Suspected short cycling. Internet enabled my 15 year old hydronic radiant system

rolan

I recently mated sensors onto a Wirsbo zone control board to monitor thermostat/circulator/boiler activity of my radiant heat system. All data is logged and sent to a central server which reports on to a web page:

https://heat.omnistep.com/?d=2019-12-11

Clicking on each zone gives detailed reports about individual events and averaged hourly cycle times.

I'm the second owner of a hydronic radiant heated house built in 1995. Most everything in the system is original. 4 out of the 6 zone thermostats are of the round mechanical Honeywell T86H variety. Examining the data, it appears the analog thermostats are short cycling and at times causing the boiler to short cycle as well (greater than 10 cycles per hour on for 3 minutes each). Is this something I should pursue fixing? Two years ago the boiler relay burned out (operating for 23 years, I guess it was acceptable). Also the flue flange butting against the basement wall is corroded and falling apart which I hear is a symptom of short cycling.

Open to hearing any feedback about the web reporting system at the link above. The sensors feed into a tiny microcontroller box that transmits data over wifi. I could post photos and more details of the hardware if there is any interest.

~Rolan

ratio

There is.

Jamie Hall

On the round Honeywells. They are unlikely to be failing. They are also very unlikely to be -- directly -- the cause of the short cycling. However, you should check that the anticipator -- I think the h model still had an anticipator, could be wrong -- is set correctly.

That said, none of the thermostats should be wired directly to the boiler. They should be wired so that they control zone valves or pumps -- whichever you have -- and the boiler should be controlled solely by an aquastat (preferably with outdoor reset). With that many zones, if they control directly, it is almost inevitable that they will be out of sync with each other, and just about the time one of them quits calling another will wake up.

hot_rod

What brand and type of boiler. Radiant in slabs?
Generally speaking you want the boiler to run 10 minutes or more every time it fires.

That becomes a challenge with a fixed output boiler and multi zones as you have.

a modulating boiler helps eliminate much of the cycling, in some cases a buffer tank is a good addition.

Tim Potter

If I might ask, what are you using for a data logger?
It looks good.

I have 7 round Honeywell mercury t-stats in my system, no problems.
They dont short-cycle. I agree on the check the anticipator solution.
The zone valve barely gets open in 3 mins, let alone heat the zone to set-point.

Thank you,

Tim

rolan

Burnham gas boiler. Radiant in slab. The boiler does not modulate. No outdoor reset. The T86H Honeywell rounds have a fixed anticipator that consists of 0.5 ohm resistor positioned against the coil. It's described in this document as is the zone control module which interfaces between the boiler and the thermostats/valve switches.

https://www.uponorpro.com/~/media/extranet/files/archived docs/t86h thermostat technical bulletin.ashx

I believe the short cycling occurs when thermostats make their brief 1-2 minute calls for heat while the boiler temp sits just below the aquastat threshold. Zone "BR1&2" is very active compared with the others. The thermostat is positioned in a hallway which may be prone to drafts.

I'm going to first try and swap out the thermostat in question with a digital where I can set the cycle rate/hr. If that doesn't exist, I could override the thermostat at the zone control module with a circuit hack that holds the thermostat lines closed for a minimum of time (5-10 minutes?) when a call for heat is triggered.

Buffer tank would probably be the last resort fix as that requires a lot of work.

Jamie Hall

Situated in a hallway which may be prone to draughts... ah, no. Move the thermostat. Thermostats -- of any kind -- do not like draughts. No matter how fancy a modern digital you get, it's not going to help, although it may mask the symptoms.

Also note that that particular variant on the T87 requires a separate 24 volt supply -- which I presume you have? As the Honeywell document you referenced states, lack of that supply can and will cause erratic operation -- particularly short cycling.

This is, fundamentally, a control problem for the boiler. There should be at least a 20 degree swing in boiler water temperature between on and off. That is, the boiler controls should allow the boiler water temperature to rise to a certain point, then the boiler should shut off and stay off until the water temperature drops at least 20 degrees. Then the boiler should come back on and stay on until the temperature rises to the upper set point. A thermostat somewhere in the house turning on or off should have no effect on the boiler operation whatsoever in an application such as this one.

