Dew Point Control

[rknauer]

I have a residential application in Houston where the humidity is oppressive during the warm months (April - November). Dew points are almost always in excess of 70F, and 75F is typical during the daytime.

I find I have to manually adjust my conventional thermostat throughout the day to get the best balance between temperature and humidity. As the evening approaches and the temperature outside drops, I have to raise the setpoint or otherwise it is too cold. But the next morning I have to lower the setpoint to lower the humidity.

I have an electronic thermostat which has provisions for 4 temperature setpoints during the day. I have experimented with different values in an attempt to hit the right combination. But I am not able to find the right combination because the humidity varies from day to day at any given time. I am shooting at a moving target.

What I need is a dual setpoint thermostat, one channel for temperature and the other for humidity. If the temp gets too high, the temperature channel turns on the A/C to cool things off. If the humidity gets too high, the other channel turns the A/C on to dry things out. If both the temperature and the humidity are in the comfort zone, then the thermostat turns the A/C unit off.

Is there such a thermostat, priced for consumer use?

Thanks,

Bob Knauer

 

[quark]

Why can't you use a humidistat? Check Johnson Controls and Honeywell etc.

Regards,


Believe it or not : There is a three dimensional figure but only with one face, called Mobius Strip.

 

[rknauer]

>Why can't you use a humidistat?

A humidistat is designed to be used with a separate humidifier/dehumidifier. Do you have a way to use it in conjunction with an ordinary thermostat to control the A/C unit?

Would it be possible to hook up the A/C control wires in parallel with the contacts of both a conventional thermostat and a humidistat at the same time. That way if either the theromostat or the humidistat closes, the A/C would go on. Of course if the humidistat's SPST switch is designed to close when the humidity gets too low, you would be out of luck.

Actually what I am looking for is a single unit that combines both temperature and humidity control together. It would have the capability to function as a dehumidifier control so that when the humidity got too high, it would turn on the A/C separate from the temperature control. It would also have the capability to function as a conventional thermostat so that when the temperature got too high, it would turn on the A/C separate from the humidity control. There would be no need for humidity control on the furnace side of the thermostate, although I see no reason why it could not provide control of a humidifier for people in dry regions.

I did visit the Honeywell and Johnson websites but was unable to find anything like this.

 

[rknauer]

After I posted the above reply I came across what I was looking for. It is called a "thermidistat" and is sold by either Carrier or Bryant (identical units). Prices were all over the map so I will have to look carefully for the best deal.

 

[quark]

First of all, your frist post was not very clear. When you said 'If the temp gets too high, the temperature channel turns on the A/C to cool things off. If the humidity gets too high, the other channel turns the A/C on to dry things out' I understood that you may be having two air conditioning units'.

What I generally practise is to cool the air below the DP temperature and if reheating doesn't take place psychrometrically, add a reheat coil. This will always take care of required temperature and humidity in the controlled space. In this case cooling is taken care by thermostat and reheating by humidistat.

Also from your second paragraph, As the evening approaches and the temperature outside drops, I have to raise the setpoint or otherwise it is too cold. But the next morning I have to lower the setpoint to lower the humidity. If the outside temperature is lower than the required temperature in the controlled space, you do require reheating. (if not, you don't require to lower the setpoint)

If you got a solution already, then good luck. This forum is just for sharing our experience and ideas.

Regards,





Believe it or not : There is a three dimensional figure but only with one face, called Mobius Strip.

 

[rknauer]

I want to be able to use only existing equipment for now. I think the "thermidistat" might be the solution, albeit limited in comparison to a profession installation that deals explicitly with humidity control, as you suggested. My goal is to get out of having to manually adjust the thermostat day and night to control comfort.

I did notice something since I last posted - the Carrier/Bryant Thermidistat has a "lower humidity by cooling" function, which is what I want, and it limits the temperature drop to 3F, which is something I also want. I would have preferred that the temperature drop be programmable, but 3F seems reasonable. That's about how much I have to lower the temperature setpoint manually to achieve adequate dehumidification on extreme days.

Here's something I will pass on to the forum FWIW. We bought our house 7 years ago and in the beginning the A/C plant worked adequately. Last season and this one, however, were not handled adequately - the house was always a bit too warm during those days when the outside temperature exceeded 95F. I had an A/C contractor look at my system and other than topping off the freon, he could find nothing wrong.

So I started asking myself what changed over those 5-7 years. There was only one significant change - we all got computers. I had my computer when we moved in but my wife and my two children did not. Over the years they each got a computer, so now we have 4 (and sometimes I run 2 at a time, so it can be as many as 5).

