FEATURED Building PWM Controller for 4 wires PWM fan

Seebs · Mar 2, 2011

bing said:
Sebastian, maybe using Lenny's pcb layout ? or the one done by Martin ?

Link please? Or is that the one that got published on the front page article?

Bobnova said:
My advice is to get everything into the PCB without soldering and then start double-checking that you have everything in the right place.
Then send all the "ruined" PCBs to me
Once you have everything laid out and double checked, start soldering.

I've tried a few times... I think the issue is not so much laying the components down on the PCB... I think it's that my hands are not as steady, and my eyesight is not as strong as when I used to do this stuff back in high school...

I'd send you ruined PCBs, but to be honest the ones that I've ruined have not lived long after the fact to be sent to anyone... They all ended up in little pieces shortly after the soldering messups. :bang head

Yep... I'm a hothead and when stuff pisses me off; it usually dies a painful death... I can't remember how many wireless xbox controllers have gone flying out windows anymore.

EDIT
I've been asked to help with a project for someone here. It involves controlling 12-24 PWM fans and a PWM Waterpump... Still don't know all the details...
Don't know what fans will be used yet, or if they will all be linked to one speed controller... or if there will be separate controllers for each fan... Need more details.
All I know is that this one needs to be as close to perfect as I can get it.

Let's say all fans are to be controlled from a single controller... Would it be possible to split the PWM signal into 24 different wires and control all the fans from a single circuit? Or will that result in too much signal degradation? What's the best approach?

Bobnova · Mar 2, 2011

lol
You need a desoldering iron, that might prolong their life a bit.
I bridge stuff constantly, bit of desoldering and presto!

jalyst · Mar 5, 2011

reposting as original wasn't very clear.

xXSebaSXx said:
I think I wrote that model number wrong when I made that post... What I meant to write was that I had ordered the two San Ace fans 9SG1212P1G01... I don't know where I got that model for the GT from, but they'd be nice to have too. Not in a long while though... I think $300 in fans is more than what I need.<SNIP>

Yeah I think you must've been confused by my earlier posts.

That's not the model no. for the GT...
The GT is D1225C12BBAP-31, whereas the 2x you listed were D1225C12BBZP-00.

This explains the significance of each character in the model no:
http://www.nidec.com/servo_part_numbers1.htm

Given the only difference between the two model no's is at character 11 (Z Vs A) and 13+ (00 Vs 31).
They'd be pretty much the same, apart from the GT not having PWM, right?

That being the case...
The AP-31 (although only LVC) should give an accurate idea of what the D1225C12BBZP-00's like, right?

I'm hoping the answers are a definite "yes" & "yes"...
Because if my AP-31 tests well, then I'll prolly also get 2x D1225C12BBZP-00

PS: Thanks for the heads up on the signature.

NP!

Bobnova · Mar 5, 2011

I've got a Nidec Beta VA450DC V34809 coming, 5400rpm and 220cfm in theory.
Much better test of my PWM thingie

jalyst · Mar 6, 2011

^ Is that a 25mm thick fan?

bing · Mar 6, 2011

As requested, below the ghetto mod

to boost the PWM signal output for those who wants to hook with many-many fans or even PWM water pump.

Actually there is already a standard technical specification on how much each PWM fan should draw the power from the pwm signal output, but alas, it seems like there so many variations from different fan makers, for example as BF's experience, the original PWM circuit from Nidec is working fine with his other normal fan, but when using that monster Sanyo fan, it needs a booster to strengthen the signal to have it works normally.

On how many fans max for each pwm signal output, the best method is to find it out through trial and error.

Attached below, with just one additional 556 ic (again, this is actually 2 555 ICs cramped in single body), it boosts the signal with two additional pwm outputs for load sharing, so if you plan to hook up 12 fans :shock:

for example, just connect those 4 fans pwm wire to each pwm output.

View attachment 92414

PS : Actually I was about to make the booster with discrete components like few transistors and resistors, after thinking Sebastian's experience and also for other 1st timer DIY-er, looks like lower components count will make it easier to build.

