interesting VCO was RE: [sdiy] Imperfect VCO

[Scott Gravenhorst]

ASSI <Stromeko at compuserve.de> wrote:
>On Monday 04 August 2003 17:32, Scott Gravenhorst wrote:
>> I was hoping that the
>> capacitor charge switches could be implemented using a 4069 or 4007,
>> but it seemed to me that the with a current source at the top of
>> that, we would have problems with the static protective diodes.  Am I
>> close with that? 
>
>You would need to make sure that the input is never more than about 
>0.4V above/below the supply terminals. The parts needed to ensure this 
>probably make it not worthwhile compared with other solutions (unless 
>of course you are going to use all inverters in the package and find 
>some external components that match in density).

Thanks Achim, for making me look at the internal schematic of the CD4007
yet again.

Looking at the 4007 (for about the 1000th time), two of the inverters
don't have the drain connected to the supply pin.  Since the gates won't
go higher than Vdd, I think the 4007 can be used to supply the "bridge"
transistor arrangement.  As you say, driving the top of the bridge with a
current source means Ron doesn't matter.

I have a rough schematic of a linear VCO at:
http://home1.gte.net/res0658s/CMOS_VCO.gif

There would be DC offset in the VCO output.  I haven't done the triangle
part, but it should be capable with another TL074 package.

>> If so, I suppose the cap switches could be built
>> with discrete MOS transistors or what about a 4066 to do that?
>
>A discrete MOS inverter is certainly possible, you don't even use it in 
>linear region, so no worries about matching. Zetex makes some dual 
>"logic" MOS in SMD packages that have low enough threshold voltages and 
>not too high gate capacitance. These should work well.
>
>The 4066 uses transfer gates, but again you feed it with current and so 
>the nonlinearities of the Ron are taken care of already. There is not 
>much point in trying to make a 4066 out of discrete components unless 
>you want to make a "power" version, as it's not easy to match the n- 
>and p-MOS of the switch.
>
>> Wouldn't that solve the diode problem (assuming there is one)?  If it
>> could be done with 4069, that seems to be simpler perhaps.
>
>I'm not sure if a 4069 is available as a single gate (tinyLogic or some 
>such), then it would be easier to perhaps work around the protection 
>diodes.

I think I found a way to use the 4007 for this.  For me, the roadblock
was how to make the transistor bridge such that a current source can feed
the drains.

>
>Achim.
>-- +<[ Q+ & Matrix-12 & WAVE#46 & microQkb Omega sonic heaven ]>+ --
>
>Q series MIDI Implementation & additional documentation:
>http://homepages.compuserve.de/Stromeko#WaldorfQDocs
>
>

=========================================================  
- Hydrogen is pointless without solar.
- Good or bad, that which gets rewarded, gets done.
- What good are laws that only lawyers understand?
- The media's credibility should always be questioned.
- Governments do nothing well, save collect taxes.

-- Scott Gravenhorst | LegoManiac / Lego Trains / RIS 1.5
-- Linux Rex         | RedWebMail by RedStarWare
-- FatMan: home1.gte.net/res0658s/FatMan/
-- NonFatMan: home1.gte.net/res0658s/electronics/