Building a portable AC for my car

[iduhfuse]

Hello everyone. This is a very simple project that I intend to build. It involves a cooler full of ice and water, two 120mm computer fans, a 240mm computer radiator and a 600 gallons per hour water pump. All these will run off the cars cigarette lighter rated at 12 Volts. I need your help to determine how to wire all of this up.

Pump 12v rated at 3 Amps
Fans 7-12V rated at 0.5 Amps

I'm guessing i need to wire this up in parallel? And if the pump is running too fast and splashing water everywhere can I add a resistor to slow it down? Let me know thanks!

 

[sophiecentaur]

If you have a good sized heat exchanger (say an old car heater matrix), then there will probably be enough circulation (thermo-syphon - look it up) if you just connect it directly to your ice tank (a big hose top to top and a big hose bottom to bottom. Your problem here will probably be that a useful AC system uses a lot of power (total energy to shift = power times time) - which corresponds to a lot of Ice to carry for a long journey or to cool down a car, initially. Be prepared to carry an awful lot (gallons and gallons) of iced water.
It could be an interesting experiment in proving that the numbers count in engineering.

 

[iduhfuse]

If you have a good sized heat exchanger (say an old car heater matrix), then there will probably be enough circulation (thermo-syphon - look it up) if you just connect it directly to your ice tank (a big hose top to top and a big hose bottom to bottom. Your problem here will probably be that a useful AC system uses a lot of power (total energy to shift = power times time) - which corresponds to a lot of Ice to carry for a long journey or to cool down a car, initially. Be prepared to carry an awful lot (gallons and gallons) of iced water.
It could be an interesting experiment in proving that the numbers count in engineering.

I don't think thermosiphoning is going to work for my application. This is meant to be done on the cheap <$100. And yes I only expect it last for an hour maximum. Those are my longest journeys in the car anyway. I need help wiring it as I am unsure how I should have it wired and if I need a fuse or not.

To give you an idea, the water pump will sit in the bottom of the cooler, I will cut out an opening in the lid for the radiator to be glued in and the fans will sit on top of the lid where the radiator is integrated. The cooler holds about 3 gallons.

 

[sophiecentaur]

I can't think of a cheaper system than a scrap radiator and its hoses and some tape with (cheap plastic) plumbing fittings top and bottom of your cooler tank. A pump involves another electric motor and another pair of waterproof joints.
I am wondering just how much heat transfer your computer radiator is designed to transfer with a temperature difference of 20C. Does it have adequate total area to make a difference to the volume of air in a car? I imagine there will be a spec somewhere to give you some idea. But remember, a computer probably dissipates around 100W, whilst the windows of a car will be admitting several hundred Watts and the initial heat content of the vehicle must be significant (i.e. for initial cooling down)
How many kW of power do you think a regular AC unit produces? (I just don't want you to be disappointed if the performance of your system is too low.

 

[sophiecentaur]

And YES! you must have a fuse, right next to where you take the power off. Changing that fuse could be a lot easier than getting under the bonnet (hood) to change the fuse in the car's electrics.

 

[iduhfuse]

I can't think of a cheaper system than a scrap radiator and its hoses and some tape with (cheap plastic) plumbing fittings top and bottom of your cooler tank. A pump involves another electric motor and another pair of waterproof joints.
I am wondering just how much heat transfer your computer radiator is designed to transfer with a temperature difference of 20C. Does it have adequate total area to make a difference to the volume of air in a car? I imagine there will be a spec somewhere to give you some idea. But remember, a computer probably dissipates around 100W, whilst the windows of a car will be admitting several hundred Watts and the initial heat content of the vehicle must be significant (i.e. for initial cooling down)
How many kW of power do you think a regular AC unit produces? (I just don't want you to be disappointed if the performance of your system is too low.

Yeah, I understand what you mean. The radiator is 11x6 inches. The two fans will cover than entire area. And the water pump is submersible. Literally just going to take vinyl tubing and hose clamp. It's not like its blasting 2000psi through the lines. It'll work. But say the car is 90F and id like it to be near 75F in the car. Cooler can only hold 3 gallons of ice water. This radiator that was in my computer on full load on the cpu was about 45F. While the regular heatsink and fan was near 65F on full load. Also the car is very small. A vw golf.

