Circuits with fans, should be pretty easy.

[Kirro]

I'm not an engineering major but I've got a fair amount of knowledge about circuits. I have a few questions about the diagrams below.

I am connecting two fans to 2 computer USB Ports. A single USB port puts out 5v at 500 milliamps. The fans are unfortunatly 12v each. Because of this I have decided to use two usb ports. Giving twice the voltage and twice the amps.

I am not sure if I should put the fans in parallel or series. I understand the resistance difference between parallel/series but I'm not sure what that's actually going to do to to the fans. (ie: heat them up, shorten their life) The powersource is very sensitive (a computer lol) and I don't want to mess it up.

The other question is.. I want to make sure the power is smooth to these fans so I figure i'll put a capacitor in the circuit. I'm not sure where to put it or what size to use. My voltage is pretty low so it will probably need an average sized one.

(Sorry if that post was confusing lol)

 

[Danger]

I know that there must be some reason that you chose to do it this way, but might I suggest that you just buy a couple of USB fans? They don't cost much, and they're designed to run directly from the ports.

 

[Kirro]

ah yes, the easy way out... lol.

I am an industrial design major and I am interested in developing computer peripherals.

The current project I am building calls for the circuits displayed above. Standard USB fans are meant for size and portability, not what I need them for.

I am opting for USB power because I've never done it before and it's pretty easy to do.. just.. limited, and I like a challenge.

 

[NoTime]

You are not going to get 12 volts out of a USB connector.
Unless you use a DC to DC converter.

And connecting the + of one USB port to the - of a second port
as shown in your diagram will likely damage your computer.

All the "-" pins are common ground.

 

[Danger]

You're out of my league with this stuff. Sorry... I was just going for what appeared to be the most practical approach. Next to it would be running 5VDC primary relays to switch 12VDC power from a wall-wart to the fans.

 

[berkeman]

ah yes, the easy way out... lol.

I am an industrial design major and I am interested in developing computer peripherals.

The current project I am building calls for the circuits displayed above. Standard USB fans are meant for size and portability, not what I need them for.

I am opting for USB power because I've never done it before and it's pretty easy to do.. just.. limited, and I like a challenge.

NoTime is correct. You will only be able to get 5V (unless you stack floating laptop USB ports or something). You can boost 5V to 12V, but the current ratios with the inverse of the voltage for a DC-DC converter (actually it's worse than that because they are not 100% efficient). So the 500mA that you say is available at 5V will net you around 200mA at 12V. Is that enough for your fans? Or maybe you should look a little harder for 5V fans.

 

[Kirro]

You are not going to get 12 volts out of a USB connector.
Unless you use a DC to DC converter.

And connecting the + of one USB port to the - of a second port
as shown in your diagram will likely damage your computer.

All the "-" pins are common ground.

Power
I put both + USB together and - USB together, they're not in series. I did this becuase the ports aren't going to act like batteries. If I put a 5v and a 5v in parallel I won't get 10v?

DC-DC
I'm not sure what a DC-DC converter is but i'll look around online. I'm not sure how many amps the fans will need.(how can I find that) The fans are just basic computer fans.

Fans
About the fans, I need fan and motor to rise no higher than 3/4". Standard motors that I found at radio shack and online are 3/4" to 1" not including the fan. I chose computer fans becuase they are very low profile. For example a 40mm fan has a rise of 1/2" which is perfect.

 

[Danger]

If I put a 5v and a 5v in parallel I won't get 10v?

No, you'll get 5V; putting them in series would give you 10. I don't know if you can even do that without slagging something inside the computer.

 

[chroot]

You cannot connect two USB ports "in series" to get 10V. By doing so, you are connecting the +5V pin of one port to the ground pin of another, thus completing a circuit between +5V and ground. They are not like batteries, and share a common ground. Your USB hardware almost certainly has fail-safe power electronics that will instantly disable both ports (and, if it doesn't, you'll damage the computer).

- Warren

 

[Kirro]

Okay, looks like the 12v fan idea has been squashed. I found a good fan with these specs: 3V DC & 230 Amp. I don't think this is going to work either becuase the USB puts out 500milliamp which is .5amp.If I take a 9v battery, how can I drop that down to 6v? And, how can I figure out how long that battery is going to last?

The batteries I found give out 9v @ 150mAh.

 

[chroot]

Two hundred and thirty amps? I think you must have read that incorrectly.

And the USB port does not "put out" 500mA. The USB port permits a maximum of 500mA to be drawn from it. You don't have to use it all!

- Warren

 

[Kirro]

Okay, looks like the 12v fan idea has been squashed. I found a good fan with these specs: 3V DC & 230 Amp @ .7v (This doesn't make much since becuase .7w/3v = 0.23333A NOT 230A as stated by the website or I'm doing something wrong)If I take a 9v battery, how can I drop that down to 6v? And, how can I figure out how long that battery is going to last?

