[Date Prev][Date Next][Thread Prev][Thread Next] - [Date Index][Thread Index][Author Index]

Re: W0FMS mods (LONG)


Okay, we can do this to the list.

First, thanks for the detailed reply. Much has happened since I sent my
question to you:

* Found that the Motorola power meter my friend gave me is flaky. Might
not even be worth fixing. This was one of the major contributors to my
tunning difficulties. I moved the test setup over to my old faithful
strip line remote head directional coupler and indicator I built back
in Feb of '89. 

* Went to http://www.g6lvb.com/images/patch003.gif and gained an
understanding of how my patch should behave when tunned.

It became clear that I needed to be more granular in my adjustments and
measurements. My friend Les N7CMJ came over and we set up the range so
that two of us could work comfortably. I ran the computer and tunned
the radio through 7 frequency steps: 1241, 1250, 1259, 1268, 1277, 1286
and 1295 loaded in memory. Les made the adjustments to the patch and
took the SWR readings. I then entered the readings into a spreadsheet
to save the data and plotted each "sweep" on a line graph. Each sweep
was plotted against the previous two sweeps, giving us a view to the
trend to our work, making our progress much faster.

In effect we were a "meat loop" network analyzer.

We found out that the adjustments are very critical and the standard
change we settled on was 1/8 turn of a 1/4-20 screw. We took a huge
amount of data but, it paid off. The specific dimensions/adjustments

	Patch reflector and patch diameters were to Robert's spec
		+/-0.010 inch

	Patch to reflector spacing 0.480 inch

	Phasing screw out (from short) 1.375 turns.

	Frequency screw out 2 turns.

The plot from above (given appropriate scaling for Q, yada, yada),
could be overlaid on the G6LVB plot and no daylight show through!

There is no substitute for good "bench" work on this stuff. One
would think that having spent over 40 years making a good living
by that rule I would have applied it earlier.

Specific comments, in-line below:

On Thu, Aug 21, 2003 at 09:59:24PM +0000, Frederick M. Spinner wrote:
> HMM.. this will be hard to answer.   First of all, your S-band patch should 
> be fine.  You might be able to get it a little better with a Network 
> analyzer, but I think that you are more than close enough to perfect.  
> Tuning it that way is completely valid.

I agree.

> I have to be honest with you.  I've been way too busy to play much with 
> anything for the last year or so, and even though I have a W0LMD L/S patch 
> 80% built, well, it is only 80% built.
> So these comments will be from my experience with the S patch.
> The 1.2:1 to 1.05:1 to 1.2:1 were only guesstamates based on the values the 

Is this not reverse, 1.05:1 to 1.2:1 to 1.05:1?

> network analyzer gave me when I switched from "Smith chart" mode to VSWR.  
> Actually these are extreme.  I should take the time to see what frequency 
> 50-j50 at 2401 translates into 50+j0 ohms at and then it would tell us 
> exactly what the "W" dips frequencies should be.  This is actually more 
> important than the SWR value... the frequencies where the dips are.

Given my resources SWR will have to make do.

> Hmm.. I really should go through and figure that out.  The algebra is a 
> nightmare, so I'll have to do this when I have some time.
> What I'm saying is that you might be fine on the circularity of the 1268 
> patch.  If not, to increase the "depth of the dip" you need to increase the 
> capacitance of the circularity screw to make it larger.  This should then 
> give you a deeper dip.  I really am going to need to figure out where those 
> dips should be.  Ideally you want to see a 45 degree shift with the 
> capacitor in one direction, and the geometry to the center screw apparently 
> gives a reflection that produces an shift 90 degrees out of phase with the 
> capacitance screw.  This makes the second resonance mode of the patch +45, 
> again giving a 90 degree phase shift, and due to the physical alignment of 
> all of it it becomes LHCP (for refelecting to RHCP)... On a network 
> analyzer the ideal situation is when you are turning from a dip in the 
> Smith chart to a loop.  Right at that point where the loop forms is your 
> "ideal" CP point.  This doesn't help much on just an SWR bridge since the 
> "scalarization" loses that info.
> So the trick is getting the two resonance modes equidistant from the 
> desired operating frequency of 2401 MHz and then getting those resonances 
> in the proper places.  I never calculated it out on the S-patch because I 
> was given the data from the G3RUH patch of where those dips should have 
> been..
> Plus I confirmed it with the Network analyzer with that dip/loop trick.

Exactly, when I finally increased the granularity of my adjustments and
measurements, the symmetry of the dips began to show up in the excel graphs
and I new I would be able to close on tunning the system.

> However, you should get some dip on your patch.  My best guess at this 
> point is to increase the capacitance of the CP adjustment by using a larger 
> screw or screw head of nut or penny soldered to the nut, etc. and see if 
> you can get at least a small "W".  I'll look at your site and I'll write 
> you back if I have any more ideas.

Well, I finally did and very close plots to the G6LVB analyzer screen capture
from the URL above. 
> Fred
> >From: Joe Ennis <joe@whiskey7.net>
> >To: "Frederick M. Spinner" <fspinner@hotmail.com>
> >Subject: Re: [amsat-bb] W0FMS mods
> >Date: Wed, 20 Aug 2003 20:27:54 -0600
> >
> >Hello Fred,
> >
> >Have read your mails w/ great interest and have built the W0LMD L/S patch
> >incorporating your suggested frequency adjust for the S-band patch. I do
> >not have a power source for S-band and did the best I could by building
> >a range using a K5GNA S-band signal source and a rigid mount for the test
> >patch where I can rotate it through exactly 90deg.  I achieved what I think
> >may be good results with the signal strength received by the patch being
> >identical at both angles of rotation. The screw positions came out very
> >close to what you and W0LMD report.
> >
> >My problem is w/ the 1268 patch. I do have a sig source and reflected pwr
> >meter and set about to get the "W" w/ no joy. I've returned to the (your)
> >writings on this and have determined that the drop SWR from say 1.2 at
> >resonance to 1.05, at +/-5% freq offset. Can't do that at 1268. +/-2.5%
> >is about all one cat swing.
> >
> >I'm wondering if your experience says that this is enough swing to see
> >the effect. I'm not having much luck seeing a peak in SWR at resonance
> >with dropoff on either side.
> >
> >I started off w/ the 1/4-20 screw settings the Robert suggested in his
> >construction article but, those settings were way off! I've been able
> >to achieve resonance with the patch with a 1:1 SWR but, small changes
> >in tuning and phase adjustment seem to just shift resonance and do not
> >put a hump in the middle.
> >
> >The patch is accuratly built. You can check my technique and the finished
> >product on my website (url in sig below). Do you have any encouraging
> >tips for this old has been?
> >
> >tia es 73,
> >


thanks again Fred,

Joe Ennis - w7net
Whitefish, Montana

Sent via amsat-bb@amsat.org. Opinions expressed are those of the author.
Not an AMSAT member? Join now to support the amateur satellite program!
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org