PWM Help

[Penguin4x4]

I was trying to figure out the PWM from Bit-Tech (http://www.bit-tech.net/article/51/), and I am really confused. In the schematics on the bottom of the page, it shows 10KOhm resistors after the MIC520. Is this the "reference" size resistor for a 12v fan? very :confused: :confused:

[Penguin4x4]

If so, then that would make the R2 resistor 22KOhms, so I can switch between 12 and 7vs?

[Veland]

Hmm.. I'm looking at the schematics and your questions, but I can't really figure out what you're asking about. The 10K, is that the potentiometer? The VA1?

And why the talk about switching about 7V and 12V, the whole point of the circuit is to give a possibility to dial in any voltage between 4.8V and 12V.

So, what is the problem? I'll try to help, but..

[Penguin4x4]

OH, so its selectable between 4.8 and 12? I thought it meant either or. R1 is 10K and R2 is 15K.

[fireflew]

I thought the point of the PWM was to vary the on time... not the voltage... so like the fan is always either getting 12v or nothing... just at a very fast rate... aka 30Hz or so... the thing that changes the speed of the fan is he length of the 'on pulses', aka the pulse width... hence pulce width modulation... BTW if anyone finds any of the MIC502BN packages let me know... best I have found is 2.12 a piece with a minimum order of 50... quite a few more than I wanted...

[Penguin4x4]

So a 7v/12v Fanbus wouldn't get too hot like a regular Rheostat?

[Crazy Jayhawk]

That diagram confused me for a different reason. I see the 12V line going straight to the fan and the ground going through an optional diode to the fan as well.

Wouldn't this bypass the control circuit completely?

[fireflew]

Originally posted by Crazy Jayhawk
That diagram confused me for a different reason. I see the 12V line going straight to the fan and the ground going through an optional diode to the fan as well.

Wouldn't this bypass the control circuit completely?

As far as I can see that is part of what the diode is doing in there... looks to me like it is stopping the +12 from 'flowing' to ground... In terms of the optional aspect... if you remove the diode D1 i think you would have to remove the lines that it connects to as well... then the ground would be coming in at least through the transister... mabye not... but that is my best guess

[fireflew]

Originally posted by Penguin4x4
So a 7v/12v Fanbus wouldn't get too hot like a regular Rheostat?

The only part that should generate significant heat would be the transister, T1, and even then only if you had multiple fans connected up to a single circuit should it come into play. That is part of the allure of this type of solution for fan speen control, you have a fairly low power loss overhead, at least compared to using linear voltage regulators or straight rheostats, thus the heat generated should be less...

edit: although I have no clue what you mean by a "7v/12v fanbus"... I will assume you mean a fan speed controller that operates using pulse width modulation...

[Christoph]

If you can't get the parts, then there's not much point in learning how to put them together. If you don't mind paying for shipping from the UK and the VAT, you can order the parts from Rapid Electronics, but it looks like it would be a little spendy (VAT + int'l shipping).
I was also interested in building one or two PWM controllers, but got turned off by the VAT and the low availability of the chips. If you really want a PWM, you can buy a kit from Radio Shack (http://forum.oc-forums.com/vb/showthread.php?s=&threadid=93052).
I ended up PMing athlonnerd and starting this thread (http://forum.oc-forums.com/vb/showthread.php?s=&threadid=106870), in which he suggests a simple, cheap and effective way to control a fan for about US$4/controller with parts readily available at Radio Shack. I'm using one now, and I'm quite pleased with it.

[Veland]

A bit tired yesterday when I look at my reply... Yes, you are correct, the PWM adjust the on cycles and therefor the voltage is always at 12V. But since this happens very rapidly the fan "see" a different voltage, this is what ranges from 4.8V-12V.

This way it generates little heat as power is either on or off, not needing to be dissipated as in voltage regulators or reostats. The problem with these types of fan control, is that it may reduce the lifespan of the fans connected. This is because the fans use electronics to drive it insted of brushes (aka brushless) and this electronics can be damaged by PWM. But give it a go!

