[SOLVED] Notches on 4-pin fan don't match - need adapter

JudMc

Reputable
Sep 8, 2020
19
2
4,525
I'm trying to replace the stock Foxconn fan in an HP 8200 with a Noctuna fan. Everything was going fine until I tried to plug it into the motherboard. The Foxconn plug has tabs over the 1 and 4 pin (black and blue) but the Noctuna has tabs over the 1 and 3 pins (or 2 and 4, depending on which way you look at it).

Is there an adapter with a female end like the Noctuna and a male end like the Foxconn/motherboard?
 
Solution
What you describe says that the fan is NOT reporting its speed to the header, and may NOT be receiving the PWM signal to control its speed. That makes me wonder whether the two signal lines for those (Green and Blue) may have been interchanged. I had recommended that you solder new fan wires to old connector wires by matching colours, which I assume you did. But that MAY have been wrong. I suggest you try these tests for that.

1. DISconnect the Blue and Green lines. We KNOW that, on the new fan from Noctua, the GREEN line is its speed signal. So, with those wires disconnected, start your system. Watch the SpeedFan display for speed of the fan. Try to connect TEMPORARILY (by touch or twisting) the fan Green tach wire to each of the...

JudMc

Reputable
Sep 8, 2020
19
2
4,525
I found the question below, which seems to be the same problem, but I don't understand the answer.


There is only one 4-pin connector on my motherboard, so this solution will not work for me.
 
Last edited:

Paperdoc

Polypheme
Ambassador
There MIGHT be more to this than you think. On all current computer fans the female connector on the end of the wires has two ridges running down one side. These enclose Pins 1 to 3. IF it's a 4-pin connector, Pin 4 is outside the range of the ridges. On the mobo MALE header, there is a plastic "tongue" sticking up beside Pins 1 to 3, and those two ridges fit on the outsides of that tongue. That way you cannot connect them wrong.

Now, you refer to those ridges as "tabs". This leads me to suspect that the header on your old mobo does NOT have that same male fitting with the tongue. Instead it MAY have a sort of "socket" of plastic with the pins inside that sleeve. And that sleeve has some slots in one side that match "tabs" or ridges on the outside of the old fan's connector. Is that right? Of course, that also ensures that you can only connect one way.

HOWEVER, the fact that the old fan and mobo use a connector DIFFERENT from what has become "standard" means it is also possible that the PINOUT they use does NOT match what is the standard today. IF you can figure out what the old system pins do, here is how the current "standard" 4-pin fan connector is wired. NOTE that this is for the newer-design PWM style of motor, and NOT what an older 3-pin simpler motor (speed controlled by varying Voltage supplied) does. Recall, Pins 1-3 are between the ridges, Pin 4 is outside.
Pin 1 = Ground
Pin 2 = Fixed +12 VDC power
Pin 3 = Speed signal (5 VDC pulses, 2 per revolution) generated by the motor and sent to header on this pin
Pin 4 = PWM control signal, approx 22 KHz freq, Pulse Width Modulation signal, 5 VDC peak)
Unless you can verify that the mobo header DOES supply those signals in some form, you may not be able to use that new fan, even with custom re-wiring.

Since this is from an older system it may pre-date the introduction of the 4-pin PWM fan design. In that case the old fan likely is of the 3-pin Voltage Controlled design, and you might be able to rig a modern fan for that system IF you know what all 4 pins of the header on the old mobo do.
 
Last edited:

JudMc

Reputable
Sep 8, 2020
19
2
4,525
Thanks, I have a volt meter but I don't know much about electronics. I've thought of cutting the plug off the old fan and wiring it onto the new fan. I can check the voltages - will that be enough?

