[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Moon-Net] RE: [linrad] Local Osc. for Linrad rec.

-----Original Message-----
From: wb9uwa@xxxxxxxxxxx [mailto:wb9uwa@xxxxxxxxxxx]
Sent: Friday, December 13, 2002 10:52 AM
To: linrad@xxxxxxxxxxxxxxxxxxxxx
Subject: [linrad] Local Osc. for Linrad rec.

Hi All,

I am developing the hardware that I hope to use on Linrad for 2M EME receive.
I am using a lot of parts on hand to keep down cost.
My questions at this time only relate to the local oscillator.
The basic plan is to receive 144 mhz using a 10.7mhz IF and a 154.7 mhz (high side injection),
local oscillator. I have a 5 pole filter tuned to 154.7 mhz that is very selective and should
provide a spur free local oscillator signal. The problem is to provide a quiet, stable LO.
My plan is to use a 13.667222 MHZ crystal and multiply it 9 times to produce 123.005 mhz.
This crystal and multiplier is actually part of a Motorola Micor, low band receiver designed to
receive at 35.74 MHZ. The crystal is temperature compensated and should be of good quality.

The 123.005 mhz output is to be added (mixed) with a 31.695 mhz (programmable) source that
is to be created by a modified CB radio. Internally to the CB, a PLL is programmable with 9
binary bits, so quite a wide frequency coverage range is possible and in as small as 10khz
frequency steps. For this arrangement it is to operate at 2.21 mhz and will not be multiplied,
so any bad aspects of the PLL will be minimized. In its modified configuration, the 2.21 mhz
PLL (variable, but 144.0 mhz RX) is added (mixed) with the output from a 29.485 mhz crystal
oscillator. This will produce an output of 31.695 MHZ from the CB.

In case this is as clear as mud, the formula is as follows:
To determine RX frequency:
(2.21 + 29.485) + (13.667222 X 9) - 10.7 = 144 MHZ.

The 31.695 mhz output can contain a spur at 29.485 - 2.21 = 27.275 MHZ, so I am providing
a 5 pole filter for the 31.695 mhz signal to pass through. This is actually the filters used
in the Motorola Micor low band RX front end. The FET mixer in the Micor will provide the
31.695 MHZ + 123.005 MHZ mix. Of course all filters normally at the output of the Micor mixer
set to 5.26 MHZ are removed, so it can output at 154.7 mhz. This is to be the LO for
the final mixer that I do not yet have.

For now, the plan is to use IF filters at 10.7 mhz and have limited, (but Xpol) use of Linrad.
For later, I may convert 10.7 down to 2.5 mhz for a full bandwidth implementation of Linrad.

Now that I have described my proposed (but partially built) system, I have a few questions.

1) Given that the output from the CB at 31.695 mhz may have some spurs, is it best to
limit the amount of drive to the mixer to provide for better selectivity or is it better to
badly overdrive the mixer.

2) Currently the 123.005 MHZ oscillator/multiplier chain only triples to 41,001667 MHZ.
This 41.001667 MHZ signal is fed into the mixer with the expectation that the 41.001667 MHZ
signal will triple to 123.005 mhz inside the mixer. This is how the X9 multiplication is
accomplished in the 155 mhz VHF Micors. Is this an acceptable solution or would
my LO be more pure if I were to interupt the signal path between the 41.001667 MHZ
signal and the mixer to triple it first to 123.005 MHZ and provide some selectivity?

3) Is high side frequency injection a problem for Linrad? The frequencies will be backwards.

73, Jim Shaffer, WB9UWA.

Moon-Net posting and subscription instructions are at http://www.nlsa.com/nets/moon-net-help.html