pc atx power supply & triton great planes

[il Zott]

hi to all :-))
for my new triton I use a pc atc power supply 235W, it works very very well, its outpu is 12v@8A, there are some modifications:

1)
to have all power I put a car stop lamp 12v 21w on 5v (red and black)

2)
a switch to power on the pc atx p.s., this switch is on black and green

3)
two connector on the case (red and black) to connect the triton


some pics

[il Zott]

2

[il Zott]

3

[il Zott]

4

[il Zott]

5
I turn af 180° the fan, now the air goes in and not goes out

[il Zott]

6

[Pir8]

Hello!
What's the car stop lamp used for?

Thanks.!

[Freddy418]

very nicely done, I have to try that next!

[Awakened]

The light bulb acts as a load on the 5V rail to bring the voltage on the 12V rail up to speed. I used two.


http://awakened.dyndns.org:8080/stuff/Planes/Charger.jpg

[Blackhawk3D]

Why did you reverse the fan direction? Does it cool better with the air blowing on it? I always though the idea was to pull the heat out of the supply.

[JoeC]

Blackhawk3D is correct, you should not reverse the direction of the cooling fan. The componets, heatsinks, and openings are arranged so the parts that need cooling the most are getting the air first. By turning the fan around you can cause parts to get overheated. Also you now have a flow that is a lot less laminar and the heat sinks may not work as well. The volume of air moved over the parts is going to be the same either way, there is no advantage to doing that.

[highflier]

It looks like the Light bulb is inside the case. I would make sure that the airflow is over the light to exit. They produce alot of heat. Also remember that the supply might not generate as much heat under a smaller 3 amp load of the Triton.

Highflier

[il Zott]

Originally posted by JoeC
Blackhawk3D is correct, you should not reverse the direction of the cooling fan. The componets, heatsinks, and openings are arranged so the parts that need cooling the most are getting the air first. By turning the fan around you can cause parts to get overheated. Also you now have a flow that is a lot less laminar and the heat sinks may not work as well. The volume of air moved over the parts is going to be the same either way, there is no advantage to doing that.

Hi :)
I preffer to invert the fan because when the power supply is in the case of pc the fan must pull out the hot air produced by cpu,hd, cd-rom... but now the power supply is out ohf the case and the fan cools better the componets.

ciao :p

[il Zott]

Originally posted by highflier
It looks like the Light bulb is inside the case. I would make sure that the airflow is over the light to exit. They produce alot of heat. Also remember that the supply might not generate as much heat under a smaller 3 amp load of the Triton.

Highflier


hi

the light bulb is quite cold, because are 12V and i give them only 5V....

[JoeC]

Since the amount of air moving through will not be different either way, reversing the fan will defiinatly not make it work any better.

As I said, the power supply was not designed to work that way. While it is no longer working with already warm air, reversing the fan may cause it to not cool properly.

Just my two cents.

[robert harik]

I know this thread is old, but for anyone else who wants to reverse a cooling fan- it really does make your equipment run hotter. By blowing the air in you are pressurizing the air inside the power supply which really slows the air flow down. As opposed to pulling the air out, which causes a slight low presssure inside. Trying to push the air in also stalls the air over the fan slighty and further reduces the airflow and cooling. You can actually hear this when the fan is installed backwards. You are also putting the heat from the fan motor into the power supply.
What was said earlier about the arrangement of parts is also true.
The voltage rating of the bulb wont make it run hotter or colder, you just have to figure wattage.

[Chippie]

Well I went the whole hog and modified an atx psu to give max output on the 12 volt......in fact I adjusted it to give me 13.8v @ 8 amps.........its been in regular use for over a year now with no problems. I load tested it with 2 65w 12 headlamp bulbs initially, then proceeded with a proof test with my Super Nova charging 16 2400 sub C's at max rates( 5A.....) current draw was a little over 7.2 Amps.......not a murmer fromt the psu and runs cool

Circuit attached if anyone wants to see how I did it.


I have some photos of the unit too somewhere......

[blueprints]

Wow... Hardcore....

[Marion]

Very nice job ! I did something similar, but used 10 Watt resistors instead of the stop lamps. Those are beautiful photos :-)

[cosmo]

I don-t want to go out of toppic but...can I use it as a car battery charger (I want one for my Winch bat!)
TIA.

[il Zott]

no because you need a power supply and not a battery charger, they are two different things :)

regards

[cosmo]

I know my english is a disaster so I try to explain.
I do my homework and make my own power supply with PC -just like here-, buttt...I have a battery car need for a charger.
The power supply have 12v 4amp output.
Can i charge my battery direct with the power supply?
sorry for re-write my question :o)

[Andrew0820]

Here is a conversion page (http://web2.murraystate.edu/andy.batts/ps/simpledriver.htm) with photos, information and some theory of operation -- link at bottom.

[Uitgeslapen]

i would bet no difference

[Terry S]

I connected a digital thermometer probe to the hottest heatsink in my supply with a cable tie and put a 6amp load on the output. I then turned the fan round and tested it again.....Guess what....No change, the meter settled around 85'c both times.

Terry

[KreAture]

Actually, as long as you can get adequate airflow to the parts that need is, a fan that sucks cold air in will have the following benefits:
- It will be more efficient since cold air has a higher density than hot air, thus moving more air.
- It will have a longer lifespan since cold air will not effect the sleave-bearing assembly as much. (Most of theese fans are either sleave+ball or simple sleave.)

Depending on the fan-grill in may be more or less noisy. I prefer to use those wire-mesh grills and reduce rpm on fan to make it a bit quieter if it has no temp-regulator. Best solution is to make sure the temp reg for fan is attached to the heatsink for 12v rail components.

[Terry S]

Thats a good point, I forgot about the built in temp regulator, maybe thats why it worked the same both ways. Anyway I dont know why we are worrying, I dont know of one person that has overheated their pc power supply conversion.

Terry

[Andrew0820]

As I said, the power supply was not designed to work that way. While it is no longer working with already warm air, reversing the fan may cause it to not cool properly.


The heat load generated by the PS is minimal WRT other components within the PC case. For many manufacturers, the direction of airflow is contingent on whether they want to have airflow directed over specific components, rather than efficiencies of cooling the PS. The vast majority of PS's are equipped with fans 80mm or larger. Guideline recommendations suggest at least a minimum of 25cfm rating. Many supplies ship with 90mm fans rated over 50cfm. The restricting factor is venting. Case design (and cowling designs for enclosed engines) will generally specify exhaust air openings between 1.5 and 2.5 times the size of intake openings. Since the fan area is the largest open area in a PS case (vent sizes are dictated by EMI suppression), the preferred direction (and again recommended by the ATX12 V2.2 Developers Guideline) is exhaust. Since ATX cases must meet fairly stringent dimension requirements, the typical ATX case will be evacuated between 750 to 1000 times a minute, depending on the manufacturers vent area. The cooling capacity is usually sufficient to effectively cool both the PS and case, except in mid-size and large towers with high heat CPU's and/or multiple peripherals.

The bottom line is: Airflow direction doesn't really matter -- cooling capacity will far outstrip the PS heat load.