Question How to move wires in a SysFan plug?

[ImWolf]

I had ordered some replacement 3-pin fans, but didn't look at the auction page close enough. The wires are connected to the little Molex plug in the wrong arrangement. Apparently the fans were intended for use with some sort of Westinghouse product, but the seller did state that the wires can be "moved to accommodate PC cooling".

So, before I start cutting wires and soldering, have you folks ever re-arranged wires on these little plugs? Are they held in with some sort of spring clamp?

https://cdn.shopify.com/s/files/1/0...-bfbc-997e13c55998_1024x1024.jpg?v=1559321959

Thanks,
Wolf

 

[Eximo]

There are little tabs that stick out on each pin that need to be depressed to remove the pin. No soldering required. You should see the metal of the pin showing in the rectangular holes. Depress with a small screw driver or pin while pulling on the wire.

https://m.media-amazon.com/images/I/61I1UyBUSdL.jpg

Then it just a matter of figuring out which wire is which.

 

[ImWolf]

There are little tabs that stick out on each pin that need to be depressed....

Thanks for the quick reply. You got me on the right track.

On the little plugs I have, I actually had to stick a pin into a small rectangle hole just above the pin connector hole, and then pry the tab up and away from the plug body. Then the wire and connector pulled right out.

Good Day,
Wolf

 

[Paperdoc]

Glad you got the wires and their connectors pulled out. The pinout diagram you posted in the first item is correct. NOTE that the colour coding shown IS the "standard" way, but what was used in the particular fans you got may not be. The important points are:

Orient the connector as show with the ridges on one side of the fan's female connector DOWN (these fit beside the plastic "tongue" of the mobo male header)

THEN the Ground is on one end as shown, and the + DCV power supply line is the middle position. Normally this is 12 VDC max for full speed, and the mobo header will reduce that as needed when it wants the fan to run slower. The "tach" or speed signal is generated by the fan and sent back to the header on Pin #3.

The diagram mentions +5 VDC as an option. This is NOT normal, but it would be IF the fan is designed for a 5 VDC power supply. So examine the fan motor label to be sure. For a normal 12 VDC fan, the power actually supplied by the header MAY go as low a 5 VDC for running at minimum speed without stalling - it should never go lower, because then the fan might stall.

 

[ImWolf]

After all the replacement fans I've bought over the years, this is the first time I came across this issue, but I'll be looking much closer at fan wiring from here on. Since this site doesn't allow me to attach images except web links, I found something close. That attached image is strange too as the diagram is correct per my MB manual, but the picture of the plug in the same image isn't wired correctly.

The fans I just bought have a Red, Black, and White wire, and all 3 were in the wrong position in the plug.

#1 was Red, #2 was White, and #3 was Black. The seller later told me that if I introduce +12V to that White wire it would damage the fan, but that didn't happen. I tested the fans using only Red & Black with a power supply, and the fans began spinning slowly at around 7V, and then spun more normally at 12V.

(They are marked 12V .07 A)

The seller also later told me that the fans only run at one speed (which doesn't seem to be true from my testing), and that the White wire is only for a PWM controller. He said the fans don't have a tach sensor and that I shouldn't use the White wire at all for PC cooling if those pin-outs don't have a PWM controller.

In the mean time I'm still running my old fans, of which only one is slightly annoying, so I guess I'll find out if these fans actually have a tach function at a later date.

 

[Paperdoc]

Yep, non-standard fans for a computer system. The colour codes you cite are not "normal". For a new-style computer 4-pin PWM type fan, the wires normally are
Pin 1 Black Ground
Pin 2 Yellow +12 VDC constant voltage
Pin 3 Green Speed pulses signal sent back to header
Pin 4 Blue PWM signal to fan

On SOME systems the colour sequence instead is Black - Red - Yellow - Blue. The first three of these is the same as on all 3-pin fans, and then the fourth pin is blue wire.

Your fans do not use these colour codes.

Now, you said the fans work when you connect 12 VDC power to Black and Red. I assume you did NEGATIVE to Black, and positive to Red. You also say that the fans run slower at reduced voltage. This IS normal. Because the new PWM fans were designed for maximum compatibility with the older 3-pin fan system, they are normally supplied with fixed 12 VDC power plus a PWM signal. Inside the fan a small chip uses the PWM signal to modify the flow of current from that supply through the windings to reduce speed. IF you connect such a fan to an older 3-pin header which cannot do those signals, the fan gets NO PWM signal to use, so it cannot modify current flow. But the header does supply power at VARYING Voltage, not fixed 12 VDC, so the fan speed IS controlled in this way. Your seller did not understand this.

