Inductors for buck converters (QUICK noob question)

[ericwilk]

Hi there,

Did a bit of searching here and didn't find a specific answer to this...

I see how to measure inductance but I don't have an oscilliscope and radio shack didn't have any inductance meters. I have a bunch of those round ones though I pulled off an old power supply for a printer, so I assume it's a good number. I'm only working with an amp or two here.

I guess my question is, does it really matter what the actual inductance is, or does it just need to be enough to hold the field? From my reading, it seems like the latter.




Thanks!

 

[BradtheRad]

Henry value is not critical. For a given value, it should be possible to find a suitable frequency of operation. (Faster with a smaller Henry value.)

You must pay attention to (a) saturation and (b) current rating. Thus if you want 1A output, then your inductor must handle 1.5 to 2 A, without saturating or overheating.

You can get slightly more Amperes through it if you operate it well into continuous conduction mode.

 

[ericwilk]

Thanks for the response!

This is from a PSU from a lazer printer I destroyed (well, it actually broke in the mail and I fixed it but only to last maybe 10,000-20,000 pages or so). It has a part number -- SW100 which comes up on Google but says "no info." I figure if it's a large component from a lazer printer the 1.5-2A max I'm looking to draw won't be too much.

I just didn't know if for some reason having TOO high a value might be bad. Basically, I'm trying to hook two power banks (the cell phone portable charger things) together to get twice the original current (peak and mAh). I want the device to last AT LEAST 3-4 hours. I have a 13,000 mAh one but it's too big for the chassis with all the components I'm putting in it. My original plan was to hook them in parallel with a couple diodes (to protect the cells) and a capacitor to evenly distibute the current.

I was told hooking them in series and using a buck converter with a regulator and some resistors would result in much less loss, however. I can always test this with a breadboard and meter first and simulate the amperage with resistors and maybe a potentiometer (to look at peak, spikes, etc). I just don't want to fry the board.



Thanks again!

- E

 

[BradtheRad]

A larger Henry value will generate more severe voltage spikes, when current is switched off suddenly.

A larger Henry value tends to go with a slower operating frequency. Your smoothing capacitor should be a higher value.

A slow frequency makes it okay to use an ordinary diode as your freewheeling diode.

 

[ericwilk]

A larger Henry value will generate more severe voltage spikes, when current is switched off suddenly.

A larger Henry value tends to go with a slower operating frequency. Your smoothing capacitor should be a higher value.

A slow frequency makes it okay to use an ordinary diode as your freewheeling diode.

Cool, thanks! I have a pile of stuff here, including voltage regulators and capacitors large and small. I'm sure I can come up with something.

Once I have the time >.<

I'm basically making a pocket-sized PC which can do many things like read and write to smart cards, GPS with navigation and a built-in lowjack system, communicate with similar devices via encrypted RF, mini bluetooth keyboard, music synthesys (that's all software though, only MIDI and output will be headphones or a crappy built-in sound system. I'm not sure I can make a good one and shield it enough from all th einterference), and small external LCD and keypad and electromagnetic locks and internal 7" LCD with touchscreen (which seems like it's enough to drive the OS and not look like garbage).

All the peripherals will probably draw a lot. I'll have to measure it once it is built. That will have to be after all the stuff I need to get done for work-related tasks.