I'm trying to find a best match for a regulator that can feed dual timer ic. The 5V circuit is dedicated to that IC which will generate a PWM pulse signal to PWM control the FAN speed. It's a 4 pin PWM fan, the 5V will not feed the fan!

Input Voltage: 24V

Input max tolerated voltage: 26V

5V line consuption: +/- 30mA, 50-100mA margin needed

Requirements: Small, good availability, easy to source, less components, low noise, cheap (Don't want isolated AIO solutions which costs a few bucks each), SMD, no zenners!

Circuit: Circuit

On my research found the following solutions:

  1. Linears regulators: From 24V to 12V L7812 to 5V L7805, i'm not a big fan of them but for the purpose they may fit?

  2. Bucks: MP1584 (overkill, big), AOZ1282CI (Smaller but still overkill). Then i come across with LT1107 and LT1111 which are simple and low current.

But i'm sure there's a better solution out there that i'm not aware. What you guys would use in that situation?

25/05/2019, Updated circuit

Circuit 2


27/05/2019, Updated circuit, Final

  • Added extra fan header to take advantage of empty space
  • Change pot circuit to use Jack solution
  • Reduced PCB overall size
  • Removed Foot switch and use pot with switch
  • Change P mosfet to N mosfet

enter image description here enter image description here enter image description here

Linear Regulators seens do the job just fine in this situation, as indicated and helped out by @DaveTweed on comments. Thank you

Edit - 25/05/2019:

End in use LM2936HVMAX-5.0/NOPB to support higher voltages to use with higher voltage fans, up to 58V (If needed)

23/06/2019, Fixed and Updated circuit v1.1

  • Added alternative XH headers for fan and power, eg. regular fan header
  • Added solder jumper for test or ignore headers
  • Added solder jumper to select 2/3/4 fan wire operation
  • Added more information about pins on board
  • Fixed wrong mosfet footprint
  • Changed VH orientation, now tab faces out the board
  • Better label positions and reorder
  • Compact board and components
  • Better thermals

enter image description here enter image description here enter image description here

  • 1
    \$\begingroup\$ Anyway, for less than 40-50 mA, I'd go with linear; above that, switcher. Specific product recommendations are off-topic for this site. \$\endgroup\$
    – Dave Tweed
    May 24, 2019 at 0:32
  • \$\begingroup\$ Yes, the fan pwm signal must be 5V, i think, not sure if it will fry with higher voltage but fan datasheet says PWM signal voltage: 5V, don't wana risk with 12V and invalidate the fan. About consumption i can't tell now but i will be happy with 100mA margin, because all i need is that generated signal from the timer. \$\endgroup\$ May 24, 2019 at 0:37
  • 3
    \$\begingroup\$ In that case, the LM7805 is fine. It's dissipating about 0.5W, so put a small heatsink on it and make sure it gets some airflow. \$\endgroup\$
    – Dave Tweed
    May 24, 2019 at 0:58
  • 2
    \$\begingroup\$ if you have several fans put a schottky diode in series with the signal to each. \$\endgroup\$ May 24, 2019 at 3:59
  • 1
    \$\begingroup\$ There is no real benefit to having linear regulators linked together. Sure, your 5V will run cooler, but all the heat will just end up on the 12V instead. So just use a 5V on its own, with a heat sink. \$\endgroup\$
    – hekete
    May 24, 2019 at 4:53

1 Answer 1


The proposed schema to control a 4-wire fan is NOT in spec.

You need to read and understand the specification for the 4-wire fan you are using.

Probably the best historical document describing the fan spec was produced by Intel on Formfactors.org, that site is long gone but you can get the document here.

In this document the PWM freq is stated as 21-28kHz and the PWM drive signal MUST be an open collector/Drain drive (in other words a pullup resistor or active pullup was NOT permitted). The voltage on the PWM pin was no more than 5.25V, but this was set by the fan ....not the driving elements. In some fans avaialble today, if you actively pullup the PWM pin or connect to two fans you get the wrong speed. Most of the early fans had a limited low speed range too, with 20-30% full speed not unusual.

Things have changed over the years and many fan manufacturers have improved their specifications. One example here for the Delta fans which can use a PWM freq of 30Hz- 30kHz, tolerates active pullup and 10V PWM amplitude.
Arctic fans are another example, with their own specification for multiple fans (up to 4) on one PWM signal line.

