Questions about Delta-T

EricPeterson

I know the definition is simple - the difference between the temperature going into the system and the temperature of the water returning from the system.
But my question is how in practice do you measure this difference?
My house is a converted gravity system with three zones, and setback thermostats for each of these, which means that at night the boiler might not run for a few hours as the house cools down, as well as all the water in the system.
I also now have a thermic bypass regulating the temperature of the water returning to the boiler, with a setpoint of 140F.
So obviously I would not measure delta-t in the morning when the system starts up.
It seems I would have to wait until a slug of heated water has passed all the way through all radiators in the system.
Having three zones complicates the picture, so I must either measure each zone separately, or wait until I know water has passed through all radiators in all zones.
The thermic bypass further complicates things, as it is feeding heated water back to the return until the valve fully opens (at 140F).
Putting this all together, I think I would need the following conditions to measure:

• Return water temperature of 140F (thermic valve fully open).
  
• All zones of the system calling for heat.
  

Alternatively I could perform this measurement with one zone at a time.

Oh and is there a similar designation for how much the water temperature rises passing through the boiler? That is, the difference between the temperature of the water leaving the boiler and the temperature of the water going into the boiler?

Comments?
Eric Peterson

EricPeterson

Oh and another option would be to bypass the thermic valve (which I can do).

When measuring delta-T, does it make any difference what the water temperature is?
For example is it equally valid to measure when the boiler supply temp is 170F or say 140F?

Thanks,
Eric

Jamie Hall

The formula works whether you use it for the system or the boiler -- just with the boiler it's giving you power input from the boiler, and from the system it's power used by the system.

And it doesn't matter what the absolute temperatures are -- it's the difference between them.

EBEBRATT-Ed

Regardless of what the boiler is doing the radiation is the driver of the system. If the radiation puts out more heat than the boiler can put into the piping the TD will rise. If the radiation capacity is less than the boiler output the TD will drop.

Most conventional Hot water systems in the past were designed with the standard 20 deg. TD which is 10,000BTUs delivered for each GPM of water circulated.

Gravity systems are different because of the large water content. They have a big lag in coming up to temp because they use a lot of heat to heat the large volume of water and the cold radiation and large piping.

The thermostatic valve you have protects the boiler from low return water temp which can cause boiler issues.

I don't know what your question n is, but if the valve is protecting the boiler properly then you should be ok.

hot_rod

The delta T is very dynamic in a hydronic system, as it should be. When the system starts from a cold condition, the ∆ will be wide, even wider than the designed for ∆ in somem cases.

Wide ∆ indicates a lot of heat is bring transfered 500 X F X (∆T)
If you have a fixed flow rate the math is easy 10 500 X 10 gpm X 30° = 150,000 btu/hr being delivered.

The ∆ will close up as the system warm as the 500 X 10 gpm X 15° = 75,000 btu/hr delivered.

When the thermostat shuts off and the pump stops 500 X 0 X ∆T = 0

To know the exact operating J at any condition you need to wait until the system reaches thermal equilibrium.
You know that when the SWT and the RWT have stablized. In a high mass, high volume system that could take some time, maybe hours when the home is under a cold condition, design day perhaps.

These Azel gauges are handy as the capture the highest temperatures. Or a datalogger if you don't have the time to sit in from of it.

When you get to thermal equilibrium, look at the boiler temperature. That is all you are going to get. Nothing you can do to that boiler to raise the temperature. Changing the aquastat, high limit settings will not, cannot change the boiler operating condition
The heat emitters are dissipating all you can supply them.

The mixing valve does not change this concept of thermal equilibrium, it basically allows the system to reach equilibrium safely, without condensing the boiler for extended periods, in your case.

It would be good you know that on your system, where exactly is you boiler temperature when you reach thermal equilibrium? Maybe it never gets above 160F? And if so, does the home maintain the temperature you desire at that operating temperature. This will indicate the heat emitter to boiler output match up.

It is the tell all to how well you did the math Load calcs, radiator output, boiler size, etc.

Another way to look at ∆T is an indication of the heat energy being transfered

EricPeterson

@hot_rod - someway somehow someday I gotta buy you a steak dinner. Or whatever your favorite meal is.
In the meantime thanks for this detailed analysis.

hot_rod said:

The delta T is very dynamic in a hydronic system, as it should be. When the system starts from a cold condition, the ∆ will be wide, even wider than the designed for ∆ in some cases.