But I can look for specs of the fans:
Foxcomm DC 12V 0.40A model PVA092G12H (and then in smaller type: P17-AE)
Noctuna NF-A9 PWM DC 12V 1.2W 0.1A

----
I found this for the old fan - does it help?

https://www.google.com/search?q=foxconn+pva092g12h+wiring+diagram&lr=&as_qdr=all&sxsrf=ALeKk00hCbsIEJI4uRDOfXsrs-zSJBRL9w:1618350210722&tbm=isch&source=iu&ictx=1&fir=-u1_3lUWTfic3M%2C_xT-vclAF3jekM%2C_&vet=1&usg=AI4_-kT2_CNFyl82JneRec91aadoSl-J6Q&sa=X&ved=2ahUKEwiDmLqrmPzvAhUnneAKHb0eBu4Q9QF6BAgNEAE&biw=1244&bih=623#imgrc=-u1_3lUWTfic3M

--------------
And when I searched for the old fan, I got this, which appears to be the same:
https://pinoutguide.com/Motherboard/mb_pwm_fan_pinout.shtml

Both fans have black, yellow, green, and blue to the same pins.
 
Last edited:

Paperdoc

Polypheme
Ambassador
Yes, those pretty much do tell us the fan wiring. On the new Noctua, turn the connector on the wire end so that its ridges are up (you see them), and the wires are pointed towards you, open holes away. Then the two ridges will be on the RIGHT, and one hole in from the left. Pin #1 is to the right, #4 left. The most common wire colour coding is:
Pin 1 - Ground = Black
Pin 2 - +12 VDC = Yellow
Pin 3 - Speed signal = Green
Piun 4 - PWM signal = Blue

It appears the old Foxconn wires use that same colour code, and the sequence across matches, so there's a VERY good chance they have used the same colours for the same functions. Moreover, that Foxconn fan IS of the new PWM design. It appears the ONLY thing different about it is the mechanical detail of the connector body. Your plan is good. Just cut the connectors off both fans leaving some wire to work with on each connector. Then splice wires of the SAME COLOURS together and tape up. Should work just fine.
 

JudMc

Reputable
Sep 8, 2020
19
2
4,525
Well, I did it today on one computer. (I have one more new fan for another one.) It was a bit tedious working on those tiny wires - they are smaller than the smallest setting on my wire stripper.

It is working but a bit strangely. It stays quiet. On 100% load it mainly bounces around 75C but it peaks at 82C. It doesn't throttle back when it goes over 80C. I think I can hear it rev up as it gets hotter, but SpeedFan doesn't show the fan.

That might be the reason for the change in the plug and socket - not fully compatible.
 

Paperdoc

Polypheme
Ambassador
What you describe says that the fan is NOT reporting its speed to the header, and may NOT be receiving the PWM signal to control its speed. That makes me wonder whether the two signal lines for those (Green and Blue) may have been interchanged. I had recommended that you solder new fan wires to old connector wires by matching colours, which I assume you did. But that MAY have been wrong. I suggest you try these tests for that.

1. DISconnect the Blue and Green lines. We KNOW that, on the new fan from Noctua, the GREEN line is its speed signal. So, with those wires disconnected, start your system. Watch the SpeedFan display for speed of the fan. Try to connect TEMPORARILY (by touch or twisting) the fan Green tach wire to each of the Green and Blue wires on the connector that is still plugged into the mobo header. Which one DOES show you a speed? Leave that connected.

2. Take the BLUE wire from the fan and connect that (twist) to the other available wire from the header connector. We'll use a feature of normal system start-up to test for speed control, but this is easiest if the system is relatively cool. You need to be able to see the fan and watch it carefully right after booting. As soon an you start up, all fans normally will be started at full speed to ensure they do start. After a few seconds of the POST process, the fan speed should decrease significantly as the system recognizes that is it already cool. Thereafter the fan speed is managed by the mobo automatic system. So IF the fan speed does change, then the speed control system is working. If not - if it appears to stay at max speed - then it is not.

This assumes that the actual power supply lines - Black and Yellow - are connected correctly. If possible BEFORE you start these tests, remove some insulation on the connections you made on these lines and connect a voltmeter. Yellow is +, Black is -. Leave these connected for the tests. For a standard 4-pin PWM fan, that should be 12 VDC at all times when the system is on, irrespective of the fan speed.
 
Solution