractions of ounces and you pick a range to be in. I'll check it out and let you know.

PMJ

@AdmiralYoda ,

It is Dwyer model 1910-00. Photos attached.

These are for sensing ultra low pressure differentials in HVAC applications. It worked for me because I just wanted to know when the dry return just cracked over atmospheric. It was exposed to vacuum all this time supposedly over the max pressure range but the vacuum never damaged it. I have two of them and the other one in service has 1000's of negative cycles on it now. Can't miss at $45.

Max on the highest range unit is 5.5 in H2O which is just over 3 oz so maybe doesn't work for you. I'd be advising stopping short of 4 oz anyway but I'm sure there are other options covering the whole range.

ethicalpaul

got my wife to leave the thermostat alone at 70 for one day.

This is the most technically impressive feat I have seen in quite some time!

KC_Jones

ethicalpaul said:

got my wife to leave the thermostat alone at 70 for one day.

This is the most technically impressive feat I have seen in quite some time!

My wife doesn't touch it, and even if she did, the ecoSteam is in control so the thermostat won't do a whole lot.

Kjmass1

I’ll be hitting design day temps in Boston Fri-Sunday, 9F with windchills of -6F in the mornings. Looking forward to putting the timer to use. 

When temps are consistently in the 20s steam heat is golden. Silent, short run times, always warm.