 

[sophiecentaur]

Yeah, I understand what you mean. The radiator is 11x6 inches. The two fans will cover than entire area. And the water pump is submersible. Literally just going to take vinyl tubing and hose clamp. It's not like its blasting 2000psi through the lines. It'll work. But say the car is 90F and id like it to be near 75F in the car. Cooler can only hold 3 gallons of ice water. This radiator that was in my computer on full load on the cpu was about 45F. While the regular heatsink and fan was near 65F on full load.

You have clearly given this some thought, which is good. However, it is the Heat (Joules) and not temperature that counts when you want to cool something.

Where did you plan to put the radiator? A pump would give you more flexibility in the position of the Cold Sink than a thermo-syphon.

 

[iduhfuse]

You have clearly given this some thought, which is good. However, it is the Heat (Joules) and not temperature that counts when you want to cool something.

Where did you plan to put the radiator? A pump would give you more flexibility in the position of the Cold Sink than a thermo-syphon.

See the attached picture. That's exactly what I'm imagining. Minus the battery because those pos and neg will be connected to the 12v adapter for the cigarette. My cooler is a bit larger than that one. And oops correction: the rad on the cpu was 45C and heatsink with fan was 65C

 

[iduhfuse]

You have clearly given this some thought, which is good. However, it is the Heat (Joules) and not temperature that counts when you want to cool something.

Where did you plan to put the radiator? A pump would give you more flexibility in the position of the Cold Sink than a thermo-syphon.

Just bought an 8A fuse, that should be sufficient as I don't see the system going over 4 amps. Plus car fuse on the 12v line is 10A or 15A i think.

 

[rbelli1]

This is all well and good but a typical cooing system in a car will be the equivalent of melting several thousand pounds of ice per hour. Your cooler will only hold several tens of pounds of ice. Your order of magnitude is off by 2.

You have to realize you are driving around an uninsulated greenhouse. The typical calculations for room size assume that you have a well insulated space with only a small percent of windows.

BoB

 

[billy_joule]

This is all well and good but a typical cooing system in a car will be the equivalent of melting several thousand pounds of ice per hour. Your cooler will only hold several tens of pounds of ice. Your order of magnitude is off by 2.

I agree. For similar time and cost to this ice contraption you could probably find a complete AC unit at an auto dismantler and bolt it straight into your car.

 

[sophiecentaur]

I agree. For similar time and cost to this ice contraption you could probably find a complete AC unit at an auto dismantler and bolt it straight into your car.

Hmmm. That's a bit of an overstatement, isn't it? Where would you fit the front heat exchanger without getting all new radiator fittings and is there a suitable fixing bracket for the pump?
However, I do agree that the project needs many more sums done on it than the OP has done. Auto AC involves a lot of Power (from the engine) and there is loads to spare (you pay in mpg but no one cares about that). As BoB says, you could be out by two orders of magnitude. These sort of projects do not work on the strength of enthusiasm but on hard Engineering figures.
PS, if your ambient humidity is low, a system based on a Swamp Cooler could be worth considering. Blowing air over a wet sponge will utilise the Latent Heat of Vaporisation, which is much greater than the Latent Heat of Melting.

 

[billy_joule]

Hmmm. That's a bit of an overstatement, isn't it? Where would you fit the front heat exchanger without getting all new radiator fittings and is there a suitable fixing bracket for the pump?

If a higher spec model came with AC then lower spec models generally will have all the mounts and space etc required, down to the unused plugs on the loom and the blank in the dash where the ac button fits, every thing just bolts straight on. At least that's my experience on a 97 corolla.

 

[iduhfuse]

Well, I suppose you're all correct. I have limited knowledge of thermodynamics which is why I consulted you guys haha. Retro fitting an AC in my car is way too much work and expensive $$$. I've already looked into it so that's a no no. I appreciate all of your responses. I'm going to proceed with this project, purely out of curiosity now as it mostly likely will not meet my expectations anyway but i will let you know how it works in the next coming days!

 

[NascentOxygen]

As a compromise, maybe water-cool a cushioning backrest on the driver's seat?

