Controlling Dell Fans >:-}

starhawk

[H]F Junkie
Joined
Oct 4, 2004
Messages
8,908
This isn't a question so much as an answer (for once).

I've known for a long time that Dell fans had an internal speed control, based primarily off a green or blue blob (almost certainly a cheap thermistor) that's in the airflow of every single Dell fan since basically Day One.

I've never really investigated that green blob. I've known two things about it... one, it's a temperature sensor of some sort; and two, it is The Thing Of Electronic Damnation Which Makes Dell Fans Not Work In Regular PCs. A pity indeed on the second one -- Dell fans are fast as shit when they get going. But they always seem to behave the same -- spin up fast for ~5sec and then quiet down to almost nothing.

Today I dragged out two fans. One, a Datech fan from one of several systems (it's Dell's P/N 7G538, for two Optiplex models, four Dimensions, and a trio of Precisions) and the other, a Minebea/NMB blower that happens to be from a P4 Optiplex (the matching heatsink is marked 7P182) although that particular heatsink and fan is a VERY popular item in SFF P4 Dells...

I didn't care about wrecking the Datech fan, I've got at least five more just like it :rolleyes: but I've only the one NMB blower so I wanted to test my theory on the Datech first. What theory? Having looked up thermistors on Wiki I had a vague idea of how they work -- just enough to get an idea. Seems that, generally speaking, as a thermistor gets hotter, it conducts electricity more easily. So: short the two bare leads on the green (or blue) blob and the fan should spin up enough to levitate off the desk (OK, I exaggerate, but only slightly!).

I took the wires off the weird shit Dell connector for the Datech fan, leaving the pins intact. Then I shoved the red and black wires into a spare drive-testing "molex power supply" that I'd received courtesy of eBay some time ago. (Alas, it's one of the new ones that's 2a per rail rather than 2.5a per rail. Oh well.) Plug in the supply and the fan cuts on. Put a flathead screwdriver across the two leads for the green (in this case) blob, and HOLY SHIT THAT'S SOME FUKKIN' WIND COME OUT THAT FAN. Same thing happened with the NMB blower. You could blow moon dust to Earth with this thing. Dell makes some serious fans!

Lazy ass that I am :p I came up with a nifty trick for solderlessly joining the two wires as well. In the NMB blower, the thermistor is on a little pair of wires, and the leads do a 90deg angle into a spot where the blower rotor isn't. This is under a bit of black paper tape. So: peel back the tape, wiggle the blue (this time) blob out and cut it off, bend the leftover leads straight, and as it turns out they're about the size of pin header pins, so I stuck 'em into a spare CD drive jumper and added a generous helping of electrical tape. Since I didn't need it, I cut off the white RPM line from the cable as well.

For once I'm not bothering with pix, it's just a Dell blower fan with a couple extra lumps of electrical tape :p

Now I have a blower that's "strong like bull" for next summer's exercise walks. Trouble is it's got a 1.34a rating on it :eek: I'll figure a portable power source out later... I think I've got plenty of time for that indeed.

Anyone else with a spare Dell fan or three, who comes across this -- now you know how to harness the power in these fans for yourself :D Have fun but be safe...
 
...actually, as far as power sources go, a quartet of old C cells seems to do quite nicely :D good thing, I've got oodles of the damn things... of course there isn't a flashlight in the house that uses em...
 
Thermistors are inverse type devices,using temperature vs. resistance (so the higher the temperature, the lower the resistance). You could most likely control speed with a pot, but you might have to do some digging on the thermistor to see its rating and set up the pot as appropriate.
 
Shit, what I do with old fans from dells is just grab the thermistor with some plyers and twist it, so it shorts at the lead. Done. 5000rpm fan go.
 
@EvilWays -- I was thinking that myself... didn't want to bother effing with the thermistor, just wanted something simple that made the fan take off for Mars ;)

@Jorona -- hey, I didn't know you hung out here! Pliers prolly work if you're delicate ;) I'm more the raging-bull-in-a-china-shop-during-an-earthquake sort tho...
 
I like Jorona's approach--easy, fast, and reversible if need be. I've got a 120mm fan out of an old PowerMac that has a similar setup, and for computer use, it's fantastic--very quiet until it needs to be otherwise. starhawk, does your fan levitate if you set it on a table? :D
 
Not quite, but powering on at full speed makes it turn around about 270 degrees... it'd do a full 360 I think if there wasn't a cord in the way :D
 
Oh, and BTW--fans usually spin up to full speed on power-up, whether controlled by the mobo or built with its own thermistor. That's intentional, to make sure the fans break the startup torque/friction.
 
That does make sense...

Just realized what you could do, altho it'd need a little soldering unless you want to rig it -- put a switch on there! Idea being the switch bridges the thermistor, so when you need it full speed you flip the switch on and FOOSH dust go bye-bye at high speeds. Flip the switch off and it goes back to regulating itself. Just an idea...
 
Can't seem to get the fan to spin up off that 6v SLA battery. Datasheet (link below) sez 8-12.8v is needed. How can I boost the output voltage of the battery? All the 6v->12v converters on eBay are rated too low...

Datasheet --> http://www.nmbtc.com/pdf/dcfans/bg0903.pdf
Mine is the BG0903-B044 model.
 
Dumb idea that probably won't work -- could I put a capacitor in series with the battery, and then use a transformer to boost the voltage...? or would that not work for some reason...?
 
Dumb idea that probably won't work -- could I put a capacitor in series with the battery, and then use a transformer to boost the voltage...? or would that not work for some reason...?

AFAIK transformers only work with an alternating current, there may be a way to boost it with a dc to dc converter circuit but you would loose a lot of energy in the process and drain your battery way faster
 
OK, idea dropped, doubly so because I tested it (very roughly) and it didn't work...

May build a charge pump for this if I can get a circuit working that can handle an amp and a half at 12v on the output...
 
You'd need a boost converter (ebay has 'em for pretty cheap). Just make sure the output voltage matches the voltage for your fan. Like this
 
As an eBay Associate, HardForum may earn from qualifying purchases.
That one won't work for me :( need one that can output 12v 1.5a -- 18w.

Can't find anything like that on eBay :(
 
Yes, you can, if you look hard enough. Most of those listings have the specs right in the title.

If a boost converter module isn't capable of giving you 1.5A@12V, it will still be useful--it'll just give you a lower output voltage than 12V. Besides, the datasheet only quotes 1.03A for your model of fan, so this or this would work.
 
As an eBay Associate, HardForum may earn from qualifying purchases.
Fan label says 1.34a.

The "15 Watt" converter is only a 15w converter at 9v input. At 5v input it's a 6w converter, below that it's a 3w converter... and the other one puts out all of 150ma at 12v... ouch.

Looking at this one tho, it seems promising --> http://www.ebay.com/itm/321199096446
I'm messaging the seller now.
 
As an eBay Associate, HardForum may earn from qualifying purchases.
1.34A is only the start-up acceleration current - steady state will be less. And if it doesn't get the full 1.34A but instead only gets 1A, it will accelerate slower, but will reach something close to full RPM anyway.
 
Back
Top