Please register or login. There are 0 registered and 1081 anonymous users currently online. Current bandwidth usage: 326.30 kbit/s March 29 - 06:12pm EDT 
Hardware Analysis
      
Forums Product Prices
  Contents 
 
 

  Latest Topics 
 

More >>
 

    
 
 

  You Are Here: 
 
/ Forums / Off Topic /
 

  Power usage 
 
 Author 
 Date Written 
 Tools 
I A Jan 28, 2011, 11:31pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List

Edited: Jan 30, 2011, 11:21pm EST

Replies: 10 - Views: 2357
I am making a cooling pad, so i am using 2 antec 3 speed system blowers which use 3.6W so about 7.2W together and then two 40mm 2 Pin 12V Cooler Cooling Fan. all of these going through 4pcs SPDT Toggle switch on/Centre off/on 3Pos. MTS103. so my question is do i need a circuit board and how much power or what type of power adapter would i need to use to allow fans to run at a nice rate.
http://www.antec.com/pdf/VGACooler.pdf

Please take a look at some pictures of the cooling pad and laptop at this(http://www.hardwareanalysis.com/content/topic/77232/) forum to understand more about the power and other, before posting comments.
Thank You :)


Want to enjoy fewer advertisements and more features? Click here to become a Hardware Analysis registered user.
john albrich Jan 29, 2011, 04:49am EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List

Edited: Jan 29, 2011, 05:01am EST

 
>> Re: Power usage

When you say MTS-103, it looks like you're referring to the actual mechanical switch
http://www.alibaba.com/product-gs/229417354/MTS_103_On_Off_On_Toggle.html

and not the semiconductor MTS-103 temperature sensor or fan controller, but I just wanted to be sure there was no confusion.

As for needing a PCB
No, you can use point to point if you want, and a nice mounting plate for the switches. e.g. a scavenged desktop 5.25" front panel bezel can make a suitable mounting plate if it's strong enough...or the cooling case itself can be used if it's strong enough. It pretty much depends on what you're going for aesthetically.

As far as adapter power goes
If you plan for all fans to be on at once and at full power, then you add up all the watts required for each fan. As long as the adapter has the output voltage and output watts capacity you'll be fine. If it can handle MORE watts than you need, that's fine, too. You should note that many cheap adapters are unregulated and if they say their output is 12volts, I've seen their output voltage range from ~20VoltsDC (no load) to ~5VoltsDC (when the maximum specified load is applied). So, as you add more fans to the circuit, the voltage drops and you don't get the speeds you think you'll get. In fact, the output could drop so low that one or more of the fans won't work, or won't be operating at maximum possible speed.

On the adapter's DC output
Current = Watts/Voltage
So if your fans take 10 Watts, the current the adapter will have to supply at a minimum is 10/12 = 0.85Amps. You'd want to go with a 12V @ 1Amp output adapter. As you add fans, the current capability has to go up of course.

Other Considerations
You also need to make sure your wiring between the fan connectors and your switches will handle any combined fan loads. For example, a wire the same thickness of one of the wires used between a fan and its connector, likely would not be suitable to connect 4 fans to your switchbox. That wire might overheat and cause problems. So if you were thinking of "recycling"/scavenging the wiring from your fans to make your switchbox connections you probably don't want to do that because you don't know how much current each wire can handle...all you know for a fact is that it was designed to handle the current going to that one fan by that manufacturer. And of course manufacturers tend to use the very cheapest components possible to keep their profit as high as possible. This is especially true of some PC fan manufacturers...so warning delivered.


The information in this section is for IF you start to see fans failing or if you add any electronics to your circuit (e.g. indicator LEDs or IC/transistor controllers, temp sensors, etc).

As long as you stick to only using mechanical switches, you generally shouldn't need to use what's called "back EMF" (aka "inductive kickback") suppression. The AC/DC adapter and the MTS103 switches should be able to handle any back EMF generated from turning the fans off and on in the low power circuit you've described. Well-designed modern PC fans also tend to produce very little back EMF.

But, IF you start seeing failures, back EMF is a possible cause.

When you switch an inductive load (like a fan) a high-voltage back EMF is generated that could damage the electronic components (possibly even other fans) and lead to instantaneous or future failure. An inexpensive back EMF suppressor can consist of dirt cheap silicon diodes wired into the switch circuit as needed to protect circuitry.

An internet search on the topic should show you what to do. As it depends on the actual circuit design, I can provide only general guideline here. You should be able to find dozens of webpages dealing with adding back EMF protection to fan switching circuits.
Here's an example: http://www.bit-tech.net/modding/2001/12/03/pwm_fan_controller/1


micro Jan 29, 2011, 01:49pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Power usage
@John, good info.

I A, when you make a post, please only make 1, not 3.

Though your question was covered. I assume that you want 4 fans so that they can be strategically placed to cool each component.