 

[jim hardy]

I'm going to proceed with this project,

I like your spirit
keep us posted

I have limited knowledge of thermodynamics which is why I consulted you guys

Well...

a car air conditioner is probably between 10,000 and 20,000 BTU's per hour
144 BTU's will melt one pound of ice

so you can estimate how many pounds of ice you'd need to provide that much cooling for the duration of the trip you have in mind.

 

[billy_joule]

Retro fitting an AC in my car is way too much work and expensive $$$. I've already looked into it so that's a no no.

Did you look at self help auto wreckers? I got an entire AC unit, every thing required, for 76 NZD, less what the parts listed in the OP would cost me.

 

[OmCheeto]

Did you look at self help auto wreckers? I got an entire AC unit, every thing required, for 76 NZD, less what the parts listed in the OP would cost me.

From the comments, it sounds like iduhfuse is my next door neighbor, and needs A/C about once a year.
That's generally how often I need it.
It strikes me as a bit silly, to spend hundreds of dollars, and countless hours, on something, that you need, once a year.

Personally, I've traditionally gone for the most scientifically & economically logical solution.

Though, I've recently been investing in all of the hardware iduhfuse seems to have collected, and am interested in the outcome of the experiment.

I'm guessing i need to wire this up in parallel?

Yes.

 

[Tom.G]

If you can direct the cool fan exhaust directly on the human cargo, it could be almost practical (for short trips).

 

[OmCheeto]

...
It could be an interesting experiment in proving that the numbers count in engineering.

And for that reason, I will be building one this afternoon.
Weather permitting, of course. The morning clouds are predicted to burn off my 2 pm.
And my vehicle is black! This should be a good test.

I've also done some of the preliminary thermodynamic calculations.
3 gallons of ice water raised to 65°F will cool 896 m^3 of air from 90°F to 75°F
896 m^3 is equivalent to 338 interior volumes of a 2015 VW Golf
Unknowns:

• thermal heat capacity of the vehicle interior
• thermal transfer rate into the vehicle
• interior volume of my truck (2009 Ford Ranger, standard cab. Just in case anyone knows this number off the top of their head.)
• whether or not the little radiator from the junk yard leaks or not
• other things I'm probably forgetting

ps. I don't think I'll be needing it this year, as my only remaining scheduled trip of 60 minutes or more is in 4 days, and the forecast calls rain. But it's a miserable drive when it's hot out, and I've got all this junk lying around to make it. So my total cost will be about 2 hours, 2 bags of ice, and some water to top off the cooler.

 

[jim hardy]

I've also done some of the preliminary thermodynamic calculations.
3 gallons of ice water raised to 65°F will cool 896 m^3 of air from 90°F to 75°F
896 m^3 is equivalent to 338 interior volumes of a 2015 VW Golf

Good work !
Pay attention to the weather lady to stay aware of the dewpoint. Air Conditioners work hard to condense moisture out of the air.
But just lowering the dewpoint makes it more comfortable.
I'm old enough to remember when Pullman cars on the train were cooled by huge blocks of ice in overhead trays.

 

[OmCheeto]

Good work !
Pay attention to the weather lady to stay aware of the dewpoint.

Found it!
[formerly] Currently:

Humidity: 74%
Dew Point: 53°F​

Doh! Spent 30 minutes looking for pictures of my dad's truck. (see below)
Currently:

Humidity: 80%
Dew Point: 55°F​

Air Conditioners work hard to condense moisture out of the air.
But just lowering the dewpoint makes it more comfortable.

That would be one of the things I've forgotten.
I'll have to review the "Instantaneous Clothes Dryer" thread to see how that works again.
I always get my "latent heats" backwards.

I'm old enough to remember when Pullman cars on the train were cooled by huge blocks of ice in overhead trays.

ps. On another old person side note, my dad had a wood burning stove in his 50's panel truck when we lived in Alaska, back in the 60's.
Someone should start a new thread: "Back when I was your age..."

 

[jim hardy]

Dewpoint 55 sounds delicious

we're 77 right now

 

[OmCheeto]

...
Weather permitting, of course. The morning clouds are predicted to burn off my 2 pm.
...

And...
Maybe next year...