[Penguin4x4]

7v/12v Fanbus (http://bit-tech.net/article/52/3)

[fireflew]

looks more like a 5v 12v fanbus... but then I am probally just missing something... but yes that should also have relativly low heat production... just less 'customizable'

[Penguin4x4]

It produces low heat? OK, OK. Most rheobus's get red hot, I thought that the PWM would eliminate this........5v/12v fanbus is the same as 7v/12v fanbus.....OK, I guess a 7/12 bus'd work fine then. Thanks all for the help. :)

[drunkmonkey]

Originally posted by fireflew


As far as I can see that is part of what the diode is doing in there... looks to me like it is stopping the +12 from 'flowing' to ground... In terms of the optional aspect... if you remove the diode D1 i think you would have to remove the lines that it connects to as well... then the ground would be coming in at least through the transister... mabye not... but that is my best guess

the NPN transistor controls the ground of the fan and the diode protects te transistor from backflow, the diode isn't really needed, just a precaution.
Can't wait to get the parts to build one:D:D:D

Got confused bout a few things, for any electronic noobs here's a quick layout of what some of the parts do:
what I just siad^
capacitor C2 is to take the noise from the line which would screw it up.
R4 protects the transistor
C1 sets the frequency of pulses(.1uf for 30hz [uF=3/Hz])
VR1 controlls the speed of the fan(10k potentometer[sp?])
I know natta about the mic502

Feel free to correct me

[Temo Vryce]

Well it looks correct, but it's been a while since I have to figure this stuff out. As far as the chip goes, you can find more information here. (http://www.micrel.com/product-info/products/mic502.shtml) I hope that helps some of you out.

[fireflew]

yes... but finding somewhere to buy the chip has been what is holding me back... oh well... I may have to settle for something of less amusement... oh well...

[drunkmonkey]

Originally posted by Temo Vryce
Well it looks correct, but it's been a while since I have to figure this stuff out. As far as the chip goes, you can find more information here. (http://www.micrel.com/product-info/products/mic502.shtml) I hope that helps some of you out.

awsome link! thanks, the pdf on the page is really helpfull. I'm gonna try a custom circuit if I can find the parts. Anyone know where one of these would be in the US?

[Penguin4x4]

Originally posted by drunkmonkey
Got confused bout a few things, for any electronic noobs here's a quick layout of what some of the parts do:
what I just siad^
capacitor C2 is to take the noise from the line which would screw it up.
R4 protects the transistor
C1 sets the frequency of pulses(.1uf for 30hz [uF=3/Hz])
VR1 controlls the speed of the fan(10k potentometer[sp?])
I know natta about the mic502

Feel free to correct me

C2 is a capacitor: It maintains a constant voltage, as the MIC502 seems to operate better at a constant voltage, as opposed to a wavering voltage.

On the other hand, the decoupling capacitor, C2 is NOT optional. During testing it became apparent that the chip does not like "noisy" supplies. E.G. My 350W " Enermax has very long leads and poor quality connectors. This does not help maintain a clean supply. Things may be different for you but I consider C2 to be necessary.

I'm just a perfectionist. :rolleyes:

[Christoph]

I just found a place that will give me a quote for the Micrel chips.
If this works, I may end up buying several and selling them. How many would be interested?

[fireflew]

well I found one place that will sell them in 50 packs... ouch... and 2 places that claim to give quotes... been 2 days and still no responses... oh well I would be interisted in a few if you get your hands on them.

[drunkmonkey]

Originally posted by IdeaMagnate
I just found a place that will give me a quote for the Micrel chips.
If this works, I may end up buying several and selling them. How many would be interested?

I'm very interested, I might buy 4 or 5 If they are not too much and I have $cash$ at the time. About how much would they be each?

[Christoph]

The first place I found hasn't emailed me back yet.
I found another place (techsonic) that responded to my request, but they're out of stock.

[fireflew]

Originally posted by IdeaMagnate
The first place I found hasn't emailed me back yet.
I found another place (techsonic) that responded to my request, but they're out of stock.

too bad they are out of stock... hopefully they get some in soon =)

[Christoph]

Originally posted by fireflew


too bad they are out of stock... hopefully they get some in soon =)
They said that they'd email me as soon as they get some in, but I wouldn't bet the farm on it.

How about that place that sold them in 50-packs (Costco, perhaps?)? If there's enough interest, it could work out. At any rate, it wouldn't be too much harder than selling US$50 worth than it would be selling 50 chips.

Edit: While digging, I found this site (http://www.cpemma.co.uk/pwm.html). It's not exactly what we're looking for, but it's got a great DIY solution for those who are familiar with electronics.

[fireflew]

oh boy an op-amp... lol heh I will see if I can find that link again... I dont remember... the only diffrence is probally the 'power up' time on the microtel IC... not certain though did not really read that link throughly... looks interisting though... too tired... 18 hours straight of moving are too much.... I will look at it later and try to say something more inteligent.