AVR-Chat

Hi, Roy
 
	Thanks for the prompt answer..

	Nice part ! Cheap and quite easy to use. I think I got a simpler
option for my project but I will keep it for future projects.

	The bandgap reference was not my main concern but the effect of fast
fluctuation on the supply voltage. It is "noise" that may induce into the
measurements in many unknown ways. The more I think about it the less I like
it. 
 
	I am finishing some worst case calculations but I will probably go
with a 2 stage shunt regulator. I will post the details afterwards.
 
Best regards,
Alexandre Guimaraes


-----Mensagem original-----
De: AVR-Chat@yahoogroups.com [mailto:AVR-Chat@yahoogroups.com] Em nome de
Roy E. Burrage
Enviada em: 12 October 2007 22:14
Para: AVR-Chat@yahoogroups.com
Assunto: Re: [AVR-Chat] Power supply dynamic range.

Talk about unregulated supply voltages...

Have you looked at a switchmode regulator like the MC34063, Alexandre?  
Some, like the '063, will withstand up to 50 volt supply voltages and won't
cost a lot to build...perhaps a couple of bucks for parts and some assembly
time, but it depends upon how critical your circuit is.  You could set your
regulator output for something less than your minimum input voltage, like 4
volts, and then you'd be good to go.  Build up a prototype and check to see
how it behaves.

Doesn't the bandgap reference get fed by a constant current source?  
That being the case, the supply voltage shouldn't make that much difference
so long as it's kept within the parameters specified for your controller.
Look at the reference variation with supply voltage specification in your
datasheet.  It should tell you if you can maintain
8 bit accuracy.


REB


Alexandre Guimaraes wrote:

> 
>Hi,
> 
>	I have an application where a simple and cheap zener regulator with 
>emitter follower would be my best ( cheap) option for powering a 
>Atmega168 with the AD converter using the internal 1.1 volts reference. 
>As the input voltage range is way to big, from about 5 volts up to 40 
>volts, the regulation will not be too good because the zener current 
>will vary too much.
> 
>	I did not find any references if I will get into trouble varying the

>supply at the AVR in the range of 3 to 5 volts. Does anyone have 
>references or opnions about this ? Would the AD suffer from those 
>variations in supply voltage even using the internal bandgap as a 
>reference ? I just need 8 bit's out of the AD...
> 
>	I would appreciate any insight. I always try to make the power
supply 
>as stable as possible and letting it out to the microcontroller with 
>such a big variation feels really weird to me... But I looked around 
>all I could and did not find anywhere saying that I could not do it 
>with a micro that is specified for this full range !
> 
>	Thanks in advance...
> 
>Best regards,
>Alexandre Guimaraes
>  
>


[Non-text portions of this message have been removed]



 
Yahoo! Groups Links