• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

12v to 48v?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

princeofdarknes

Member
Joined
Nov 21, 2005
Location
MS, The redneck south -.-
Alright, here's my thought: I found this website that sells FFB1248SHE and EHE delta's for $4 each, the only problem is that they are 48v
http://www.excesssolutions.com/cgi-bin/item/ES2815 SHE
http://www.excesssolutions.com/cgi-bin/item/ES3393 EHE
(and there's a $25 spending minimum on the site, so you can get 6 for the price of... however many $25 will get you ;))
This site also has other Delta's (and all their fans) for $4 and $5 12v if you look, i'm gonna be gettin a 40mm just incase my chipset fan dies, a couple 80's and a couple 120's, but they dont sell FFB's in 12v or 24v. So is there an easy way to convert a 12v input rail to 48v that doesnt create too much excess heat? It be great if i could find all the parts on that site XD
Remember the converter will have to support an output of 48v at 0.6A

mmFFB
-prince
 
not positive but i have used 24v fans at 12v, so it may work for a 48v fan to come to 12v...not sure but it could be worth a shot, or else i think your gonna have to get a 24-48v psu
 
no, i doubt you can run 12v on a 48v fan, its less than half, i've run a 12v fan on 5v, but any lower and it doesnt like it, the 48v fans have a minimal voltage of around 30v i'm pretty sure
 
You best bet would be to build or buy a AC mains to 48v DC supply.

You can convert 12v to 48v, but to get a few amps on the 48v rail you'll be getting quite expensive...

Tom
 
Well this might work...
http://www.allelectronics.com/cgi-bin/item/TC-100/790/100W_UP_DOWN_TRANSFORMER_.html
Just use it to step down 115/120 to about 60V AC, and put in a rectifier bridge and smoothing caps, and you'll probably be getting about 50V smoothed DC out of it.

However, some old AT style computer supplies used to have a 2:1 transformer on the 115/230 switch, so you could use one of those like that too.
 
i think the main thing would be rewinding the motor's armature, so it will run at lower voltages with more current, then the other components should handle lower voltages fine, unless the hall effect sensor wont switch at the lower voltage or somethin.

i havnt sucessfully rewound a fan, i always get distracted and set the fan aside and end up losing it or ill do something to break it. i have reversed the polarities to make one spin the opposite direction, but never reduced number of turns, at least on one with an RPM sensor where i can actually collect data. maybe some day i'll actually do it and get some info, current draws, voltages, RPMs, all that fun stuff. it would be a nice writeup, and if done right (theoretically at least) one could make a fan spin as fast or slowly as they wanted, as long as the hall sensors and transistors and the other internal components in side the fan could handle the load.
 
RoadWarrior said:
Well this might work...
http://www.allelectronics.com/cgi-bin/item/TC-100/790/100W_UP_DOWN_TRANSFORMER_.html
Just use it to step down 115/120 to about 60V AC, and put in a rectifier bridge and smoothing caps, and you'll probably be getting about 50V smoothed DC out of it.

However, some old AT style computer supplies used to have a 2:1 transformer on the 115/230 switch, so you could use one of those like that too.

That won't work like that. What you will get out of the 2:1 step down is 60 Vac which is actually about 85 volts peak or 170 volts peak to peak. When you rectify in a silicone bridge rectifier the resultant voltage is about 83.6 volts peak. (1.4 volt drop over the bridge) Depending on the load and cap size, the average DC voltage will vary. Given a good sized cap for a reasonable load and the transformer at full load, I would expect the DC voltage to be within 75-85 volts. Since the transformer you suggested is 100 watts, the actual output voltage will be higher than printed since they overcompensate to get the correct voltage at full load.
 
IMHO Your best bet would be to find a suitable 12 volt fan rather than trying to save a buck buying a 48volt fan, then spending time, energy and more money trying to get it to work.

Just my .02 cents
 
I gotta go with Sam on this. Time is money, you'd be better served spending that time on looking around for a suitably cheap 12v fan than spending that time rewinding a fan or building a 120VAC to 48VDC step down circuit if your only goal is cheap cooling. Of course if the intent behind a step down circuit or fan rewind (or whatever else) is to educate yourself along the way then by all means, make a good solid design, then post it on the forums. It might help others (could get stuck too).
 