I believe you are better off with just two 120m fans running at 5-9v. They will run fine at lower then 12v and will be quiet running slower and still push more air then 4 small fans that have to run faster to move the same amount of air.

This should still cool close to what your 4 fan set up will do if not the same or more, due to the increase in air flow. This would be a simpler design to make. You could easily use a home phone adapter to run these.

GA-Z68X-UD3H-B3, 2600k @ 4.0
16 gig Corsair Vengeance
Evga gtx 260 216, samsung 2253lw"
Baracuda 7,200.12 CoolerMaster 212 +
Win 7 64, fsp fx700-gln, Razer DA,G15
Evil is powerless if the good are unafraid
Silence in the face of evil is it's self Ev
john albrich Jan 29, 2011, 02:23pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Power usage

micro,

I'm thinking he's doing something more complicated for some reason...probably wants some way to switch the fans in series and parallel configurations, possibly with different voltages...based on his use of 4X 3-position (on-off-on) switches. That's the only logical reason I could come up with for using that many of those particular switches.

He also might be thinking of doing something like that earlier laptop cooler project of a couple weeks ago? The one where the guy was also designing his own case and using separate fans to cool his laptop display panel as well as the notebook base.

Does seem a bit over-complicated, though.

I A Jan 30, 2011, 10:50pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Power usage
the reason for four fans is to check out this link/ foum i posted before to see the design of the cooling pad and other info.
http://www.hardwareanalysis.com/content/topic/77232/

I A Jan 30, 2011, 10:56pm EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List

Edited: Jan 30, 2011, 11:19pm EST

 
>> Re: Power usage
complicated yes but for a reason. many cooling pads are designed to cool but do not do the job so you end up wasting good money on something that did not help much. they pull air from the bottom, where there might be small amount of clearance for the cooler air to pass through. but it also might be pulling in the warm air, but when cool air is coming in from the side (not being blocked by you legs or table , etc... ) hot air is being pushed away from the back as to keep the back area from heating up. if you have used a toshiba qosmio x505 q885 then you would understand. or please take a look at x505 series.

couple of degrees so i am trying to go for as much low heat to laptop as much as i can.

connecting through usb port causes more heat to laptop, two at the bottom to allow cool air from the side the the cpu and gpu fans. and two in the back allow hot air to stay away from the laptop itself, so the hot air is not going back through the bottom

i did not mean to post three times it was an accident, and if you think there is a better way of doing this project then please give me advice. :) and i do not intend to sound angry or anything it is that people do not understand. so if you have a better idea/design please post.

john albrich Jan 31, 2011, 03:58am EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Power usage

Didn't remember that was you, and I didn't realize it's been a whole month since the last post to that earlier thread. Not too surprising considering the meds I'm on!!! :)

As for my "does seem a bit over-complicated" remark, don't take it too seriously. Everyone has their own ways to approach problems, and some just do stuff for fun, or to learn, or to see if they can make a difference. No probs.

I didn't realize the heat problem with that particular model notebook was so pronounced. Trying to prevent heat generated by one area of the computer from pre-heating other areas of the computer seems like a good idea on general principles.

I A Jan 31, 2011, 07:26am EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Power usage
thank you for understanding and you have been helpful before so thank you for that.
The heat problem is high with high activity.
So i decided to use the fans suggested on the other thread and fans on the back and just for looks and show i decided to use toggle switches and i am trying to spend as less money possible, so i could not find normal toggle switches for low price so i decided to buy the on/off/on switches

john albrich Jan 31, 2011, 08:16am EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List

Edited: Jan 31, 2011, 08:17am EST

 
>> Re: Power usage
A couple of good places for electronic parts at decent prices/shipping in the US, are

http://www.digikey.com

http://www.mouser.com

Over a number of years I've never had a single problem with either company.

There's another I've not used recently, and I can't recall the name. I'll edit this if I remember.

john albrich Feb 05, 2011, 10:12am EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Re: Power usage

Case in point.

Corroborating info on some fan companies using ultra-cheap wiring to save a penny on BMC.

Reviews of this fan
http://www.newegg.com/Product/Product.aspx?Item=N82E16835220014
say the manufacturer used/uses very thin wires, and the fan wires weren't even a decent copper alloy.

john albrich said:
...So if you were thinking of "recycling"/scavenging the wiring from your fans to make your switchbox connections you probably don't want to do that because you don't know how much current each wire can handle...all you know for a fact is that it was designed to handle the current going to that one fan by that manufacturer. And of course manufacturers tend to use the very cheapest components possible to keep their profit as high as possible. This is especially true of some PC fan manufacturers...so warning delivered....

john albrich Feb 12, 2011, 05:33am EST Reply - Quote - Report Abuse
Private Message - Add to Buddy List

Edited: Jun 13, 2012, 05:04pm EDT

 
>> Re: Power usage
Something else that might be interesting to you re: fan speed control. A way to control relatively high-power fans with a simple and cheap circuit (about $3 worth of parts).

