[sdiy] =?ISO-8859-1?B?UmU6IFLpcCA6IFtzZA==?=iy] Hmmm...OK NOW you can panic :(

[jhaible at debitel.net]

There's no need to match more than all corresponding pairs in an 
OTA. No need to match the mirror transitors to the input pair etc.
As long as you can buy npn and pnp pairs, everything's fine.

JH.

(currently preparing experiments with a 6BC8-based VCA (;->) )



> Hi
> 
> Long ago, before CA3080 were available, people were using CA3096 (2 
> PNP+3 NPN) transistor arrays
> to build OTAs. A good example is the paper by Sergio Franco, Electronic 
> Design, september 1976,
> that you can download at my site :
> http://www-timc.imag.fr/Yves.Usson/personnel/SDIY/archives/index.html
> 
> The problem is that the CA3096 is even more obsolete than the CA3080.
> The good new is that Intersil sells a very good substitute ()
> for it, with improved performance, which is referenced HFA3096 and
> is available at a price of $2.00 per unit.
> OK, this is fine for american citizens, but not for us european 
> citizens, since these
> have been classified by the US department of defense we must fill in an 
> export compliance
> form in wich we have to certify that we will not use these chips to 
> build evil devices
> and that kind of stuff !... Well I promise I'll a good boy Uncle Sam, 
> can I buy
> these chips now ? :-)
> 
> Cheers
> 
> Le mercredi, 29 sep 2004, à 09:48 Europe/Paris, mark verbos a écrit :
> 
> > Tom G's method was simply a pair of NPNs, a pair of PNPs and an extra 
> > NPN at the bottom for the control path. The single NPN can be removed, 
> > as in Don and Ian's oscillators, allowing an NPN pair for expo 
> > control. It doesn't seem too bad to end up using 6 transistors insted 
> > of an 8 pin chip and a matched pair. They do need matching tho, so 
> > that makes it a little harder. You could use 3 matched pairs. How does 
> > the noise spec compare though (not that the 3080 was a shining example 
> > of how to be quiet)?
> >
> 
> -------------------------------------------------------------
> Yves Usson
> 
> 
> email : yves.usson at imag.fr, yves.usson at ujf-grenoble.fr
> 
> 
> 




-------------------------------------------------
debitel.net Webmail