 

[iduhfuse]

From the comments, it sounds like iduhfuse is my next door neighbor, and needs A/C about once a year.
That's generally how often I need it.
It strikes me as a bit silly, to spend hundreds of dollars, and countless hours, on something, that you need, once a year.

Personally, I've traditionally gone for the most scientifically & economically logical solution.

Though, I've recently been investing in all of the hardware iduhfuse seems to have collected, and am interested in the outcome of the experiment.
Yes.

Alright you have me, I don't "NEED" AC every day as it is a luxury but it sure is nice to have. I do use it ever day during the summer where I live in both my home and previous vehicles I've owned. However, still waiting for my parts to come in but will definitely keep this thread updated.

 

[OmCheeto]

Alright you have me, I don't "NEED" AC every day as it is a luxury but it sure is nice to have. I do use it ever day during the summer where I live in both my home and previous vehicles I've owned. However, still waiting for my parts to come in but will definitely keep this thread updated.

hmmm... "every day during the summer"?
In that case, I would ignore everything I've said, and listen to everyone else.
A big ol' cooler full of ice water, sloshing around for an entire summer, is just going to ruin you car.

ps. I have a truck, and can put my 12.4 gallon cooler back in the bed, and have zero moisture intrusion. And like I said, I only need it one day out of the year.
pps. My local forecast indicates that I will not be able to perform the experiment for at least another 10 days.

 

[NascentOxygen]

For DIY types, I wonder how practicable it would be, for just a handful of times per year, to pre-cool your car's interior before leaving home using a external mains-powered A/C unit? Perhaps filling some empty passenger space with thermal mass would assist here, to keep the interior cool for a trip of, say, 30 mins (and this needn't rule out carrying some drink containers of frozen water, too).

Being mains-powered, energy costs would be economical, but a big attraction is that you'd avoid entirely the power-sapping drain when a car's aircon is switched in (especially undesirable when driving a 4-cylinder car on winding or hilly roads, where there is just no power to spare). It might take an hour or so, running a split-system A/C to sufficiently chill the interior mass of the car.

Okay, I've glossed over the return trip...

and haven't mentioned window shades and reflective coatings.

 

[iduhfuse]

For DIY types, I wonder how practicable it would be, for just a handful of times per year, to pre-cool your car's interior before leaving home using a external mains-powered A/C unit? Perhaps filling some empty passenger space with thermal mass would assist here, to keep the interior cool for a trip of, say, 30 mins (and this needn't rule out carrying some drink containers of frozen water, too).

Being mains-powered, energy costs would be economical, but a big attraction is that you'd avoid entirely the power-sapping drain when a car's aircon is switched in (especially undesirable when driving a 4-cylinder car on winding or hilly roads, where there is just no power to spare). It might take an hour or so, running a split-system A/C to sufficiently chill the interior mass of the car.

Okay, I've glossed over the return trip...

and haven't mentioned window shades and reflective coatings.

My garage is pretty much AC'd but within 10 minutes of driving the car is baking like an oven so yeah it's pointless. The car had zero tints before 2 weeks ago when I got it tinted 35% all around. I definitely noticed a difference but it still is pretty hot. AC only saps like 5hp nowadays. My car is only 110hp tiny 2.0 that I have no problem revving to the moon if I need power. Anyhow, I am excited, all the parts are in except for the radiator. Pics shall soon come!

 

[arydberg]

And for that reason, I will be building one this afternoon.
Weather permitting, of course. The morning clouds are predicted to burn off my 2 pm.
And my vehicle is black! This should be a good test.

I've also done some of the preliminary thermodynamic calculations.
3 gallons of ice water raised to 65°F will cool 896 m^3 of air from 90°F to 75°F
896 m^3 is equivalent to 338 interior volumes of a 2015 VW Golf
Unknowns:

• thermal heat capacity of the vehicle interior
• thermal transfer rate into the vehicle
• interior volume of my truck (2009 Ford Ranger, standard cab. Just in case anyone knows this number off the top of their head.)
• whether or not the little radiator from the junk yard leaks or not
• other things I'm probably forgetting

ps. I don't think I'll be needing it this year, as my only remaining scheduled trip of 60 minutes or more is in 4 days, and the forecast calls rain. But it's a miserable drive when it's hot out, and I've got all this junk lying around to make it. So my total cost will be about 2 hours, 2 bags of ice, and some water to top off the cooler.

