12V DC Speed Controller 1

[Chuff]

Can anybody help with a simple schematic for a (simple) speed controller for an automotive fan. It consumes about 15A at full voltage (12V DC).

A PWM design with low losses would be preferred.

 

[automatic2]

A pot will be the 'simpliest'. Relay switched resistor grouping is practical for several fixed speed possibilities. PWM is certainly a good way to go, but not necessarily 'simple' nor required. You'll need to vary the reference voltage to the controller to change speed which could put you right back to step one (varying voltage to control speed).

Check TI's TL494 spec sheet for some simple control circuits built around a pretty standard controller. You'll need a drive element capable of your 15 amps (darlington)and a PWM controller (TL494). This device has onboard drivers capable of driving your switching element, as well as error amps for feedback control. You'll simply feedback the motor voltage to an onboard error amp. The difference between this feedback and your reference input (speed signal) results in motor drive.

Back to the original problem, how do you derive a speed signal?. Maybe a stacked comparator that switches a relay(s)the select a resistor combination is adequate. This could easily give you several speeds as opposed to full on/off.

 

[circuit1952]

Chuff,

Look at these:

http://www.4qdtec.com/pwm-01.html

http://www.solorb.com/elect/solarcirc/pwm1/

http://www.google.com/search?as_q=p...s_occt=any&as_dt=i&as_sitesearch=&safe=images

http://www.discovercircuits.com/M/motor-cont.htm

 

[melone]

What have you done to solve your own problem? More specifically, what circuits have you tried? What were your results? What are your FULL design constraints, i.e. size, cost, power, etc.?

I do not presume to speak for all of the Eng-Tips community, but I find it exceedingly irrating when someone posts a question and expects us to know the best solution. Every solution has its pros and cons. When posting a question, pretend that you are asking your boss for help. Prove that you've done at least the most basic of groundwork, and that you don't want someone to actually do your work for you. I am sure that everyone here would like to point you in the right direction, but don't want to solve the problem for you!

 

[CarlPugh]

At 12 amp you will require an extremely large pot. Switching regulator is probably the way to go.
Check out power supply circuits, there should be some that meet your requirements.
Of if efficiency is not important, place a power transistor with a heat sink in series with the motor and vary the drive current (Or voltage if using a FET).
DISCLAIMER: The preceding will not give accurate speed control and may not work with some motors.

 

[OperaHouse]

WOW! A TL494 lover. Actually, I'm going to be driving an automotive fan with a TL494 and a couple BUZ71 this summer when I pull out my stupid clutch fan. You can drive these in a fixed PW mode or with a temp sensor. I've recommended these to others for home projects because it can be found in many old computer power supplies. Unfortunantly they seem out of favor now. Digi Key doesn't stock them.

 

[automatic2]

Newark carries stock. The surface mount device is quite popular. Good idea with direct temp feedback.

 

[maged541946]

Hi, at 12 Amps and a motor, the best way to go will be a FET driven from a simple 555 timer that change pulse width, data sheets have good examples for one or better yet 2 LM555 timers. The fet help because it is a resistor when open , unlike bipolar transistors, you also need the freewheeling diode to bupass the switching currents.
I used a similar way to speed control a fork lift runs on 2 dc 2 motors 200Amps 48 volt each

 

[melone]

The problem with FET's is that they act like a resistor when they are closed as well. Let's assume that the FET you use will have a Vds(on) of 100mOhms (pretty low for a high power device). The voltage drop across the FET will be V = 12 * (.1) = 1.2V. If you compare that with a PNP BJT, then the drop will only be in the 0.1 - 0.3V range.

FET's are nice, but they can also create other problems. Try a IGBT (Insulated Gate Bipolar Transistor). These devices have the drive strength and voltage drop of a BJT, but the input drive levels of a FET.

 

[cbarn24050]

Hi Melone, i'm not sure your right there, even a big bjt (245302) can't get below 0.7v at 15A. 2 cheap old mosfets like IRF541 can do better or maybe a IRFZ40 which is even better. IGBTs are not very good at all when it comes to saturation voltage, very few are under 2v at 70% rated current.

 

[melone]

Did you look at PNP's?

 

[cbarn24050]

Hi Melone, there is no significant difference between them, all the pnp transistors I can find are slightly worse that there complementry npn. Maybe you could quote some part numbers.

 

[IRstuff]

OK, the old standby 2N3055 has a VCEsat of around 3V @ 10A:

http://www.onsemi.com/pub/Collateral/2N3055-D.PDF

Others are not that much better:

http://www.onsemi.com/site/products/parametrics/0,4446,805,00.html

Compared to Hexfets:

http://ec.irf.com/v6/en/US/adirect/ir?cmd=catNavigateFrame&punchInID=600560

TTFN

 

[maged541946]

Hi, a low voltage FET (60 volts)will have on resistance of .02 to o.05 ohms, those in the 0.1 range ususlly for higher voltage like 100-200volts, the ones I used made by IXYS generated less heat than the older disk SCR's that I had to replace (along with control circuit)