Wide ∆ indicates a lot of heat is bring transferred 500 X F X (∆T)
If you have a fixed flow rate the math is easy 10 500 X 10 gpm X 30° = 150,000 btu/hr being delivered.

The ∆ will close up as the system warm as the 500 X 10 gpm X 15° = 75,000 btu/hr delivered.

When the thermostat shuts off and the pump stops 500 X 0 X ∆T = 0

To know the exact operating J at any condition you need to wait until the system reaches thermal equilibrium.

I am unfamiliar with the term "operating J".

You know that when the SWT and the RWT have stablized. In a high mass, high volume system that could take some time, maybe hours when the home is under a cold condition, design day perhaps.

These Azel gauges are handy as the capture the highest temperatures. Or a datalogger if you don't have the time to sit in from of it.

Yes this is what I was trying to say, how to know when the system reaches that state as to when to measure the temps.
Those Azel gauges look like just what I would need for this.

When you get to thermal equilibrium, look at the boiler temperature. That is all you are going to get. Nothing you can do to that boiler to raise the temperature. Changing the aquastat, high limit settings will not, cannot change the boiler operating condition
The heat emitters are dissipating all you can supply them.

The mixing valve does not change this concept of thermal equilibrium, it basically allows the system to reach equilibrium safely, without condensing the boiler for extended periods, in your case.

It would be good you know that on your system, where exactly is you boiler temperature when you reach thermal equilibrium? Maybe it never gets above 160F? And if so, does the home maintain the temperature you desire at that operating temperature. This will indicate the heat emitter to boiler output match up.

The tricky part with my system is that the three zones do not heat equally, because of the different heat emitters and insulation levels in the house. I would probably have to start from a pretty low temperature to have equilibrium with all three zones open and the boiler running. I could also lower the boiler setpoint.
Alternatively I could measure each zone independently.
Another factor it seems to me is the boiler will cease firing when it reaches its setpoint, then resume only after the temperature drops by the differential amount. So for example the boiler will cease firing at 180F then resume at 165F. So what value would be used for the SWT?

It is the tell all to how well you did the math Load calcs, radiator output, boiler size, etc.

Another way to look at ∆T is an indication of the heat energy being transfered

I look forward to doing some measurements this winter.

Eric

hot_rod

If you are doing the math, use the AWT. Fin tube for example, if SWT is 180, and the RWT is 160, the average is 170.
Look up 170 in the output chart and you get an accurate heat output of the fin tube.

With old radiators you would need to find output charts and see what temperature range they show for output. Still use AWT regardless of the type of heat emitter.

Regardless of the SWT at any point, what you want to see is the difference, ∆T between the SWT and RWT not moving.

I know what you mean with high mass systems. I was trying to get some output data from this radiator at different flow rates and SWT. Just this one radiator would need hours of run time to stabilize. Maybe overnight.

EricPeterson

So I hooked up the temperature sensors yesterday. While it's interesting to watch the numbers change, the device is disappointing in that it only records max/min values for supply/return - it does not for example record the min and max deltas between supply and return.

The numbers vary wildly as I expected based on my three zones and and the varying temperatures of the water in all those zones. Also with the boiler setpoint at 180 and a 15 degree differential, the supply will heat up to the setpoint and then drop before the boiler fires again. This alone results in a 15 degree swing.
For some of the zones it takes a long time for the heater water to circulate all the way through the zone.
It seems to me there would need to be some set procedure to follow in order to measure delta-T, if such a measurement is even feasible.
As you watch the display, the supply temperature will fluctuate, and in general the return temperature will always be slowly rising.

We setback the thermostat at night, here are measurements from this morning when the house was heating up.
This is the last reading after the call for Zone3 was satisfied.

This is the last reading after call for Zone1 was satisfied.

This is the last reading after call for Zone2 was satisfied.

Zone1: 32' of 9" CI baseboard, 80 ft run.
Zone2: 14' of 9" CI baseboard, 3 free-standing radiators, 78 ft run.
Zone3: 48' of 7" CI baseboard (31' active), 6 free-standing radiators (4 active), 104 ft run.

Just now I watched as the delta was around 30 running zone3, then drop down as low as 14 after zone1 fired, then creep back up into the 20s. Of course I cannot watch all the time, what I would really like is a temperature monitor that provided history over time so that I could graph the behavior.

