Terry Given
Guest
|
 Posted:
Wed Feb 23, 2005 5:16 am Post subject:
Re: Gyrator paper |
|
|
Jim Thompson wrote:
| Quote: | On Fri, 11 Feb 2005 22:01:42 -0800, qrk <SpamTrap@reson.com> wrote:
On Fri, 11 Feb 2005 16:15:58 +0100, martin griffith
martingriffithX@Xyahoo.co.uk> wrote:
Just found this, (I got sidetracked !)
http://www.bbc.co.uk/rd/pubs/whp/whp093.shtml
Its got loads of 2*PI*F's in it so it must be correct.
Also on page 4of the pdf:
I'm sure glad I didn't use a patented circuit that screws up the
performance of the gyrator. I think these guys missed an obvious
solution. Referring to their Figure 2.1, inductance is:
L = (C1*R1*R3*R4)/R2
If you place equal valued compensation capacitors (Cc) across R2 and
R3 and make R2 = R3, the net effect of the capacitors cancels out,
i.e.
(Cc || R2)/(Cc || R3) = 1.
Adding capacitance across R2 and R3 will help stabilize the circuit
without screwing up the Q.
Nice student paper, though.
Well, In Brit-land it garnered a patent ;-)
Perhaps I should take a patent out on this trivial concept.
Mark
Sounds good to me !-)
...Jim Thompson
|
Hi Guys,
hilarious paper. It could be summarised thusly:
1) I built a bog-standard gyrator, using whizz-bang opamps I pretended
were ideal
2) it didnt work - the opamps werent ideal
3) I carefully didnt bother analysing the circuit properly (having never
read Jiri Dostals book) and am uncomfortable with laplace transforms, so
fluffed around with some baby maths before coming to the obvious
conclusion of slapping a cap across Rf to tame the oscialltions caused
by opamp/stray capacitance (ie the industry standard solution)
4) by studiously paying very little attention, I missed the blindingly
obvious symmetrical approach kindly pointed out by Mark, which magically
happens to solve the problem I created with my assymetric compensation.
I shoulda looked harder at the maths
5) them nice folks at the pommy patent office let me patent it anyway,
probably because they a) had no idea what the hell I was talking about,
and b) are trying to emulate the USPTO in patenting any old nonsense
(the older the better)
OTOH the frequency response plots are very nice. Just goes to show a
$50,000 network anlayser doesnt compensate for a profound lack of skill.
Cheers
Terry
|
|
FAQ
Memberlist
Register
Profile
Log in to check your private messages
Log in