How low can indoor temp go

tonelow54

Residential Customer with 3 ton 13 seer split a/c system unit is properly charged and proper cfm
2 story home never is comfortable upstairs so runs thermostat at 65 degrees all the time unit occasionally
Freezes up how can this problem be fixed are air conditioners designed to cool this low ?

Jamie Hall

They can cool that low -- but it sounds as though despite the cfm being OK you have a distribution problem. First thing I'd look into is where are the returns and how are they ducted? Then look into the cold air distribution...

GW

I would say "you can't do that". 75 is the standard with residential ac. I have some experience with this topic, I'm pretty sure every hvac contractor does.

Sounds like they have humidity issues, and or their second fl is warmer than the 1st fl?

I bet it's freezing at night/morning

Is this a new system? New ducts? Is this a changeout? Did their "needs change", in other words.....when "were' they comfortable? Under what conditions were they comfortable before you entered their lives? You need to know those answers or you'll be an unhappy camper. You need to determine whether you're at fault or not

Increasing blower speed will help some if you're looking for more sensible and less latent.

Gary

ChrisJ

Wow, I'd never be able to live with 75F standard or not.
I keep my house at 70F, often 68F at night.

75F is too damn warm IMO.

At the same time, 65F seems a bit cold even for me. I guess my theory holds true, no matter how extreme you think you are at anything (good, bad, cold, hot etc), there's always someone out there that will go further.

GW

Ahh the beauty of what the energy code says vs what the HO says, if it were easy everyone would be doing it

JUGHNE

For every situation with more than one occupant there are two temperatures.......too hot and too cold.

Just had a change out that is 6 months old, cooling not good. Air bear filter plugged. Usually get nearly a year out of those. But not the first year because of disrupting old dirt in RA panning.
Cool froze up in this case, young couple want 72 degrees.

GreenGene

Check the unit specs, most ac units have a limit specified by the factory of what conditions they can run in, you say 2nd fl ng, they need to run the fan constant and I'd check the attic for insulation, check ductwork going to attic, if it's coming from the first floor it may be losing temp, super insulate and increase flow to 2nd fl if possible.

I just added 15" insulation on top of my rafters and I put foil faced polyisocyanurate over all flooring and it is 1:15pm at 85 outside and my ac still is not required.

The house is part of the system.

ratio

Cooler than 68 isn't air conditioning, it's refrigeration. Attempting to set the thermostat below 68 is indicative of some other problem, usually airflow/balance in my experience.

Harvey Ramer

Right on. Every manufacturer and trainer I've ever talked to says all bets are off if you run the return below 68 degrees.

Fred

I've always kept my temp at 68 degrees. No problem but I have two systems, one cools second and third floor and a smaller one cools first floor.

jumper

If not enough air moves across evap coil it frosts. How do you know you have correct cfm?

JUGHNE

Fred, with your 2 systems, especially with dedication to the upper floors, your systems probably get to cycle on and off with any coil frosting not building up.

For new T-stats I use the HW, even the 5000 series has programing for max and mini temp. The lowest cooling temp I program in might be 70 or 72. This gives the system some cycle off time and also try to set up the blower for long post cycle run on time. Or try to convince HO to let fan run continuously for AC season.

Some people see a digital number and it will control their desires. The upper end T-stats have a thermometer display that can be programmed to show an offset up to 5 degrees. ie. it could actually be 74 in the house but the T-stat tells them it is 69. Then they feel better and in control, the butter won't melt either.

ratio

Never had to dial in a delta on the indoor temp, but I have fixed a few problem areas with wireless stats...

I set all my stats up with 68 cooling/80 heating limits. Only had one complaint so far, & they were trying to get away without dehumidification in an exercise joint. @#%& injuhneers. The manager wanted to set the temp down to 66 to dry the place out. I don't think it works like that...