1 calorie is required to rase 1 cc on water 1 degree centergrade but 80 calories is requirde to melt 1 cc of ice to water at 0 degrees. My point is if you want lots of cold use ice, not cold water. Also if you want it go for it. Be true to yourself. There should be no problem in running everything off the cigarette lighter butif you use wires from the battery position fuses close to the battery.

 

[sophiecentaur]

Dewpoint 55 sounds delicious

we're 77 right now

HAHA. I first read that as 77 years of age and the 'Royal We'.

 

[OmCheeto]

The sun decided to burst out early today.
Currently collecting preliminary data.

 

[sophiecentaur]

There is another significant factor which I don't think has been dealt with here. That is the flow of air through the car. It's the main factor in Central Heating calculations and it is very important here. If you want the car not to be stuffy with recirculation air, you have to cool down all the warm fresh air coming in.
But we can really ignore the basics of a particular invention. The bottom line is to compare the heat transferred with a normal car AC with the 'negative heat' available in a few kg of ice. A car AC (Wiki figure) uses around 3kW of mechanical energy from the engine. A 1kW source of heat will melt about 10kg of ice in an hour. If you had 10kg of ice and could actually bring it into good enough thermal contact with the air in the car, you could have an effect almost comparable with a regular AC if you could actually melt it in one hour. That would have to involve some pretty clever heat circulation system which would be based on crushed ice.

 

[jim hardy]

uses around 3kW of mechanical energy from the engine

Just for estimating purposes...

3kw into a machine with EER of 8, same as a low priced room airconditioner, would be 24,000 btu/hr, two tons of refrigeration,
a ton of refrigeration being able to make a ton of ice in a day
which is 83.3 pounds in an hour

3kw making two tons of cooling should be equivalent to melting 167 pounds of ice per hour ?

check my arithmetic ?

 

[Planobilly]

I would run the wire direct to the battery with a inline fuse and a switch, The lighter is not the most well made device on a car and prone to failure to begin with.

With no engineering calculations at all, I can assure you that air being passed through a cold heat exchange unit blowing on your neck will be cooler than warm ambient conditions...lol

Everything you are talking about doing will work. How well? Who knows. What are the weather conditions where you live. Here in Florida at this time of the year would take a lot of ice to have much effect in this type of arrangement ..lol

Yes, fresh air is good but not a requirement.

If after you try all this and it does not work well enough for you. there are other less complex ways to deal with the issue.

I assume you only want to cool down people. Frozen bottles of water wrapped in cloth work fairly well. I do this some times on my boat which has no AC, and it is hot as hell here in Florida, so I know it works. How well does it work? A lot better than nothing, that is for sure.

Ask an engineer to build you a parking lot and leave him along and when you get back there will more than likely be a twelve story building there...lol Engineers are truly wonderful people, but most of them were asleep during economics class...lol

Have fun and go try your ideas and see how well they work for you.

Cheers,

Bily

 

[sophiecentaur]

there are other less complex ways to deal with the issue.

That is right without a doubt. The requirement was not stated at the beginning and any detectable change in the cab could be thought of as a success by the guy doing the project. But I wouldn't say that the possible outcome should be considered as "Air Conditioning as we know it".

check my arithmetic

I would do, if the problem had been stated in SI units. (I finished with Imperial Units, ten years after we did away with our Empire (lol) when I left school in the late 60s and they tend to bring me out in spots.) My calculation that 10kg of ice could produce a mild cooling effect over an hour seems to tie in, near enough with your answer.

 

[sophiecentaur]

The best value of cooling effect from ice would probably involve local cooling of the driver's body with partially insulated ice packs inside the driver's clothing.

 

[OmCheeto]

The sun decided to burst out early today.
Currently collecting preliminary data.

The data does not look good.

Thank god I used duct tape to put everything together.

