Go Down

### Topic: zener diode to power an attiny? (Read 2970 times)previous topic - next topic

#### stevethatsmyname

##### May 28, 2011, 11:03 pmLast Edit: May 28, 2011, 11:24 pm by stevethatsmyname Reason: 1
Hi all,

I'm planning a project to make a "Choker" (i.e. wearable) using the 12V RGB LED strips.  I was planning on using a small 12V battery to power it.

My current plan uses a zener to drop 12V to 3.3V for the microcontroller, so that I can use the same battery for both the uC and the LED's.

Here is my planned overall circuit:

#### stevethatsmyname

#1
##### May 28, 2011, 11:07 pmLast Edit: May 28, 2011, 11:19 pm by stevethatsmyname Reason: 1
And here was just the power schematic

my main questions are
1) will it work?
2) with the way I have it laid out, with the zener in parallel with the Microcontroller, the current through R1 will be approximately constant, correct?  And it will depend on the value that I actually pick for R1.  Current = (12V- 3.3V)/R1
3) Any way I can calculate the approximate max current draw through the microcontroller, in order to figure out the ideal R1 value? I'd like to not waste so much power through the zener, as the 12Vbattery only stores 55mA-h.
4) Does the max impedence of the zener diode matter so much as long as it is greater than what I need for the voltage divider to work?
5) Am I just better off powering the uC using a separate battery?

#### RuggedCircuits

#2
##### May 28, 2011, 11:10 pm
You've got pins 4 and 8 reversed. Pin 4 is ground and pin 8 is VCC. Also, all your LED's are reversed.

--
The Gadget Shield: accelerometer, RGB LED, IR transmit/receive, speaker, microphone, light sensor, potentiometer, pushbuttons

#### stevethatsmyname

#3
##### May 28, 2011, 11:26 pm

You've got pins 4 and 8 reversed. Pin 4 is ground and pin 8 is VCC. Also, all your LED's are reversed.

--
The Gadget Shield: accelerometer, RGB LED, IR transmit/receive, speaker, microphone, light sensor, potentiometer, pushbuttons

Thanks for that, I was looking at a physical circuit I used as a reference upside down, which is why I had 4 and 8 reversed.

I fixed the schematics... I updated using the fritzing (RGB LED) symbol but I really mean the RGBstrip.  I mainly want to know whether using a zener to drop 12V to 3.3V is worth the effort and current loss, or should I just use a separate ~3V power supply for the uC.

#### RuggedCircuits

#4
##### May 28, 2011, 11:31 pm
If you've only got a 55mAh battery the shunt regulator you're building will be pretty wasteful. I'd go with a 3.3V regulator if you really want 3.3V (any reason you don't want 5V?) You can get a standard 3.3V regulator (LM1117 and others) or try to find one with lower operating current to get a bit more battery life.

--
The Flexible MIDI Shield: MIDI IN/OUT, stacking headers, your choice of I/O pins

#### stevethatsmyname

#5
##### May 29, 2011, 12:22 am

If you've only got a 55mAh battery the shunt regulator you're building will be pretty wasteful. I'd go with a 3.3V regulator if you really want 3.3V (any reason you don't want 5V?) You can get a standard 3.3V regulator (LM1117 and others) or try to find one with lower operating current to get a bit more battery life.

--
The Flexible MIDI Shield: MIDI IN/OUT, stacking headers, your choice of I/O pins

I was mainly thinking zener due to space savings... although if I can get a 3.3V or 5V regulator in a small package (it only has to power the microcontroller so it wont draw much current) that would certainly do.

All regulators i've seen so far come in really large packages (which makes sense for medium to high current draw applications).  If there's a smaller one you think would work well for this purpose I'm all ears.

#### RuggedCircuits

#6
##### May 29, 2011, 12:35 am
Small surface-mount package or small through-hole package?

--
Beat707: MIDI drum machine / sequencer / groove-box for Arduino

#### floresta

#7
##### May 29, 2011, 02:19 am
How about some resistors to limit the base current ?

Don

#### stevethatsmyname

#8
##### May 29, 2011, 05:47 am

How about some resistors to limit the base current ?

Don

good idea.  the ones i am using don't have this.  I included it in my 'real' prototype circuit but forgot to put it in schematic

Quote
Small surface-mount package or small through-hole package?

I am probably preferring through-hole right now.  I found some on mouser between 100mA and 300mA in the TO-92 package which ought to be small enough.

#### Grumpy_Mike

#9
##### May 29, 2011, 06:01 am
Going from 12V to 3V3 means that the regulator has to burn a lot of power. Therefore it will get hot. Make sure you don't exceed the power rating of your regulator. You can be within the current ratings but exceed the power ratings very easily on most devices.

See:-
http://www.thebox.myzen.co.uk/Tutorial/Power.html

#### stevethatsmyname

#10
##### May 29, 2011, 07:04 am

Going from 12V to 3V3 means that the regulator has to burn a lot of power. Therefore it will get hot. Make sure you don't exceed the power rating of your regulator. You can be within the current ratings but exceed the power ratings very easily on most devices.

See:-
http://www.thebox.myzen.co.uk/Tutorial/Power.html

good advice.  I am expecting a fairly low current (I will be clocking the tiny at 1mHZ)... lets say 30mA max? at 7V drop that is 210mW, the datasheet for one of the Dinky 250mA LDO regulators says around 200degrees per watt.  So about 40 degrees above ambient, which is 60 degrees C.  That's a little toasty but still within the device specs. I'll probably have to double check the current draw.

#### pwillard

#11
##### May 29, 2011, 01:22 pm
Mike is not talking about current.  He's talking about what happens when a regulator must make the voltage difference between 12 and 3.3 "go away".  It becomes HEAT.

#### stevethatsmyname

#12
##### May 29, 2011, 08:01 pm

Mike is not talking about current.  He's talking about what happens when a regulator must make the voltage difference between 12 and 3.3 "go away".  It becomes HEAT.

hi,

according to the datasheet the power dissipated = the voltage drop times current.  And the temperature depends on the power dissipated. This means that is is, indeed proportional to the current, and also proportional to the voltage drop.  From 12V to 5V is a 7V drop.  Times ~30mA gives me 210mW power dissipated in the device.

Also the devices datasheets have general formulas for thermal properties.  The TO-92 packages seem to be around 130-200 degrees Celcius / watt.  divide that by 5 for  210mW... this gives 26 - 40 degrees C (this is measured as degrees above ambient).  Ambient is 25C so 50C - 65C.  As I said, pretty toasty, but within device specs.

Go Up

Please enter a valid email to subscribe