Your proposed PWM fan control

I would suggest that the operation using a 556 timer may have problems.

  1. You are using active pullup, which may not work depending on the fan.

  2. You mention using 16kHz which may not work depending on the fan.

  3. Your freq and pulse width controls will interact.

  4. T1 does absolutely nothing. I think perhaps you meant to create a CC source with T2, but what you have does not work that way. At high temperatures you may find T2 is turning on sufficient to reset your 556. If you are depending on simply the R3C5 time constant your timing is way out, T2 may be on for 8-10 Tau.

As a suggestion you may be better to design a solution based on an ATTiny85 driving an AO3400 or AO3402 for the PWM output. It would be a lot less components and board space. The ATTiny would require about the same current (12-15mA) as you might use for the 556 timer so a small SOT-89 78L05 would be adequate to power the MCU.

Update: Since the OP has now posted details of the fan being driven, it's possible to show some of the hazards to not following the 4-wire fan spec.

The fan is a Sanyo-Denki 9BMC24P2G001. The datasheet has little detail of the PWM signal, but the Catalog shows on page 551 the following information on the PWM signal:

enter image description here

Clearly from this information you can see that providing an active pullup may actually damage the PWM signal input or in your case would simply shut down the +5V regulator. The only way to drive this type of input is open collector/drain. To use active pullup you would require either a series resistor to limit current or a series Schottcky diode.


The new circuit is a vast improvement over the old one, with just a few comments:

  1. If you want to drive multiple fans then simply add more AO3400 FETs

enter image description here

  1. It appears you want to switch between an internal Pot and an external Pot. I would suggest the following change to ensure safer wiring and predictable failure modes.

enter image description here

With the circuit above you don't feed +5V into your wiring, so it's safe from short circuits.
Using two A/D inputs allows you to easily separate both the function and scale of the external versus internal Pot.
Providing a limited scale for the pots means you can detect an open circuit wire, or a faulty pot.

I have no idea what your intent is for this piece of the circuit:

enter image description here

  • \$\begingroup\$ Thanks for pointing out, i will take a look at formfactors document. I built that circuit based on this: overclockers.com/forums/showthread.php/… The post says: "Both circuits are using PWM frequency that is set according to FormFactors standard for 4 wires PWM fan -> formfactors.org/developer...e_PWM_Spec.pdf" I have that circuit implemented on a test board and so far it worked with every fan i have toss in, PC fans, agressive delta and san ace fans like 9BMB24K201, 9BMC24P2G001... \$\endgroup\$ May 24, 2019 at 15:56
  • \$\begingroup\$ The fan i'm using now for the project is 9BMC24P2G001, 25Khz PWM which is what circuit is generating, so ignore 2. as that is just a note about 16hz, not used on circuit. About 3. i also saw that, not sure why they put it like that but the function works, i didn't try to remove T1 and test out but i can do that and see the results. Using a MCU will require software burn and extra work, i prefer keep it only with hardware \$\endgroup\$ May 24, 2019 at 16:04
  • \$\begingroup\$ @TiagoConceição I'm sorry , but the link to overclockers seems like a good example of what NOT to do with your fans. While mentioning the Intel spec, they then break all the rules. Still if it works for you, go for it. I'd suggest you not connect PWM signals together (multiple fans driven by one signal). If you want to drive multiple fans use Schottcky diodes as already suggested in the comments. \$\endgroup\$ May 24, 2019 at 17:18
  • \$\begingroup\$ It works, but if not the correct way i rather fix it now and use the correct spec. So for the fix i must remove that R1, and exchange traces from 9 and 12? \$\endgroup\$ May 24, 2019 at 17:30
  • 1
    \$\begingroup\$ @TiagoConceição I added an update to the answer. You have insufficient specification to give much more help, but I think I have shown what you proposed is likely not to work or be reliable. If you want to tackle this problem with a small MCU, then you have to do some study I assume. Read up on driving a 4pwire fan from Arduino to start with, and then how to read a potentiometer using the A/D inputs. If all this is beyond you them hire professional help. \$\endgroup\$ May 25, 2019 at 5:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.