On an especially hot day I asked everyone to turn off their machines. You would think I was asking to them to stop breathing. But I got them to do it and about an hour later the interior started cooling off like it did 7 years ago. It's those bloody computers that have added enough heat load to cause my A/C plant to be a bit on the inadequate side. But this should not be surprising. After all, each machine has a 350 watt power supply and a large display unit, both of which consume 500 watts together at a minimum. All 4 units consume at least 2 kW, and that heat has to go somewhere.

I once measured the load caused by the A/C unit. I turned everything electrical off so the meter was completely stopped. I recorded the readings and the time and then turned the A/C unit on. I ran it for a while, then shut it off and recorded the readings and the time. From that I could come up with what I believe is a reasonable estimate of power consumption for the A/C unit. I got about 2.2 kW.

If these figures are anywhat close to being realistic, then the A/C unit is consuming about as much power as those 4 computers. If the A/C unit is 100% efficient (which is unrealistic) then all the power it consumes goes into removing heat from those 4 computers alone. No wonder I can't cool the place - all my cooling capacity is being used to cool off computers.

The upshot is that HVAC contractors are going to have to make provisions for computers when they design cooling plants for residences. I do not know what an extra 2 kW does to the specifications of an A/C unit, but I suspect it is somewhat significant.

Thanks for your help - your clue about "humidistats" gave me a seach keyword that eventually led me to "thermidistat".

 

[Cooky]

If your AC system is absorbing 2.2 kW of electrical power and assuming it is correctly maintained and in good working order, it will produce about 6.6 kW of cooling.

As a rule of thumb, for each kW of power input, an AC system will develop 3 kW of cooling (or heating).

Regards,

Cooky

 

[rknauer]

That's remarkable, particularly in light of the conservation of energy: 1 joule of energy per second input and 3 joules of heat per second output. It appears that the First Law of Thermodynamics is being violated.

Or am I missing something here, like the actual meaning of the expression " 3 kW of cooling". For example, does it mean that the heat flow at the condenser coils is 3 kW? Is that how much heat (measured in kjoules) is being discharged at those coils per second?

NB: 1 watt of power is 1 joule of energy/heat per 1 second of time.

I wonder if it would be possible to make a perpetual motion machine out of such a device. That is, could some of that 3 kW of cooling power be used to operate the device at a rate of 1 kW energy consumption?

Please explain to this erstwhile physics major.

 

[Cooky]

A clue to what's happening is given in the name of the process, the refrigeration system or 'heat pump' is a closed loop system. The compressor moves the refrigerant around the system, but the refrigerant is carrying the energy from the evaporator to the condenser. The system works by lowering the pressure of the liquid refrigerant in the evaporator, causing it to absorb energy from it's surroundings, alowing it to boil-off. This energy is 'forced out' of the refrigerant by condensing the hot gas back into a liquid.

If you can't follow my very simplified description, any good heat transfer book will explain the process in a little more detail.

Regards,

Cooky

 

[ChasBean1]

The COP represents two energy-in terms. The heat energy absorbed into the refrigerant at the evaporator is about three times the work energy in by the compressor. The COP neglects the heat rejected term, which is the latent and sensible heat lost by the refrigerant in the condenser. The first law is conserved for the cycle by considering the rejected heat.

As far as dehumidification, this is often a result of oversized DX systems; but it sounds like this is less than a two-ton system, which might be good for a 1,500 to 2,000 square foot house. Do you have any additional house and AC unit size data? Regarding the dehumidification concept, there are issues in trying to do both with one unit, in that dropping the dry bulb temperature raises the relative humidity with the same moisture content in the air. I’m a little skeptical about cooling-only systems that claim to do both without the use of reheat…

 

[rknauer]

ChasBean1 wrote:

The COP represents two energy-in terms. The heat energy absorbed into the refrigerant at the evaporator is about three times the work energy in by the compressor. The COP neglects the heat rejected term, which is the latent and sensible heat lost by the refrigerant in the condenser. The first law is conserved for the cycle by considering the rejected heat.

---

If I understand what you just said (it's been 35 years since I studied Thermodynamics and it was not one of my favorite subjects so I managed to forget most of what I learned), this typical system inputs 1 unit of mechanical energy and 3 units of heat energy at the evaporator coils. Therefore it must reject 4 units of heat energy at the condensor coils. Is that a reasonable estimation of what happens?

