Cooling fan that looks like resistove load?

[Flyback]

We are looking for a 48VDC (25W) Axial fan to cool our electronics with. We need it to look like a resistive load so that we can supply it from a small value ceramic capacitor as we do not want any electrolytic caps on the PCB.
As you know, many BLDC fan motors do not actually have “high frequency PWM” controlled fan coil current, -rather they simply have “commutation PWM’’ which switches from one coil to the next in order to spin the fan.
As you know, if there is no “high frequency PWM” acting on the fan coils, then the fan coils can resonate with the supply capacitor if the supply capacitor is not big enough, and to be that big the supply capacitor has to be an electrolytic one, which we do not want. As you know, if the fan coil commutation period is greater than the LC resonant frequency of the fan coil inductance and supply capacitor, then the resonance causes problems for driving the fan.
Do you know of any fans which have the high frequency PWM’ing of the fan coils such that the fan coils effectively look like resistive loads from the supply capacitors point of view?

 

[kubeek]

It´s interesting to see how you allways come up with an overly complicated solution to problems that don´t need that much attention.
If you don´t like electrolytics, then why not use a film cap, or multiple ceramics? Increase the pwm frequency to get better filtration using small components?

But to answer your question, maybe a DC motor with mechanical comutator could be better, but it also may have the same problems as bldc fans regarding the pwm.

 

[MrAl]

Hi,

Buy a 36v fan and put a big resistor in series with it

 

[dr pepper]

Als idea allthough inefficient would probably work.
A well designed lc filter would take out your noise issues, but that involves a electrolytic, or a really big choke.
Even a brush motor doesnt appear pruely resistive.

 

[ronsimpson]

Digikwy.com 490-5786-2-ND 22uF 63V X7R $1.50 at 500 pieces. **broken link removed**
**broken link removed**
The 10uf 50V cap is under $0.10 at one real. 50V (is too low)

 

[Flyback]

As you can see on page 3 of the following, the current in most brushless dc fans is highly resonant, this causes much ripple current in the supply capacitor...the resonant appearance to the waveform that you see on page 3, is actually the fan coil inductance resonating with the supply capacitance.

https://www.nmbtc.com/pdf/engineering/cooling_fan_behavior.pdf

 

[ronsimpson]

Flyback,
Your peak current is 2x the average at 460hz. Wow that wants a very large capacitor.
Ron,

p.s.
You need a picture like this above your name.

 

[Flyback]

The current that you see on page 3 of the link below is highly resonant. As you know, it will draw high ripple current from the supply capacitor, that ripple current being equal to the following...
SQRT{[RMS value of the current]^2 - [average value of the current]^2}

https://www.nmbtc.com/pdf/engineering/cooling_fan_behavior.pdf

I realise that its not massive AC RMS ripple current, but when you are trying to get rid of electrolytics then it is not wanted.
Also, the problem as you can see, is that if the supply capacitor that supplies the fan is of a small faradic value then the resonant period of the supply capacitor and fan coil inductance is less than the commutation on time of the fan, and that gives problems of resonating current inside the fan which wears out its bearing.

I am sure you agree that the resonant appearance of the waveform on page 3 is the resonance between the supply capacitor and the fan coil inductance.

 

[MrAl]

Hi,

I would think that even if it does not resonate it is still something to think about.

I always say that more data is better than less data, as long as it is well organized. I think they did a decent job in the pdf. I thought about similar things in the past because i started using big fans to cool my computer case guts and i wondered about the effect on the +12v line. What happens when we turn off, there must be some inductive current coming from the fan, and if it is enough it could charge the filter caps up to some level which would have to be determined.

I thought about suggesting a series inductor similar to what dr pepper said, but as you know before you can insert an inductor into a circuit you have to know the effects during any and all modes of operation, which includes turn on and turn off, so you cant just stick an inductor in between the P/S and the fan or the fan could be destroyed on the first run.

It's also very hard to design an inductor capacitor filter with a single inductor and one capacitor without knowing the detailed dynamics of both the source and the load. It might be possible to cheat here by using a pi filter, but that will take decent sized caps too.

So the only thing left is to use a lower voltage fan and use a regulator circuit to regulate the voltage or maybe the current. It would be operating in the 'limiting' mode so it should not have to store any energy. Unfortunately this will act like a largish resistance so there will be power loss. But that's life. If you are not allowed to store energy then your only choice is limiting. Think about this for a while. Also, added series resistance decreases resonance as the damping factor is altered.

A 36v fan that draws 0.667 amps will cause a 12v drop in voltage with a 18 ohm resistor in series with it, and the resistor will dissipate 8 watts, so the efficiency will be 75 percent in driving the fan. 32 watts in with 24 watts out means instead of the power supply powering 24 watts it will have to supply energy to power 32 watts. It's a loss but that's life. Of course you need to test this as well.

 

[ronsimpson]

I am sure you agree that the resonant appearance of the waveform on page 3 is the resonance between the supply capacitor and the fan coil inductance.

No I don't agree. I don't see resonance in this picture. or at least not LC resonance.

I think you need to store enough current to power the fan over one cycle. There are only a limited number of ways to store energy.

I like the wind up spring method of sorting energy but you should use a capacitor.

 

[alec_t]

I am sure you agree that the resonant appearance of the waveform on page 3 is the resonance between the supply capacitor and the fan coil inductance.

Sorry, I don't agree either. What the waveform shows is the coil current dropping to zero at each commutation, then building up due to the coil inductance. The current varies as a function of the rotation angle, due to back EMF; hence the current profile. I've obtained similar waveforms with a BLDC fan run from a 12V battery, with no supply capacitor.

 

[Flyback]

a 12v battery is a big capacitor.
The resonance on the waveform is clear to see...it curves one way then the other, like a sine, its definitely LC resonance.....it makes sense too.....you have an L and a C and they are being pulsed, and inevitably they ring with each other.
If you use a small supply capacitor, you will see the effect of the ringing very clearly indeed, as there will be multiple ringing periods within 1 commutation on time event....this is bad news and means torque pulsations and bearing wear in the fan......as well as extra core loss in the fan's core.

 

[MrAl]

Hi,

Well did you try adding a small resistance in series with the fan? Sometime even a small resistance damps out an oscillation enough to make it work. Maybe even 1 ohm would do it really.
A buck circuit can even be wild but add a little series resistance and it's nice and smooth. That's because a pure L and C with even a small amount of injected energy is an oscillator...not a filter. Add a little resistance somewhere and the oscillations damp out and it becomes a filter. All buck circuits have at least some series resistance.

 

[alec_t]

Here's my effort at simulating a fan motor commutated at 100Hz. The coil current waveform is not a million miles away from that shown in post #10 . You can play with varying C1, but it makes very little difference; there is no apparent LC resonance. The sinusoidal part of the curve is due to the back-emf. In reality the curve shape would depend on the rotor and stator pole profiles.

 

[ronsimpson]

My thought:There is a oscillator in the fan. The zero current time is the dead time of the oscillator. The blue line is the inductive charging if the coils. I think the red line is the coils saturation.

There is a common power circuit, using two transistors and a transformer. The transformer is driven in one direction until the core saturates. That causes the other transistor to turn on. Saturation is the key trip point for the oscillator. Any way the current wave form is close to you motor.