The batteries I found give out 9v @ 150mAh.
==================================
EDIT:

Watt/Volt = Amps
BattAmps/Amps = battery life..

So.. first convert all the units..

Battery
9v @ .15A

Motor
3V @ .230A @ .7w

-------
And plugin the numbers

.7w/3v = .233333 Amp (not 230Amp on the website)
.15A/.23333A =.64Hours

So if I hook up this 9v battery to the motor, it will die in .64 hours? That's pretty crappy. I found (http://www.gizmology.net/batteries.htm) to calculate battery life and that says 0.7hr if I plugged numbers in correctly.

 

[chroot]

The 9V battery is not going to be able to produce enough current to satisfy the motor.

I thought you were going to use your PC's USB ports? I'm very confused as to what the application is here.

If you can so easily switch from using the computer's USB ports to using a 9V battery, why can't you just use a standard wall transformer to produce 12V for the fans you originally selected?

- Warren

 

[Kirro]

I thought you were going to use your PC's USB ports? I'm very confused as to what the application is here.

I came to the conclusion that usb ports do not put out enough power for a motor. period.

The application of this power source is a computer peripheral that is mobile and does not require power from a wall socket. It can be battery, usb, or anything that is small and lightweight. Basically something easily portable in an average/small laptop bag.

This is the motor I am getting specifications from:http://www.batteryspace.com/index.asp?PageAction=VIEWPROD&ProdID=1524

 

[chroot]

The computer's USB ports provide more than enough power to run the 3V 230mA fan you just listed.

A laptop's battery is usually around 1000 mA-h @ 12V (or sometimes 16V), for a total energy capacity of 43kJ. If your fan consumed 0.7W, or 0.7 J/s, the laptop's battery would run it for about 17 hours. (Of course, the laptop itself consumes much more power.)

- Warren

 

[Kirro]

The computer's USB ports provide more than enough power to run the 3V 230mA fan you just listed.

Alright then, the device I'm working on needs two of these motors. Can you suggest a wiring scheme? Two ports are fine if another is needed.

 

[chroot]

You could probably put two of the fans in series and run them off 5V. Each fan will see only 2.5V rather than 3V, but I doubt this would be a serious problem.

- Warren

 

[Kirro]

Sounds good, I'll order them and see how things go. If it doesn't work for some reason $10, boohoo.

Thanks for the help!

 

[berkeman]

I am an industrial design major and I am interested in developing computer peripherals.

Kirro, to be a good industrial designer in the computer industry (or elsewhere for that matter), you sometimes need to step back and ask if there are other ways to accomplish what you are trying to do. For example, having a fan in a product is usually a disadvantage (power, noise, reliability problems). You only use a fan if there are no other cooling possibilities.

So, is it a requirement of your assignment that you include a fan, or could you take care of the cooling chores in another way. With a good overall design, you can remove heat from the internals of a product and just sink it to the ambient air. You haven't really told us what you are trying to do, so I don't know what-all might be possible. Just think a little outside the box every now and then, and you'll be surprised at what a little creativity can result in.

 

[Kirro]

I wouldn't have expected such a design conscious person to respond in an electrical thread.

I am honestly not sure if the fans are needed however I do know the following about the project:

Right now I have a product that produces 135-155F of temperature. It holds this temperature for about 5-7 hours continuously. If it reaches and holds this temperature the life will be significantly shortened. Right now it gets that hot and stays that hot which is REALLY bad.

A fan may not be required for the application however I would like to have one just in case the heatsink itself gets too hot.

 

[Danger]

Have you considered Peltier devices?

 

[Kirro]

Have you considered Peltier devices?

Wow, didn't know those existed. Very interesting theory.. "The Peltier effect" Anyways, they use a rediculous amount of power and wouldn't be too effective for this application. Thanks for the refrence though! Added some knowledge to my book!

 

[Kirro]

And the results are in

The results are in... Check my circuit diagram and you can see what I ended up with.

Any thoughts or final suggestions?

Couldn't find a fuse lower than 250mA so I went with that. It may not be doing much but it makes me feel better

 

[chroot]

1) It's very likely that your regulator will have problems with only 0.5V of extra voltage to deal with. The difference between the input voltage and desired output voltage is called "dropout," and most regulators cannot deal with only 0.5V of dropout.

2) The AA batteries likely cannot withstand the current draw that the 12V fan will require to run properly. The batteries will get very hot, and their internal equivalent series resistance will limit the current that can flow through the fan, limiting its performance.

All in all, I'd say this is a pretty abhorrent circuit.

- Warren

 

[Kirro]

1)All in all, I'd say this is a pretty abhorrent circuit.

ab·hor·rent (b-hôrnt, -hr-)
adj.

