Heat loads

Jim Pompetti

on every job we do a heat load. But with this recent cold spell , we had a few problems. We had a house or two only get to 62 * F , both with condensing boilers, First one we were running the curve to top out at 160 * F , no problem till it hit 0* F( our design temp. is 15* ). We increased the top end to 180@ , but it was slow to respond . I'm sure all will be find in the future ,but the customer is a little upset . Any suggestion on talking to her.

Before we install the boiler , she have a 30 year old monster in the house.

Zman

More info

What type of radiation? Was the boiler reaching it's target? If the boiler is reaching it's160 target and you are using radiators, it sounds like you should increase the design day water temp.

If the boiler is not keeping up... That is another issue.

Is it a lack of radiation or boiler?

Carl

Robert O'Brien

Thermostat setback?

Huge problem with people not grasping the incompatibility of outdoor reset and room setback

Mark Eatherton

Infiltration wild card...

Might consider having a local company do a blower door test on the home and see if it matches your numbers…



ME

SWEI

Slow to respond

I have seen two causes of slow response when changing parameters on a cold day.  The first is simple inertia, which most installers should understand.  The second is the PID parameters in the boiler controls, which is a bit trickier to diagnose.  Even when the setpoint (whether ODR calculated or manually set) moves, the boiler can take 20-30 minutes ramping up to high fire.  While there are generally ways to tweak those coefficients in the controls, it's just too easy to cause major problems there.



Assuming the boiler has the BTUs, I think you're right -- the best answer is probably to change the maximum water temp and the outdoor design temp to extend the curve.  The old fashioned ratio controls would pretty much do the same thing -- up to the high limit setting.



Given that our heat loss calcs already have at least 25% headroom in them, this ought to work in pretty much any system.

Harvey Ramer

This is why boilers need to

become smarter. We should not have to program a reset curve. The boiler should have an outdoor sensor and an indoor sensor and using those reading the control logic should determine the firing rate and the BTU delivery rate to the system.



In any building, the internal mass needs to be maintained within a narrow temperature margin to achieve occupant comfort. That is why daily setbacks don't work good. In any kind of system, whether forced air or radiant, interior thermal mass has everything to do with comfort. There is always some manner of radiant heating and cooling going on in any type of system, that is directly related to the temperature of the interior thermal mass.



If the interior thermal mass temp drops to low due to setback or incorrect ODR, a correctly sized heating system will struggle to recover. Getting the air temperature in a space to a certain set point is not as relevant as a lot of people think.



It's all about the temperature of the mass surrounding you.





Harvey

Jamie Hall

I like Harvey's comments

and he is quite right; if we are to achieve significant increases in efficiency, it is necessary for the equipment to be smarter.



Ideally, there would be some direct way to measure the instantaneous heat loss rate of the structure.  We use the outdoor temperature as a surrogate for that, and do both our heat loss calculations based on information about the structure and the difference between what we hope will be the lowest outdoor temperature (design day) and the design indoor temperature and we also, on the more modern systems (and some older ones!) use the outdoor temperature to set the heat output of the boiler, when we have boilers which can modulate.  The problem is, of course, that the outdoor temperature is only part of the picture; depending on the structure, there may be appreciably less heat loss at some times than would be suggested (if there is significant solar or occupant gain) or considerably more (if there is considerable wind -- which is worse if there is infiltration, but which is a factor anyway).



So -- it would be nice to be able to directly measure heat loss rate.  Bright ideas are welcomed any time!



Then there is the other factor Harvey mentions: if you try to raise the inside temperature, that is going to take additional heat -- sometimes a remarkably large amount of it -- and if our system is running so as to exactly match the heat loss and we turn up the thermostat, nothing will happen -- there is no additional heat available to warm up the contents.  So that has to be factored in.



It is a very interesting problem in control theory -- which is a fascinating subject all by itself.

Mark Eatherton

Flux sensors...

http://www.omega.com/pptst/HFS-3_HFS-4.html



This would tell you exactly what the structures heat loss is, but which wall do you place it on?



We Americans have become so focused on air temperature, and the ability to manipulate it upward and lower, most American homeowners would NOT know what to do…



Witness what happens with a TRV. It has "Comfort Settings" between 1 and 5. Put it into an American home without explanation and you are guaranteed comfort complaints. No read out, no feed back, but if you set it and forget it, extremely high comfort.



Kudos to Harv for recognizing the fact that the MRT has much more meaning and influence on human comfort than air temperature alone has. Now if we can convince our industry, our customer, and our government, we will have it made…



We (The RPA) are working towards that end…



Join us in our efforts, won't you?



ME

SWEI

MRT sensors

I stumbled on the phrase "Black Bulb Thermostat" (of all places, in an RPA mention of radiant windows) and managed to track down a couple of sources.  I've ordered a few from the UK to test http://www.sontay.com/sites/default/files/productdownloads/TT-915_3.pdf



So far, outdoor reset combined with indoor feedback (or indoor proportional valves, e.g. TRVs) is about as good as we can get.