 

[rbelli1]

two tons of refrigeration,
a ton of refrigeration being able to make a ton of ice in a day
which is 83.3 pounds in an hour

2 tons == 4000 pounds
4000 / 24 is about 167 pounds per hour.

about what I calculated after you correct tons per hour to tons per day in my estimate in post #10

If you put a heat exchanger in a block of ice and another directly on the occupant you should get some relief.

https://www.zoro.com/igloo-full-siz...TY3_FlLUZF_qXiP-SAVCVBoCBR3w_wcB&gclsrc=aw.ds

Fill that solid with ice and somehow melt it in 2 hours and you have the equivalent of a regular car AC unit.

Moving a 300+ pound cooler into your car is left as an exercise for the reader.

BoB

 

[iduhfuse]

Update: Canada Post is going on strike and therefore my rad is being delayed. Son of a gun! It might be a while until you hear from me next but don't worry. There will be pics and all. Also, cig lighter is 15amps. That's easily enough to power my pump 3A + 2, 0.5A fans. Total 4 amp draw lol.

 

[OmCheeto]

Update: Canada Post is going on strike and therefore my rad is being delayed. Son of a gun! It might be a while until you hear from me next but don't worry. There will be pics and all. Also, cig lighter is 15amps. That's easily enough to power my pump 3A + 2, 0.5A fans. Total 4 amp draw lol.

It's been my rule to never to pay more than $20 for equipment for an experiment, unless I've done the maths first.

My electrical power draw yesterday was 24.5 watts. (solar powered, measured with separate volt and amp meters)

one water pump: 0.194 kg/sec
one air pump: 0.0185 kg/sec​

My system thermal power was 67.2 watts (based on the temperature rise of the ice/water bath)

My guess is that the water flow rate was not the problem.

Still only into this problem for about $2.

 

[iduhfuse]

Update: all parts are in! Going to build this thing tonight! Will post pics of building process and final product. Testing will be performed tomorrow. Stay tuned

 

[iduhfuse]

Assembly and soldering went well. Ran out of time to finish cutting the lid properly and have to run out to get longer screws for the fans and rad. Wired it all up in parallel and did a test run. Runs perfect, but haven't tested water pump yet. Will finish tomorrow and post final build!

 

[Planobilly]

This is cool stuff! No pun intended...lol As you were describing all this, the photos you posted are just what I had in envisioned.

I also like to build stuff. Sometimes stuff I build works well and sometimes it end up in the trash..lol

Win, loose, or draw, it is always fun!

Cheers,

Billy

 

[rbelli1]

No pun intended

Seriously?

Win, loose, or draw, it is always fun!

Seriously.

BoB

 

[Planobilly]

LOL

I can't tell you how many times I dreamed of making some sort of "space suit" contraption to keep me cool in the boat and other hot places. Actually that idea comes up a lot about this time of year here in Florida!

Cheers,

Billy

 

[iduhfuse]

Okay it's finished! But I'm having problems
:(((...when I first flicked the switch on the pump ran and the two fans ran beautifully. Later when I adjusted everything switched it on again and one of the fans sort of took a while to turn on? Then I switched it on again later and only one fan came on. Again, now both fans are not operating ...wth?! Now, only the pump runs, I double checked all the wiring and its solid. I have no clue what's going on. Thought maybe the battery was draining so there wasn't enough voltage so I had the engine running for a bit and still both fans not operating. I'm stumped and need your help!

 

[Tom.G]

Methinks the H₂O entered the prime air mover. (At least that's better than the brown stuff entering the air conditioning system.))

 

[iduhfuse]

Methinks the H₂O entered the prime air mover. (At least that's better than the brown stuff entering the air conditioning system.))

You think water shorted the fans? *Cries*

 

[Tom.G]

First-pass guess based on next to zero knowledge. Dry 'em out, they may resuscitate. Or maybe there's just mechanical interference.

 

[jim hardy]

Got a datasheet for those fans ? Any notes in it about mounting plane ? Are they brushed or electronic ?

Fans 7-12V rated at 0.5 Amps

12 V ? Car runs 14, surely they're not that tender.
However - inductance of water pump could be generating a turn-off transient...reverse polarity spike when you flip the switch off
one of those cheap bicolor LED voltage testers would probably show, look for a green flash at turnoff
http://www.walmart.ca/en/ip/auto-voltage-tester/6000119380174try a flywheel diode ?

old jim