IF the WHITE wire on your fans really is solely for the PWM signal as your seller says, then certainly connecting a +12 VDC supply to that could damage the fan's internals. So do NOT try that. THEN, if you assune the seller is correct about the proper finction of the WHITE wire, you could slip those three wires into a female 4-pin fan connector body with the white on the Pin #4 end. Since your info is that the fan has NO speed pulse signal output to go back to the header, then NONE of your wires would go to the Pin #3 slot of the connector. With those connections made, if you plug that into a mobo 4-pin header that is configured to use the new 4-pin PWM Mode of control, the header should be able to control the fan speed. Alternatively, if you plug it into a 3-pin mobo header, that ALSO should control the fan speed as I said above. But in both cases you simply would not have any speed reading at the mobo header, and it MIGHT give you a Fan Failed warning because it has zero speed reported. NOTE that the header does NOT need or use the speed signal to control fan speed. It uses that only to detect fan failure.

FYI a small note since you obviously have some knowlege in this field. PWM control of fan motor speeds in a computer is NOT done the same as a PWM speed control system for larger motors in industry. In those systems the electrical supply fed to a motor IS already modified into a fast pulsing DC Voltage at the motor Controller unit. That design could not be adapted easily to the older 3-pin fan system in computers for easy compatibility, so it was changed for computer fans. For this reason computer 4-pin fan makers specifically tell you NOT to try to feed power to their fans from a common industrial PWM speed control system.

 

[ImWolf]

Thanks for all that info PaperDoc.... (I read some of it a few times).

My old machine MB has 2 of the 3-pin headers for SysFan, and one 4-pin header for the CPU fan, which is the only one that has the PWM controller. From previous reading, I got the impression that the PWM function does not change fan speed at all, but instead gives intermittent power to the fan which effectively keeps turning the fan off and on again. But that's not what you're describing.

Currently, my bios does report fan speeds, so when I get tired of listening to this one fan slightly humming I'll put one of the new fans in to see how correct the seller was about the white wires real performance. I can't think of any reason attaching a PWM lead to the "Sensor" pin on the MB header would cause a problem.

 

[BFG-9000]

You can directly plug 3-pin fans into a 4-pin header. Just go into the BIOS and change the speed control from PWM to Voltage unless you want the fan to always run at full speed.

If it's not a tach wire I would just leave that wire unconnected. You aren't going to be able to monitor RPM without a tach wire anyway.

 

[SkyNetRising]

this site doesn't allow me to attach images except web links

You can upload your image to imgur.com (press on new post - upper left corner)
and post link here.

 

[Paperdoc]

OP, you understood that the PWM control system "gives intermittent power to the fan" correctly, but the TIME factor in this is important. In that system the fan's own chip uses the PWM signal from Pin #4 to switch On and Off the power connection from the fixed +12 VDC supply line (Pin #2) to the windings. The PWM signal is like a square wave except that a square wave is "On" exactly 50% of the time. In a PWM wave the information IS the "Time On" and can vary from 0% to 100%. In computer fan systems, that PWM signal has amplitude 5 VDC and frequency of about 22 KHz. So the pulses of 12 VDC applied to the windings (and hence current and motor torque) happen at 22,000 times per second. Obviously the motor rotation does not start and stop - it keeps turning with very rapid "kicks" of torque, each "kick" lasting from 0 to 1/22,000 of a second. Thus the AVERAGE torque (and hence speed because of work and friction) varies. If the motor is fed from an older mobo header with 3 pins it gets NO PWM signal from Pin #4 and the chip cannot do its switching function. But that header supplies power from Pin #2 at a VOLTAGE that is not always 12 VDC. The voltage is reduced when the header wants the fan to slow down, so the motor current and speed are slowed that way.

You said, "I can't think of any reason attaching a PWM lead to the "Sensor" pin on the MB header would cause a problem." Correct. There would be no signals on such a connection. BUT what if your seller was wrong on this point, and the White lead really IS supposed to carry a speed pulse signal generated in the motor back to the header for counting. IF you connect such a wire to Pin #4 of the mobo header that IS using the new PWM Mode of control, then that header will send out the PWM signal to the motor and it will arrive at the point in the motor's circuit that expects to be sending out its own speed signal. THAT scenario could cause a problem in the motor. So only ONE of these is hazardous. If you leave the White lead from the motor on Pin #3 of the connector, OR if you do NOT connect that White line to anything, the motor cannot send any harmful signal to the header, and the header cannot send a harmful signal to the motor. When you make the connection this way you get a result. IF that white lead really is the speed signal, it will be received at the header and show you a speed. IF, instead, that White line is supposed to carry the PWM signal to the motor, you will NOT get a speed reading from the header. That would not GUARANTEE that the seller was correct in calling the White line the PWM signal line, but it would be proof that is it not the speed line.