My question with all this is not academic, but to help in determining which circulator would be better than the NRF-22. Regardless of the delta-T value, the system works great, the only issue is the whining noise when the NRF-22 starts up and the Caleffi valve is closed. That's been mitigated temporarily by allowing some return water to bypass the Caleffi valve.
The house is heated well and evenly, there is no longer any air accumulation, and the boiler cold return water problem has been fixed.
So all in all I'm a happy camper.

Eric

EdTheHeaterMan

Operating J is a typo. Bob was trying to type a ∆, not a J. ∆ is typed on an apple by holding down the command key and then the J key at the same time.

jesmed1

EricPeterson said:

what I would really like is a temperature monitor that provided history over time so that I could graph the behavior.

I've had good success with these data loggers. I foil-tape the probes to metal pipes, press the "record" button, and leave it for as long as I want it to record. Then plug the logger into a computer and upload the data. The interface software is a little odd, but once you get the hang of it, it works fine. It produces a graph of temp vs. time, so if you have one logger on the supply and one on the return, you can graph both, overlay the graphs, and see how the delta varies.

https://www.amazon.com/gp/product/B01MY9A0G1/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&th=1

hot_rod

Sounds like more of a balancing issue than a circulator. If everything heats adequately and comfortable.

Zone 1 is easy, pull out the data sheet on that bb and see what the output shows. I know copper fin tube generally shows two flow options. The 4gpm eeks out a bit more.

The tricky one will be determine the proper flow for the loops with bb and standing radiators

Then, ideally you would have a balance valve on each circuit to match the gpm to the load that those emitters require.

If the 3 zones are controlled by zone valves, delta p circulators can provide some pretty good balancing

EricPeterson

Thanks @jesmed1. I'll have to give this a try.
For the adventurous, one could also build up a device with Raspberry Pi.

I know that Zone1 is over-radiated and clearly will have a lower delta-T, whereas Zone2 and especially Zone3 would take a lot longer to stabilize for any meaningful measurements.

@hot_rod - looks like I'm back to the Grundfos circulators.
SupplyHouse has the UPSe 15-58FR and the more sophisticated ALPHA 15-58FR.
Do you think the extra features of the ALPHA are worth it for my situation?

But like I say the system works well now except for the initial whining (and some may my continual whining on this forum), if I can address that with a different circulator that still satisfies the needs of the rest of the system then that will be money well spent.
All the work I put in to tuning this system (pumping away, replacing finned baseboard with CI, adding emitters for an addition, splitting original house into two zones, adding zone valves and TRVs, adding insulation) has been to provide even heat throughout the house. Other than a heat-loss calculation done when replacing the boiler, I have not done much in the way of static analysis. The one nagging issue of low-temperature return water has been addressed along with the other issue of air accumulation.
Thanks to lots of help on this forum over the years I have made much progress.

Thanks everyone,
Eric

hot_rod

Use and rearrange the Hydronic Formula to answer some of the questions. Confirm observations, predict future additions, troubleshoot, etc.
Q is the rate of heat transfer into or out of a fluid stream

Q= 500 X flow rate X delta t


We went through an example a few posts up.

Rearrange that formula to estimate the temperature drop to required deliver the heat.

Yow want 50,000 into zone 2 lets say. You know or measure a 4 gpm flow

∆T = Q ÷ 500 X flow rate 50,000 ÷ 500 X 4= 25 delta t is where you are transferring that 50,000 at a fixed 4 gpm.

OR if you what to know what flow rate you need to deliver a 50,000 btu/hr load with the temperature drop at a 25 ∆

f= Q÷ 500(∆T) 50,000÷500(25) = 4 gpm

So this show you what is happening as those delta T change as you open and close the 3 zones.

With a fixed speed circ the flow rate doesn't change, so the less restrictive, lower head, zone gets more flow the others drop. The delta moves, the heat output changes.

What the delta P circulator will attempt to do is maintain the required gpm to each zone as the 3 open and close. That assures you get the gpm, the heat flow you always need.

So plug in those changing delta T you noted a few posts up and you see how the thermodynamics change in you fixed speed circulator zone valved or TRV system.

So to make a long answer long, yes the delta P circulator will make the system operate more to what you want or expect.

If you like or want the additional adjustability and like to get data from the circulator, operate or adjust it it while you are in bed, then the Alpha 15-58 is the better option.
Or how much $$ you want to get a more perfect system:)

Maybe formulas are easier to understand on the attachment.

All these come from Idronics 16, with additional examples.



500 X 1.5 X