If so, then I can burn a 300 watt light bulb in the air duct and it will only take 100 watts of mechanical (electrical) power to remove the heat created by the light bulb, and will discharge 400 watts of heat at the condensor.

Somehow I have this feeling I am leaving something out of the equation, like an enthalpy or entropy term. The only reason I want to know what is going on is to be able to estimate how much it is costing me in terms of cooling power to remove the heat from the computers. If the 3:1 rule is valid, then it takes about 700 watts to remove the 2000 watts created by the 4 computers.

+++++

ChasBean1 wrote:

As far as dehumidification, this is often a result of oversized DX systems; but it sounds like this is less than a two-ton system, which might be good for a 1,500 to 2,000 square foot house. Do you have any additional house and AC unit size data?

---

I was told that the unit was 5 tons. The house is single story "patio ranch" style with lots of full-length sliding glass doors. I have thermal film on most of them and we keep the drapes closed during the heat of the day. The house is listed as 1825 sq. ft.

+++++

ChasBean1 wrote:

Regarding the dehumidification concept, there are issues in trying to do both with one unit, in that dropping the dry bulb temperature raises the relative humidity with the same moisture content in the air. I’m a little skeptical about cooling-only systems that claim to do both without the use of reheat…

---

As I said, the A/C plant works satisfactorily when we shut off all the computers. But that is not possible in our household, so I have to deal with the extra heat load - and one consequence of that is that I have to tweak the thermostat day and night to contol the humitidy. It is for that reason I want a "thermidistat" even if it is not the ideal solution.

 

[lilliput1]

If humidity is your problem, not °F db temperature control, I would suspect your AC unit as being oversized. The compressor do not run long enough to dehumidify the air. This is assuming you have a reasonably tight house and do not get a lot of outdoor air infiltration load. Is your house wee insulated? Did you add more insulation lately?

 

[DouginMB]

I agree that the unit is oversized. You're looking at 365 sqft./ton which is low for a residential unit. If the system was not large enough, it would run day and night and you would still be uncomfortable. There would be no way to "tweak" the system as it would always be trying to bring the house down to setpoint.

Did you verfiy that your air handler fan is operating at low speed? This will help the system to run longer and improve humdity control. I still don't get how it operated adequately prior to adding computers. Computers will add maybe 40% of their nameplate value in heat (and thats when thier in use). It shouldn't make that much of an impact. Besides, the computer heat load would make the A/C run longer and subseqeuntly lower humidity in the space.

 

[rknauer]

I have two separate problems that are related.

1. Computers cause me to have to raise the setpoint because if I don't the A/C unit will run all the time on days when it is above 95F outside. Even when I run the A/C unit at a higher setpoint the humidity is marginally comfortable. The unit cycles on 80% and off 20% at setpoint 81-82F. If I shut off the computers then I can lower the setpoint to 79-80F and the unit will still cycle.

2. Because I have to set the setpoint to 81-82F when conditions are as described above, it gets too humid at night when it cools off outside. When that happens the setpoint is too high to cause the unit to operate at all. Because the unit does not run at all, humidity builds up and so I have to lower the setpoint to 79-80 to get the unit to come on and lower the humidity.

When the computers are not running, I can set the temperature to 79-80 and the unit will cycle 80-20 on a very hot day. But with the computers on, I have to raise the setpoint to get 80-20 cycle. If I left the setpoint at 79-80, the unit would not shut off at all, IOW it would cycle 100-0, which I do not want.

That's why I have to manually adjust the setpoint day and night, which is what I am trying to get around by using a thermidistat. It can be set for daytime cooling at 81-82F and when it comes night it will turn on the unit based on an increase in humidity with the same setpoint. Because it will not allow the temperature to fall more than 3F, it's as if I lowered the setpoint manually.

As far as the size of the unit is concerned, when I do not run the computers the unit is just adequate on a very hot day. Although the official airport temperatures are the ones people look at, they are often inaccurate for residential comparison. IAH (Bush Intercontinental Airport) is north of the city in the piney woods of East Texas, where it is typically 10 degrees cooler than in the city. I live on the outskirts of the city so my outside temperatures run about 5F above the official temperatures. That figure agrees with local school-based weather stations and my own crude backyard thermometer.

That means that the typical hot summer day in Houston, which officially registers 90-100F, is more like 100-105F in my yard. It takes 5 tons to cool this house off when computers are not on. When they are on, I have to raise the setpoint or else the unit will run continuously.

On a very warm day, when the official temperature is 90F, it will get to 95F in my yard and the unit can handle the heat including the 4 computers. When the temperature in my yard drops to 90F, the unit runs 50-50 with the computers on.

