[sdiy] Run standard CMOS on 3.3 volts?

[Roy J. Tellason]

On Wednesday 11 August 2004 07:13 pm, WeAreAs1 at aol.com wrote:
> In a message dated 8/11/04 1:49:57 PM, rtarcan at superonline.com writes:
>
> << Yes it works. But the  nonlinearity of the on resistor (the resistor
> when the 4066 swiched on)and the on resistor value increases...
> increases. It best works on 15 volt. (low on resistor and nonlinearity)
> Cheers
> . .
>
>
> Hello guys,
>
> Thank you for your responses.  I'm glad that the old standard part will
> work. I'm trying so very hard to keep from ever having to use surface mount
> parts. So far, so good.  I'm hoping that the ON resistance at that low
> power supply voltage will be low enough for my application (switching
> digital signals from a front panel switch matrix to simulate button
> presses).  If not, I will simply parallel two or more 4066 switch sections,
> which should effectively lower the total ON resistance.  Since I'm just
> gating some low-to-medium speed digital signals, I don't think linearity of
> the switch element is going to be an issue.
>
> I am kind of surprised that none of the data sheets I found for the 4066 or
> 4016 specified a minimum operating supply voltage.

I don't know what you're looking at,  but I have a bunch of data books here 
dealing with CMOS that go back quite a ways,  some to the real early days of 
these parts. And almost *all* of them specify a range of supply voltages as 
being usuable,  most often 3-15 volts.  You may notice that the backup 
battery commonly used for CMOS computer setup chips and other memory backup 
applications are usually 3V coin cells these days,  though I've also seen 
somewhat higher voltages used -- that's not a coincidence.

> They do give various measurements and parameters at 5 volts,

Probably because that was the most common logic supply voltage around.

> but they don't mention that the chip can be used at lower voltages.  Maybe
> it's because the data sheets were created back before people started
> designing stuff around 3.3 volt supplies?

Don't think so.

> Or maybe.... it's because they just want us to purchase the specialized
> (more expensive) 3.3 volt parts?

I'm not even aware of these.  <shrug>

> Well, I'm not going to be tricked by them.  When I come out with my new line 
> of cell phone/PDA's, they're going to be built with all through-hole parts, 
> discrete logic, and the polyphonic ring tone generator is going to be pure 
> analog -- probably something like a 16-voice ASM-1, except with more
> modulation routing options.  Stay tuned.

That gonna  come with a back-pack to carry the batteries in too?  :-)