Seebs · Mar 6, 2011

bing said:
As requested, below the ghetto mod to boost the PWM signal output for those who wants to hook with many-many fans or even PWM water pump.

Actually there is already a standard technical specification on how much each PWM fan should draw the power from the pwm signal output, but alas, it seems like there so many variations from different fan makers, for example as BF's experience, the original PWM circuit from Nidec is working fine with his other normal fan, but when using that monster Sanyo fan, it needs a booster to strengthen the signal to have it works normally.

On how many fans max for each pwm signal output, the best method is to find it out through trial and error.

Attached below, with just one additional 556 ic (again, this is actually 2 555 ICs cramped in single body), it boosts the signal with two additional pwm outputs for load sharing, so if you plan to hook up 12 fans for example, just connect those 4 fans pwm wire to each pwm output.

View attachment 92414

PS : Actually I was about to make the booster with discrete components like few transistors and resistors, after thinking Sebastian's experience and also for other 1st timer DIY-er, looks like lower components count will make it easier to build.

Thank you kindly good Sr.... This will come in handy for that project I had mentioned to you via PM.

You are a gentleman and a scholar...

Thanks again
Sebastian

bing · Mar 6, 2011

My pleasure, glad able to help. :grouphug:

xXSebaSXx said:
Link please? Or is that the one that got published on the front page article?

Sorry, missed this question, here the one from Martin -> http://www.overclockers.com/forums/showpost.php?p=6674854&postcount=154

Looks like its quite clear from his pic how he soldered all those components.

Bobnova said:
I've got a Nidec Beta VA450DC V34809 coming, 5400rpm and 220cfm in theory.
Much better test of my PWM thingie

Hey Bob, don't forget the video at your e-pot circuit working with that fan please !

Btw, how do you change the e-pot's value, using up & down buttons ? Really curious how is your circuit looks at the final finished shape.

Seebs · Mar 6, 2011

No problem bing...

This is the one I was talking about... Super small, clean, compact... Perfect for the project that I've been asked to help on.

I realize that with the booster circuit that you've laid out ti won't be as small, but still I figure it can be fitted in an area small enough.

m0r7if3r · Mar 6, 2011

Jez, nice photo...you a hobbyist or a professional?

Bobnova · Mar 6, 2011

At the moment it's rigged to a light sensor, I need to buy a sack of misc. buttons and such. Button control would be easy enough, given some buttons

Seebs · Mar 6, 2011

m0r7if3r said:
Jez, nice photo...you a hobbyist or a professional?

LOL... I wish that were the case... Neither the circuit, nor the picture are mine.

That is one of the example circuits that were built throughout this thread; it was used on the front page article that was published a little while ago.

http://www.overclockers.com/pwm-fan-controller

I just used it because that's what I'd like to be able to accomplish with mine...

S

I.M.O.G. · Mar 6, 2011

And the gentleman who took that photo, is either professional, or semi-professional when it comes to photography I believe (my memory is failing). If not, he's damn good.

MIAHALLEN · Mar 6, 2011

xXSebaSXx said:
This is the one I was talking about... Super small, clean, compact... Perfect for the project that I've been asked to help on.

I realize that with the booster circuit that you've laid out ti won't be as small, but still I figure it can be fitted in an area small enough.

Looks great Sebas...hopefully you're as successful with yours. Thanks a bunch for helping me out :thup:

bing · Mar 6, 2011

m0r7if3r said:
Jez, nice photo...you a hobbyist or a professional?

I.M.O.G. said:
And the gentleman who took that photo, is either professional, or semi-professional when it comes to photography I believe (my memory is failing). If not, he's damn good.

