PDA

View Full Version : Switching ON/OFF a 3.3V Regulator with propeller



DavidM
01-15-2008, 09:11 AM
Hi,

I need to switch a 3.3V regulator (LM2937-3.3) ON or OFF via the propeller to power a AEROCOMM RADIO MODULE ( AC4490 )
My regulator currently runs from the 5V supply of my board.

Which is more suitable, a transistor or a mosfet?



Thanks

Dave M

Phil Pilgrim (PhiPi)
01-15-2008, 09:18 AM
Instead of the LM2937, use a different regulator with an enable pin. They can be found on DigiKey by choosing a package with more than three pins (e.g. TO220-5). If that's not possible, a PMOS-FET with an open collector NPN transistor, driven by the Prop pin, to pull the gate down will also work.

-Phil

DavidM
01-15-2008, 09:36 AM
Hi Phil,


Your idea of a regulator with enable pin is a good idea, it will take up less space!

I now just have to choose one from gazillions available!


regards


Dave M

Dgswaner
01-15-2008, 12:19 PM
at least for the few regulators with an enable I looked at. the mA rating was 2-300mA Not the 1A of the LM29 series. so make sure you get one big enough to meet the "650 mA typical" requirement of the radio module.....

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"A complex design is the sign of an inferior designer." - Jamie Hyneman, Myth Buster

DGSwaner

DavidM
01-15-2008, 04:08 PM
I Found these two..

LP3875ET-3.3 1.5A - NATIONAL SEMICONDUCTOR — IC, REGULATOR LDO 3.3V

LP3964ET-3.3 — 0.8A - NATIONAL SEMICONDUCTOR — IC, REGULATOR ULTRA-LDO +3.3V

Are these overkill, I think they are expensive!

thanks

Dave M

Nick Mueller
01-15-2008, 09:52 PM
> Are these overkill, I think they are expensive!

Bugetary price is $1.20, so you can expect a street-price of $6 for low quantities.


Nick

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Never use force, just go for a bigger hammer!

The DIY Digital-Readout for mills, lathes etc.:
YADRO (http://www.yadro.de)

ohararp
01-16-2008, 01:12 AM
In this case I would use the FDN304P. You can switch 2.4A with this tiny sot-3 part. I am switching 3.3V in the schematic put you can use this to source 5V without a problem. Because this has a 1.8V gate it is no problem to drive it with a pin from the propeller running at 3.3V. Just remember you have to pull this pin "low" for current to flow to your device.

Datasheet:
http://www.fairchildsemi.com/ds/FD/FDN304P.pdf

Mouser: ($.55)
http://www.mouser.com/Search/ProductDetail.aspx?qs=MA%252b40jw2YMpyjgZQyVD8yA%3 d%3d

You can see a schematic for how I use this via the attachment:


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Regards,


Ryan O'Hara
www.ohararp.com (http://www.ohararp.com)

Post Edited (ohararp) : 1/15/2008 6:20:39 PM GMT

SteveW
01-16-2008, 03:32 AM
>In this case I would use the FDN304P. You can switch 2.4A with this tiny sot-3 part. I am switching 3.3V in the schematic put you can use this to source 5V without a problem. Because this has a 1.8V gate it is no problem to drive it with a pin from the propeller running at 3.3V. Just remember you have to pull this pin "low" for current to flow to your device.

I'd be very tempted to fit a pullup resistor to the gate, though. A half-turned-on SOT23 will desolder itself fast, if your output pin is floating for a while.
Even with the gate driven hard, I^2R losses in the FET are getting close to the thermal limits, if you're really hoping to switch the full 2.4A. Careful PCB layout would be called for.

It would be brave indeed to switch a 5V rail with a 3.3V IO pin from a propeller straight to the gate. Even when 'off', you'll have a Vgs of 2 volts or so. That's mostly on. Expect crispy smells. A second transistor stage to get the levels, and defaults, safe would be advisable, wouldn't it?

Steve

Phil Pilgrim (PhiPi)
01-16-2008, 03:41 AM
Or you could switch the output side of the 3.3V regulator, downstream of the filter cap. Most modern regulators have a low enough quiescent current that letting the regulator run without a load would add little to your power drain. Just be sure to use a PMOS-FET with an RDS(on) in the milliohm range at a -3.3V VGS to ensure a minimal post-regulation voltage drop.

-Phil

Post Edited (Phil Pilgrim (PhiPi)) : 1/15/2008 8:46:46 PM GMT

SteveW
01-16-2008, 03:47 AM
Yeah, that'd be safe. I'd still be tempted by an LDO with an enable pin, though, unless I was building something with some serious sales volume. I like easy...

Steve

DavidM
01-16-2008, 01:15 PM
Hi,

Thanks guys for all the tips,

I think a regulator with enable pin is they way to go, As I can easily add this to my existing layout.

regards


Dave M.

mahjongg
01-17-2008, 07:55 PM
Texas instruments REG104, extremely low dropout, 1A, SOT223-5 package, and on/off input pin. REG104-A = adjustable.

Drone
01-17-2008, 10:25 PM
The REG104 is a nice part - been around for a few years & I've looked at it before. But the SOT223 lead pitch is 0.05", not 0.1". The 230mV drop out for all but the 5V part is nice (5V is 320mV), max input is 15V. These parts do not require an output capacitor. Couldn't find a unit price on the 3.3V part but the -A adjustable from Newark in unit quantity is $4.53 USD ea. (ouch!), TI budgetary 1k price for the 3.3V part is $2.35 USD (again-ouch!). For experimenters, aside from the 0.05 lead pitch is the difficulty in heat-sinking this device. Perhaps a TO220 "clip" heat sink can "capture" this device and then you can solder some leads to the pins to "break-out" to a 0.1" plug-proto-board. Hmm...

Regards, David

DavidM
01-24-2008, 06:41 AM
Hi,,

I found this device FAIRCHILD KA378R33

http://www.mouser.com/Search/ProductDetail.aspx?qs=RAZo5F3HLrnWbTu8pZ2dtA%3d%3d

Is it just a matter of connecting the enable pin to one pin on the PROPELLER , with a pulldown resistor to ground, so on start-up this regulator is OFF, and when I choose to, I ENABLE it via HI on the prop pin?

I guess I only need 1 output CAP as usual, I will be running this from a regulated 5V device which is about 8cm away from it . So I hope I don't need a HEAT-SINK as well.


What do you think?

thanks

Dave M