Some of the cheaper case fan variable speed controllers (both bracket and front panel mounts) use a simple potentiometer (a variable resistor, also called a "pot") to change the voltage on the fan from about 6 to 12 volts. From what I can tell, this is the case with fans like:
http://www.newegg.com/Product/Product.aspx?Item=N82E16835200021
(see the note I've added at the bottom of this post)

One problem with this method is that the more powerful the fan, the more power the resistor has to be able to dissipate. This is a problem with some units and the "pot" literally burns-out partially or completely within a short period of time, or if you plug-in a much too powerful fan is even destroyed immediately. Some of the "feedbacks" from customers reported this happened to them. It also means you can't control more than one fan with one of these cheap controllers.

For example, if a 12V fan requires 6 watts (0.5Amps) it can take a relatively high-power resistor/potentiometer to handle the power. Some manufacturers might use a cheaper, low-power pot, leading to early failure.

But there is an alternative that can handle very high-power fans without problems. The downside is that it doesn't provide smooth control of the fan. The speed changes are in discrete steps...BUT they are small ones. Think of it as the equivalent of a typical "3-speed" case-fan switch but it handles much more power, can handle multiple fans, and provide more speeds.

You can use a simple, dirt-cheap silicon diode to reduce the voltage going to a fan by 0.7volts. You simply add more diodes in series to further reduce the voltage...each diode taking away 0.7volts from the fan.

In addition, this method lets you connect fans in parallel, and as long as the total doesn't exceed 1Amp (you add the amps of each fan), you'll be ok. If your fan(s) can require more than 1Amp, you simply use a higher-power diode (e.g. a 6A6, which is a 6Amp silicon diode). The same rules apply for the 6Amp diode as apply to the 1Amp diode. (and of course the wires and switch must also handle the maximum current).

I just "bread-boarded" a controller that uses 1X 10-position rotary switch and 8X "1N4001" silicon diodes (1Amp capacity). So, I have ten speed settings; from "off" to "maximum" with eight speeds in between.

If you don't want 10 speeds, just use a 1X 5-position switch and tap into the diodes at just 3 interim spots for a 5-speed fan. For example, instead of tapping into after every diode (0.7v voltage drop), tap after every other diode (1.4v voltage drop).

And, using 1N4001 diodes handles a 0.9 Amp (11 Watt) fan without any problem. The diodes didn't even get noticeably warm.

Diodes are polarized so they have to be installed in the correct orientation or no current will flow. It won't damage anything...just the fan won't get any power.

For example:

+12V--->To Red Wire Of Fan

Gnd--A-|diode--|diode-B-|diode--|diode-C--|diode--|diode-D

Black Wire of Fan<------->To Point A,B,C,D,E (no connect for off)

pointA=max speed
pointB=brisk speed
pointC=medium speed
pointD=low speed
pointE=off (no connection)

Point ~VoltageToFan
A 12.0v
B 10.6v
C 9.2v
D 7.8v
E 0.0v

--|diode-- the "|" represents the cathode polarity stripe on the diode
(for more info on diodes http://www.allaboutcircuits.com/vol_3/chpt_3/1.html )


edit to add: Re: cheap fan controllers--Resistor v. MOSFET control
However, I have seen even VERY cheap rear-slot-mounted controllers that do in fact use MOSFET control circuits, and do NOT use just a resistive potentiometer to carry the main fan power. In fact, as of the date of this edit, even some of the cheap Rosewill controllers now use MOSFETs. These controllers can handle higher power fans (e.g. some handle over 30 Watts) quite well, without generating any meaningful heat. UNFORTUNATELY, they are typically VERY poorly designed and/or constructed, and the fan/board connector breaks all too easily. For example, one controller mounts the 3-pin connector to its circuit board using ONLY the 3 solder points for the pin connectors. That is structurally very, very weak. Unless one is exceptionally careful, that will not last more than a few connects/disconnects, and may fail even the first time one tried to connect a fan. I'm convinced that design flaw is responsible for most of the "intermittent" or DOA reviews on these controllers. For example, of 4 of these Rosewill controllers I bought, two of the fan connectors were already failed when I took them from the box, The damage was probably done when the controller board was tested by the manufacturer. So it worked for the test, then they broke one or more of the three solder points when they pulled their test fan's connector off the controller board's connector and they ended up shipping a bad board.


Write a Reply >>


 

    
 
 

  Topic Tools 
 
RSS UpdatesRSS Updates
 

  Related Articles 
 
 

  Newsletter 
 
A weekly newsletter featuring an editorial and a roundup of the latest articles, news and other interesting topics.

Please enter your email address below and click Subscribe.