LennyRhys said:
Yes I do have a badass camera (well, two) but I'm a photographer, so that's ok

Those shots were taken using canon eos 1-ds mark II as the exif says, definitely not an average dslr camera, its a pro league cam ! :rock:

jalyst · Mar 7, 2011

jalyst said:
That's not the model no. for the GT...
The GT is D1225C12BBAP-31, whereas the 2x you listed were D1225C12BBZP-00.
This explains the significance of each character in the model no:
http://www.nidec.com/servo_part_numbers1.htm

Given the only difference between the two model no's is at character 11 (Z Vs A) and 13+ (00 Vs 31).
They'd be pretty much the same, apart from the GT not having PWM, right?
That being the case...
The AP-31 (although only LVC) should give an accurate idea of what the D1225C12BBZP-00's like, right?

I'm hoping the answers are a definite "yes" & "yes"...
Because if my AP-31 tests well, then I'll prolly also get 2x D1225C12BBZP-00

@bing you own some D1225C12BBAP-31 & D1225C12BBZP-00 fans.

Can you please advise...
Are they exactly the same aside from one being LVC and the other PWM?

Assuming yes, D1225C12BBAP-31 should be a good indicator of the D1225C12BBZP-00's performance characteristics right?

I realise a D1225C12BBAP-31 won't have the same RPM range as a D1225C12BBZP-00 paired with a decent PWM controller.
But at the same RPM they should perform equally right?

Thank-you.

bing · Mar 7, 2011

I have only D1225C12BBZP-00 which is the PWM version of D1225C12BBAP-31 which I don't have.

Mechanical wise, both are identical as stated at the datasheet, so cooling performance must be identical as well. I don't believe they deliberately made them different since there is no incentive for that in business perspective.

At same rpm, they should perform equal too.

Now, when it comes to electronic, PWM version is much more complex and needs more electronic components, hence more expensive than it's 3 wires brethren.

For ease of rpm control for fine + large rpm range and also the power efficiency, pwm 4 wires version simply beats 3 wires fan out of the water.

Hope this helps.

jalyst · Mar 8, 2011

bing said:
I have only D1225C12BBZP-00 which is the PWM version of D1225C12BBAP-31 which I don't have.

Mechanical wise, both are identical as stated at the datasheet, so cooling performance must be identical as well. I don't believe they deliberately made them different since there is no incentive for that in business perspective.

At same rpm, they should perform equal too.

Now, when it comes to electronic, PWM version is much more complex and needs more electronic components, hence more expensive than it's 3 wires brethren.

For ease of rpm control for fine + large rpm range and also the power efficiency, pwm 4 wires version simply beats 3 wires fan out of the water.

Hope this helps.

Thanks mate, same mechanical wise is all I wanted to know!
I'll soon test my linear voltage controlled AP-31 to see if I like it more than my G1238B12BBZP-00, for some applications I have in mind.
The G1238B12BBZP-00 will have to rely on the limited PWM control I have built into my MB for now.

LennyRhys · Mar 8, 2011

I.M.O.G. said:
And the gentleman who took that photo, is either professional, or semi-professional when it comes to photography I believe (my memory is failing). If not, he's damn good.

Heh many thanks for the kind words guys :thup:

and yeah I'm a professional photographer who likes to take hobby pics in his spare time, and as bing says the camera plays a huge part in the quality of the images

And since this is the "cooling" thread I thought I should share an image of cooooool - a nice frosty sunrise which I took on holiday last week

Nice to see the thread continuing to develop. I really should get back on the PWM controller wagon, but I've been so busy lately...

LennyRhys · Mar 8, 2011

Sebastian - I just dug this out for you and I hope it is of use to you.

This is the sketch I made after breadboarding, and it's the design I chose for the final stripboard layout.

Tracks run horizontally, and all the red circles denote breaks in the tracks (under or between components, etc.); and the caps are all labelled accordingly.

The 556 is pointing up, as it appears in the schematic in the article.

FEATURED Building PWM Controller for 4 wires PWM fan

Fronting as a Mod Member

Senior Member

Member

Senior Member

Member

Low Profile Senior

Fronting as a Mod Member

Low Profile Senior

Fronting as a Mod Member

Member, Water Cooling Sticky Reading Enforcement O

Senior Member

Fronting as a Mod Member

Glorious Leader

Senior Member

Low Profile Senior

Member

Low Profile Senior

Member

Member

Member

Similar threads