1. Disgusting, loathsome, or repellent.
2. Feeling repugnance or loathing.
3. Archaic. Being strongly opposed.lmao.

I tested the regulator and it worked fine. However I learned that it drops voltage by converting it to heat. Removing heat is what I'm after so I took the regulator out and took out one battery leaving me with 11v; just under the 12v max.

Right now I've got that circuit running and I'm taking voltage readings on the batteries as well as their temperature. So far nothing bad has happened: the fan is running at 11 exactly, all batteries have nearly full charge and there is no heat.

I'll get back to you after a few more hours.

 

[Integral]

Kirro,
One would have to wonder why you are posting this here, If, when you get feedback from a working EE, you laugh at it.

The regulator is NOT removing any heat, it is GENERATING heat. It's only function in your circuit is to WASTE current from the batteries/USB by turning it into heat. Your batteries would last longer and your fan would work just as well WITHOUT the regulator. It serves no meaningful function. Fans do not require a regulated voltage, close is good enough.

When you ask for and get advise from someone knowledgeable you really should listen.

 

[Kirro]

I was laughing at the word, that's all.

And yes, after reading up on regulators, I realized they create heat. The whole project is to remove heat... that's why I took it out of the circuit. I have listened a lot to what everybody has to say, that's how I've gotten so far. Sometimes things work for me when the math is questionable.

I learn by doing and I found this circuit to work perfectly.

Results after 1hr 40min:

Heat: all components have no rise in temperature.
Batteries: dropped .25v (but I think that was due to multimeter probe error)
Fan: consistent speed

I'll be running this for another 6 hours and should have the most accurate results as well as a final conclusion.

 

[Kirro]

Image of my setup

 

[berkeman]

I am an industrial design major and I am interested in developing computer peripherals.

Kirro,

Chroot and Integral are being a bit too kind. If you presented this to me in an interview for a job, that would end the interview (not well). If you presented this to me as a project in school, you would get an F. I think it's great for an ID major to learn more about the electronic aspects of what "goes inside", but you clearly are hacking away without trying to learn the underlying details. And please don't say things like, "I've been running it for an hour and it seems to work". That's definitely not how things are designed and tested in the real world.

I think we gave you some viable alternatives earlier in this thread. There's a reason that some folks get A's on projects and others get F's. Go for the A -- learn about the electronics that will go inside, and get creative and knowledgeable about how to generate some synergy between the ID and the electronics.

 

[Kirro]

Alright, I do know that it is a very bad circuit. There are a few limitations of the project that have not been considered.

1) I am limited to power sources that portable computers have. Nearly all laptops have USB power and this gives consistant power unlike Firewire.

2) It is very difficult for me to produce a fan/motor combination that won't wobble. Using fan/motor combinations that I can find are best. ie: computer fans.

3) The fan and motor combination must have an extremely low profile (<3/4in), low power consumption, and low to medium amount of air movement.

If I only have 5v, how would you propose to generate extra voltage to bump it up to 12v? I say 12v because that's what all the fan/motor combinations that I can find require.

DC-DC Converters that I have found are super expensive and too bulky to be portable. The required budget for this project is minimal, around $30-$40 USD.

I've found a low power motor, but it is very overpriced. On top of that I then have to figure out how to securely connect a fan to it.You can't assume I made a loser circuit becuase I felt like it. There are many factors affecting why I did what I did. I havn't worked with circuits in over 7 years, I'm trying my best.

 

[Kirro]

What's actually going on

I don't think I ever posted what this was actually for. CAD renderings posted below. Copper heatsink with a rubber shell. Fans are inside pushing air through the grooves.

 

[Danger]

3) The fan and motor combination must have an extremely low profile (<3/4in), low power consumption, and low to medium amount of air movement.

You just kind of shot yourself in the foot with that one. If low-medium is all that you need, then running your 12VCD fan motor at 3-6VDC is all that you need. The USB port, therefore, is adequate.

 

[Kirro]

When I put 5v (USB) through a 12v fan it won't even spin. I also tried 6v (4x AA), that also did nothing.

I know this becuase I tried.

 

[Danger]

Hmm... that's unusual. I routinely run DC motors at reduced voltage with no problems. Yours must have a really high starting resistance.
Now that your CAD pictures are visible, I have to ask if you're sure that the design will work as you want it to. It looks to me as if it would just create a static region in the centre.

 

[berkeman]

If I only have 5v, how would you propose to generate extra voltage to bump it up to 12v? I say 12v because that's what all the fan/motor combinations that I can find require.

DC-DC Converters that I have found are super expensive and too bulky to be portable. The required budget for this project is minimal, around $30-$40 USD.

http://www.national.com/an/AN/AN-1066.pdf#page=1

You design your own boost converter for a couple of dollars, and include that in the overall design. (I think I mentioned this already, but it could have been in a different thread...) A DC-DC converter is basically a constant power converter, with some efficiency loss. So the input current will be more for your boost application than the output current. That should be part of your calculations for the current demand that you will place on the USB port.