 

[chicopee]

You may be miserable between April and November but I bet you the rest of the year must be great.

 

[quark]

rknauer!

Some points to note before constructing a PMM1 (PMM of first kind).

1. Electricity is a high grade energy and heat a low grade energy.

2. The COP of an ideal Carnot engine is about 6. i.e every 1 unit of work input results in heat removal of 6 units.

3. One kW of electricity will give heat output of 1 kw by a resistance heater. But when used in a closed loop to run a heat pump it is 5 times high (thanks to Mr. Joule, and this basic derivation is available with basic thermodynamics)

4. 3:1 rule is a good approximation. (A DX unit has better COP, around 4.5)

5. Maximum heat load is contributed to the controlled space from moisture and fenestration. (otherwise, you wouldn't have required an AC when you shutoff the computers. A ceiling fan may work well)

Coming to your problem, it will be good if we are provided with typical data of DBT and WBT day and night. During day time, when the ambient DB is high, RH will be generally low(if you are away from a large water body) and you have to mostly take care of sensible heat (well above the dew point). During nights when the ambient DBT is low, RH will be high and to reduce this you should cool the air near to DPT. So DBT of controlled space also comes down and you may require reheat. In this case if you start your computers, you are more comfortable.

I still doubt how you manage with a two point controller when you have to control only one parameter.

Regards,


Believe it or not : It is believed that Archimedis single handedly sailed the 4000 ton ship Syracusia from the dock when dock workers failed and quoted his famous saying.

 

[rknauer]

chicopee wrote:

You may be miserable between April and November but I bet you the rest of the year must be great.

---

It all depends on ENSO (El Nino Southern Oscillation) and the subtropical jet stream. Some "winters" (late Nov to early April) are great, whereas some are cold, wet and cloudy.

In any event, "winter" is the time to get out hunting and camping. There are more Texans, men and women alike, with a gun in their hand on opening day of deer season than all the armed forces and reserves of the US military plus all the federal law enforcement agencies combined.

 

[rknauer]

quark wrote:

>Some points to note before constructing a PMM1 (PMM of first kind).

You are using unfamiliar jargon. What is a PMM?

>(A DX unit has better COP, around 4.5)

DX? COP?

>it will be good if we are provided with typical data of DBT and WBT day and night.

I can tell you DBT. It is around 81-82 day, 79-80 night. Those are the setpoints I use and the A/C unit manages to meet them even on a very hot day with all computers running.

>During day time, when the ambient DB is high, RH will be generally low (if you are away from a large water body)

Houston has a subtropical climate during the summer months (late April to early November). The entire region is one big water body. DPs run typically 75F during the day and 70F at night. Per capita consumption of electricity for cooling is the highest in the world. BTW, the Gulf of Mexico is only 25-50 miles away and the sea breeze is very active, carrying in abundant moisture.

>I still doubt how you manage with a two point controller when you have to control only one parameter.

I do not have a two point controller - I am only considering one. I want to be able to avoid having to manually reset the setpoint to control the RH. My thermostat even has provision for 4 programmable setpoints each day but try as I might I cannot get a combination that works reliably because the RH is a moving target.

If, for example, the RH is high at night, I need to lower the setpoint to 79F. But if the RH is not that high, 79F will cool the house too much, so I have to use 80F. You might think I am nitpicking over 1F, but that is enough to cause the unit to run excessively and chill the place too much at times.

At over $300 per month for electricity during the summer months (for a house that has 1825 sq. ft. and gas appliances), you can see why I am such a control freak when it comes to A/C costs.

 

[quark]

PMM1 - Perpetual Motion Machine of first kind (which violates first law of thermodynamics)

COP - Coefficient of Performance = (Heat removed from the controlled space/work input)

DX Unit - Direct expansion unit (a generally misused term for the cooling units, where you don't have any secondary refrigerant- like window and split ACs)

We are just showing here where you can go wrong in load estimation and control philosophy.

As you already worked on the problem manually, your failure chances will be less even if you go for simple automation.

Good luck and give us your experience on thermidistat when it gets to work.

Regards,


Believe it or not : It is believed that Archimedis single handedly sailed the 4000 ton ship Syracusia from the dock when dock workers failed and quoted his famous saying.

 

[rknauer]

Thanks for all the information. I will wait until next summer to implement a thermidistat and then only if the price comes down significantly - which will tell me the mass market has bought them. If that happens then local a/c contractors will have gained some experience and can tell me if the device will do what I want.