The first place, then, that I would start looking is at what controls your boiler. Forget the thermostats for the time being until you get the boiler operation straightened out.

hot_rod

Observe the operation, is the t-stat still calling when the boiler cycles? If so the issue is the large size mismatch, boiler to load.

Really a t-stat is merely a high limit device, calling for heat until it's setpoint is reached

Electronic radiant specificities stats have some logic onboard, PWM for example.

rolan

Jamie Hall said:

...A thermostat somewhere in the house turning on or off should have no effect on the boiler operation whatsoever in an application such as this one.

The first place, then, that I would start looking is at what controls your boiler. Forget the thermostats for the time being until you get the boiler operation straightened out.

The thermostats call for heat to the Zone Control module which sends 24V to the corresponding zone valve actuators. When the valve actuators open (closing the feedback switch) the Zone Control module responds by energizing a relay that switches on the boiler. The boiler then fires up until the water reaches the desired temperature. If at any time the zone valves close, the zone control module relay shuts off and the boiler shuts off (even if it never reached full temperature). So in summary as long as any one valve is open in the system, the boiler comes into action (I discovered this when some of the older motorized valves began to fail in the open state). When the zone valves shut off, so does the boiler.

What's happening is that the thermostats+valves are short cycling (between 69F-70F every 1-5minutes) and the boiler is hovering around 120F but never reaches 140F before the call for heat is shut off.

Oh and I did check the thermostat to confirm 24V is running to the 3rd wire that powers the fixed anticipator.

JUGHNE

FWIW, I have seen the old HW round tstats have a nearly shorted heat anticipator that went to "high heat" within the tstat when ever it called.
(Sorry Jamie )
This caused short cycling regardless of where the anti setting was.

Jamie Hall

@rolan, your description of how your system operates is, no doubt, the way it is wired to operate. Unfortunately, at least from my point of view, it's wrong -- particularly where you have multiple, unsynchronized thermostats.

The boiler should be controlled by its aquastat, and nothing else. The way you have it wired is guaranteed to give exactly the behavior you are seeing, which is very undesirable.

hot_rod

under low load conditions the boiler may not reach its setpoint before it shuts off the burner. ideally we like to see the boiler run 10 minutes on each light off, but it may not on low loads.
A conventional cast boiler should see 140 return temperature for a period of time also to prevent corrosion inside and in 1the flue piping. It really should get above condensing temperature condition on every cycle in a perfect world..

If you set the stat for 70 it should run until 70 is reached. It will start again at 69 if the stat has a 1° differential built in. Some stats are adjustable to 2 or 3 degree differential, that may help, but comfort levels may be noticeable at wider differential settings.

A boiler ideally is sized to run non stop on a design condition day. With a fixed output boiler, of course it will cycle more often on low load conditions, as it has only on, full speed ahead operation mode.

Jamie Hall

@rolan 's problem, @hot_rod , is that he has seven (I think..) independent thermostats fighting for control. Not going to work. He's pretty got to pipe the thing pri/sec and use the boiler's aquastat alone to control the boiler.

rolan

https://heat.omnistep.com/cycle_detail.php?b=1&z=32&d=2019-12-14

Looking at this thermostat log, I think I'm going to first try inserting a delay circuit between the zone control module relay and the boiler such that it only passes the 24V if there has been a request that lasts for at least 3-5 minutes. That way the valves and circulators can short cycle as much as they want with the residual 120F water in the system but the boiler will only kick on when really needed.

rolan

Just as a followup, I inserted a ICM102 delay on make timer between the control board and the furnace on the 24v line (5 minute delay). Cycle times look a lot better now.

Before: https://heat.omnistep.com/cycle_detail.php?b=2&z=8&d=2019-12-14

After: https://heat.omnistep.com/cycle_detail.php?b=2&z=8&d=2019-12-19