 

[chroot]

National Semiconductor also has a (relatively) powerful online power supply tool called WEBENCH:

http://webench.national.com/appinfo/power/webench/index.cgi

- Warren

 

[Kirro]

National Semiconductor also has a (relatively) powerful online power supply tool called WEBENCH

Interesting stuff here... I was checking out the LM3224:

InputMin Voltage 2.7 Volt
InputMax Voltage 7 Volt
Adjustable Output Yes
Output Range Vin to 20V

I'm guessing that's the sort of DC-DC converter I'm looking for.

Looks good on the search page however when I add it to my order it reads $4.95 for the part, $5 for the handling, and then $2 for the shipping? Is that possible? I'm really even sure that is the same part that I selected on the search page because the minimum order is 1000 units. (lol)

Similarly, http://www.national.com/pf//LM/LM2588.html also looks good, however.. at the bottom of the page is 24 different variations of the same chip. I know I want a 12v one but there are 3 of those.As a side note, how big are these things? What do they look like? There aren't any real pictures on the website other than the circuits.

 

[Kirro]

http://www.national.com/an/AN/AN-1066.pdf#page=1

You design your own boost converter for a couple of dollars, and include that in the overall design. (I think I mentioned this already, but it could have been in a different thread...) A DC-DC converter is basically a constant power converter, with some efficiency loss. So the input current will be more for your boost application than the output current. That should be part of your calculations for the current demand that you will place on the USB port.

hmm, I read through that PDF and found it pretty interesting. However, I don't think I'd be able to make a working circuit after reading through that.


[sorry for double post]

 

[chroot]

Looks good on the search page however when I add it to my order it reads $4.95 for the part, $5 for the handling, and then $2 for the shipping? Is that possible? I'm really even sure that is the same part that I selected on the search page because the minimum order is 1000 units. (lol)

You're not going to want to order directly from the manufacturer. The manufacturer only deals directly with big contracts, like Sony. National Semiconductor is not interested in dealing with you directly unless you're buying a significant number of devices. For small orders, you want to go through a distributor, like DigiKey or Arrow or Newark.

DigiKey, for example, will sell you a single unit for a price of $4.53.

As a side note, how big are these things? What do they look like? There aren't any real pictures on the website other than the circuits.

Look at the datasheet. All datasheets will include a package mechanical drawing, often on the last page. The LM3224, for example, is an 8-lead surface mount package. (You might find these difficult to solder, unfortunately. They're meant to be soldered by automatic machinery.)

- Warren

 

[NoTime]

Hmm... that's unusual. I routinely run DC motors at reduced voltage with no problems. Yours must have a really high starting resistance.

Likely a brushless DC fan those aren't quite as flexible as straight motor.

 

[NoTime]

hmm, I read through that PDF and found it pretty interesting. However, I don't think I'd be able to make a working circuit after reading through that.

As long as you are thinking of ordering parts.
Why not just get a sutible fan instead.
Here is an example
http://rocky.digikey.com/scripts/ProductInfo.dll?Site=US&V=259&M=KDE0504PFS2-8
Lots of other choices

5vdc brushless <.200 ma
And less than $7 plus whatever they want for shipping.

Also you might want to do some calcs on just what you really need for airflow.

 

[Danger]

Likely a brushless DC fan those aren't quite as flexible as straight motor.

Ahhh... I don't have any experience with those. Good point.

 

[Kirro]

Sorry, I stopped getting notified of new posts for some reason.

NoTime, those fans are PERFECT.:!) :!)

As far as calculating airflow goes.. I'm not really sure how much airflow I'm going to need to cool off the heatsink. I am fairly familiar with about how much air a certain rpm will give with various types of fan blades. I would estimate I need something around 4K-5K rpm to achieve my goal with minimal noise.

Thanks very much for the link!

 

[Kirro]

Every site I've found that carry these fans have minimum orders of 200+ or handling fees of $15 + shipping.

 

[NoTime]

Every site I've found that carry these fans have minimum orders of 200+ or handling fees of $15 + shipping.

?? Where are you looking ??
Digikey is only $5 handling for orders < $25.
$0 if > $25. And the will sell you just 1.

Search the catlog there is probably something else you need to actually finish the project.
Or just order 4 fans

 

[Kirro]

... You're exactly right. I'm not sure what I was doing before...

I checked out other items they have and I was very excited to see some of their other products! They've got shrink wrap tubing and EVEN... USB A Male to open end wire. I can't even believe it...

I'll be ordering up some fans, usb cables, shrink tubing and whatever else I can find in their catalog soon!

Thanks for all the help!