Content-Transfer-Encoding: 8bit
Message-ID: <3.0.1.16.20011213095456.2d3f78a0@pb623250.kuleuven.be>
Date: Thu, 13 Dec 2001 09:54:56
To: rsgb_lf_group@blacksheep.org
From: "Rik Strobbe" <rik.strobbe@fys.kuleuven.ac.be>
Subject: Re: LF: Chips
In-reply-to: <3C17885E.8063C550@netscapeonline.co.uk>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii; format=flowed
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

At 16:39 12/12/01 +0000, you wrote:
>Hello All
>There are numerious chips available for dividing frequency but what is
>available for multiplying, eg.

Hi Mal,

Frequency multiplying is a complete different story than frequency
dividing. While the last is done digital, multiplying is a non-linear
analog thing.
If you want to multiply a signal source that doesn't change is frequency
(or has only a minimal change) you can do it the classic way (class C amp
followed by a resonant circuit on the desired harmonic). If you want to use
a 'black box' there are some frequency doublers ,eg. by MiniCircuits, that
perform frequency doubling or alternatively you could use any double
balanced mixer where you would actually mix the frequency by itself. Using
an active DBM - such as MC1496, SO42P, NE612 - will give a pretty good
efficiency and surpression of the source frequency. But additional
filtering is still recommended (also with the MiniCircuit stuff).
Anyway, for you purpose (x8) you would need a chain of 3 doublers and have
to handle a lot of unwanted signals.
An alternative path might be a PLL, at the frequencies you mention a
CMOS4046 should do fine. With a PLL you won't suffer from any
'sub-harmonics' but a proper design is needed to keep phasenoise within
acceptable limits.
Using a PLL system you wouldn't need more than 2 CMOS IC's, a 4046 PLL and
a divide-by-8 IC (many possibilities, 4018, 4020, 4022, 4024, 4029, 4040,
4516, 4520 etc...).

73, Rik  ON7YD