AMservices

Central air conditioning refrigeration is based around a 20°degree temperature drop. If you measure the air temp across the cooling coil (the return air temperature to the supply air temperature) you should see a 20° difference.
If your not getting 20° delta-T and its freezing up on you, with a clean air filter and coil, then I would say your under charged.
To be accurate about checking the delta-T, the temperature should be taken right at the equipment, in the return and supply plenum's.
If the humidity is high (around 80%) and it's 85°F in the return, you might see a 18-15 degree difference at first. As the humidity drops, so will the delta-T.
If you have a lot of return ducts work in the attic thats open or full of holes, you could be conditioning 120° air. That's why it's important to check temperature at the equipment, not in the space. Also if the return is full of holes, fix it.
Central air conditioning can't drop the indoor temperature more the 20° from the outdoor temperature. So if it's 90° outside and the thermostat is set to 68°, the system will never shuts off.
I can't say I've gotten a house down to 60° on a 80° day, but I've seen 64°.

jumper

>>If you have a lot of return ducts work in the attic thats open or full of holes,<<

More often I see inadequate return so that you can't get enough CFM no matter how hard you blow.

ratio

My Old Guy once told me, "they're called blowers, not suckers. There's a reason for that." Restrictions on the return are a lot harder to overcome than on the supply.

AMservices

An old guy I work with used to call them "Monica's"
It took me all of 2 seconds to figure out what he was talking about.

ratio

Heh. I'll have to remember that...

ChrisJ

Fans of all types perform better pushing, by far.
I think it's because as you drop the pressure, there's less "stuff" for the blades to push.

When you restrict a vacuum cleaner, it's current draw drops big time. I assume a blower on a HVAC system is similar.


Curious, do normal air conditioners slowly drop power consumption as the evaporator drops temperature or does it remain fairly constant assuming the ambient conditions don't change?

GW

I don't get how some duct systems are clearly lacking is 1st grade mathematics. If joe metal installed two 16x8 supply trunks and your returns are a 12x8 and a 16x8, and the ret drop is a 20x8.....I just don't get it.

We use 10x on all jobs. We don't do many but some 2 to 4 ton jobs here and there.

I don't get why people use 8x all the time. A round duct is best, a square duct is second best, a pancake is dog poo.

Basic rule of thumb don't exceed 50% of your duct height, but less is better.

10x is just a few bucks more.........

Gary

AMservices

8x fits better running alongside beams in basements. Every inch of head room makes a big difference with customers.
I find 8x duct to be readily available, with more fittings on the shelf. 10x duct and fittings are less popular, so I can't always find what I need in the event my original master plan needs to change.
12x10, 16x8, 6x22, or 12" round, it's all the same when talking 100' of straight duct.

As the installer, you have to decide what makes the most sense to fit the home.
Round pipe is better for resistance around bends. But if I can save space with square duct, I'm sizing everything anyway, might as well make it look nice and tuck it in spots that can have a soffit easily built to hide it.

jumper

Chokes and turns.....It's amazing AC works at all

GreenGene

It's clear what the problem is, the 2nd floor is way off, probably on cfm, ductwork, insulation, etc.

You need to do a new load calc and make sure the unit is sized right.

You need the total sq ft of the home and the sq ft of each room.

From the correct size of the unit figure 400cfm per ton and come up with the total cfm.

Then for each room

AREA OF ROOM DIVIDED BY THE TOTAL AREA X TOTAL CFM =
CFM FOR THIS ROOM

So for example if you had this unit as the right size and a 2000 sq ft house with a 12x14 master bedroom the numbers would look like this.

168 / 2000 x 1200 = 100.8

Then check your actual cfm for each room and see what you're really getting.

The other problem you're probably also having if the unit is in the basement is temp drop, see what your supply temp is on the 2nd floor compared to the first, my guess is you're losing big time, you might have 55 on the first and 67 on the second, you need a 20 degree drop.

good luck

phasers on stun

GW

Gene, east-south-west...overhangs on the glass, attic installation...where the ducts are located, aren't you missing a few items with your basic formula?

GreenGene

I said do a load calc. Then the rest

I said " You need to do a new load calc and make sure the unit is sized right."

THEN

You need the total sq ft of the home and the sq ft of each room.

THEN

From the correct size of the unit figure 400cfm per ton and come up with the total cfm.

Then for each room............

it's quick and pretty accurate for splitting the cfm, some load programs will do each room for you some just do the whole house

but with this, 2 story, 3 tons?.........could be under sized, I have a raised ranch so my lower level is half in, no real cooling load, I'm doing just the upstairs with 3 tons and I'm tight, a home that isn't well insulated could be 4-5 tons or more, you really have to do that first or you're just chasing ghosts

JUGHNE

Do we know where this problem house is located? CDD and humidity are big factors.

GreenGene

Seen a lot of houses like that, unit in the basement, ductwork run 32 miles to the attic and then down, it generally doesn't work.

Too much temp loss on the way, not to mention velocity. Many times it will work ok for heat but not cooling.