I usually like to learn upon the way, it'll probably cost the same if i were to get the $4 FFB and downvolt it or make a 48v powersupply
I didnt know that the voltage is based on how much the wire is wound, is it the more turns the higher voltage or the other way around?
And i'm curious, what's the highest rated Amperage one should put on a motherboard fan header? Because i wanna know if these 80mm Delta's wont blow the headers.

Oh well, i'll figure out what i'm gonna do with that 48v fan, if i get it, later, thanx for suggestions guys
-prince
 
YES finally i can apply rc cars to computers...The more windings around the armature (stature in this case cus its brushless) the lower rpms and higher torque you will get. The less turns the more rpms and the less torque. This is why i run a 55 turn or windings around my Rc rock crawler for the torque (it spins at 8000rpm) and i run any where from 19turns or windings (37000rpm) through 11turns (55000 rpms) in my street and offroad buggies.
 
hm, so i theorize i'd do more turns, so the voltage to RPM ratio would even
ah well, i got a small 40mm fan that the wire goes around one central point and i'll mess with that if i feel like it before i mess with a large, more important fan ^^;
 
you could build a switchmode psu to convert 12v to 48v. just oscilate the 12v at a high frequancy like 24Khz or so and run it through a 1:4 turns ratio transformer (you could wind one on the core of an old transformer from a computer psu.) the core doesnt need to be big because of high frequancies. The just use high speed rectifiers to rectify the secondary winding of the transformer and then a filter cap to smooth out the ripples. You could also try building a charge pump converter or a buck converter. Im not too familiar with those though. All i know is the charge pump does something by charging a capacitor to the input voltage and somehow using a diode to charge a second cap to twice the input. Cant realy explain it. There is also a way to use a diode and a capacitor to make a volatge doubler. You'd still have to oscilate the input using some power transistors or mosfets but the diode+cap method might make it easier if you dont want to wind a transformer for a switchmode supply. I am currently thinking about building a circuit to power my 24V 3 cm fans off of the pcs 12v.
Keep in mind that if you are doubling the voltage the input current will be doubled.
1Amp of 12v will only supply 500ma of 24v at 100% efficiency. so 1Amp of 12v will only supply 250ma of 48v. As you can see the current is an inverse relationship to the voltage. Ill see if I can pull up some schematics and info on charge pumps, switch modes, buck converters and the diode+cap array for ya. But for now im tired so im gonna catch some sleep. Ill post back tomorrow.
Peace out.
 
(at least i theorize)

to run at a lower voltage, but keep the same ammount of power, we need to increase the current. saythe fan will draw 28 watts at 48V (its just a beastly fan), it will be drawing about .583 amps. to get the same power output at 12 V, the amperage has t obe increased to 2.33 amps. this seems like a nice advantage of using higher voltage fans, you wont have to have as much current on the wires powering them.

on my RC i had a 17 turn motor, while it was fun i kinda wish i had a 30 turn or so so my batteries would last longer, 3 mins run time isnt as fun, when it takes 15 mins for them to charge. the fans shouldnt be TOO bad to rewind, since you just have to keep taking them off till it gets right.
 
Alright I did a bit of searching and heres what I came up with.
Here is a nice PDF about charge pump circuits. Its fairly technicaly advanced but it should give you a good understanding of how they operate.

Here is a slightly less advanced page on charge pumps
These are mainly used as voltage doublers. But building 2 of them and connecting them respecively should result in 2 seperate doubling of the voltage with an output of 48v. Make sure you check the voltage characteristics of the parts used. If a device is built to double 12v to 24 volts it might not be able to handle a 24v input to get a 48v output. Slight modifications might be needed.

Here is a nice page on a DC-DC converter that will supply 30v from 5v input. Larger transistors could be used for higher power output.

Hope that helped a bit. Ill let you know what I decide to do to run my 24v fans from my 12v rail and give you a heads up on my design that may help you along.
Good luck
Peace out!
 
I thought the same thing with the voltage:amp ratio earlier, so i understand that. I'd Have to figure out somewhere to put the extra slack of the wire, or should i cut it and solder it to rid of the excess?

I'm likin that 5v-30v DC-DC converter, since one would probably run the fans at a lower voltage anyway, or i could use it so i wouldnt have to take as much wire off the fan, and make the fan native to ~30v but i'm not sure of 20mA would be enough
 
Back