[PSDR] Spooky October HamWAN Update

[Bart Kus]

Hello,

A lot has happened recently, and I've been lagging on communication, so 
here's an oversized update.  If you would like to take some of the tasks 
below off my hands, let me know!  The to-do list is kind of overwhelming 
right now.

*Administrative*

HamWAN is now officially a non-profit corporation in the state of 
Washington.  We've also got a business license and an IRS EIN for tax 
purposes.  Because of the nice legal statuses, we've now got a bank 
account, a PayPal account and a credit card!  Just like a real boy!

This has all allowed HamWAN to launch the DONATION PROGRAM!  Right now, 
yes right now, you can hit up HamWAN.org and make donations! You can 
also become a long-term supporter and donate on a recurring monthly or 
yearly basis.  I don't want to drown the rest of this status update 
email with donation details and why they're important and what donating 
gets you, so please read the details at the Service Levels 
<https://www.hamwan.org/t/tiki-index.php?page=Service+Levels> page.

The next steps on the legal front will be making a 501(c)(3) filing.  
The process has begun and the board should be approving it in early 
November.  Processing time with the IRS will be a few months.

Speaking of the board, it is composed of:

Position
	Name
	Callsign
Bart Kus
	Chairman
	AE7SJ
Curt Black
	Director
	WR5J
Benjamin Krueger
	Director
	N2IHK


I would like the board to grow, so if you or anyone you know is serious 
about contributing, please get in touch.


*Technical*

Here at HamWAN Labs (aka: my home) engineers (aka: me) were successful 
in setting up an MPLS network with 4 MikroTik routers:

[admin at D] > /tool traceroute 192.168.44.10
  # ADDRESS                                 RT1   RT2   RT3 STATUS
  1 10.0.2.1                                10ms  8ms   8ms <MPLS:L=29,E=0>
  2 10.0.3.1                                13ms  7ms   10ms 
<MPLS:L=23,E=0>
  3 10.0.1.1                                10ms  9ms   8ms <MPLS:L=28,E=0>
  4 192.168.42.1                            9ms   8ms 9ms
  5 192.168.44.10                           11ms  10ms 9ms

MPLS-TE is next up on the network test agenda.  IPv6 was tested 
successfully.  Maximum over-the-air data throughput was tested as 123Mbit/s.

On the RF side of things, maximum throughput was measured on two RF 
links simultaneously, within close proximity to each other, and 
configured to use adjacent spectrum.  The results showed interference 
and slow-downs.  So now it is important to measure the inter-antenna 
coupling and relate it to some minimum guard band size.  In this 
harshest of conditions (no inter-antenna shielding) a guard band of 
20MHz was sufficient to eliminate all interference and slow-downs.  
Although the tests were only being done @ 10mW not the full 1.3W, so as 
to not saturate the receivers.

Speaking of antennas, HamWAN Labs has a couple candidate models in stock 
and is about to perform analysis!  Doing work at 6GHz is no easy feat, 
but for about $700 it looks like we're all geared up to pull it off.  
The test setup is something like this 
<https://www.hamwan.org/t/tiki-index.php?page=Antenna+Testing&structure=HamWAN>.  
But due to budget constraints only one power meter is available 
(although it's dual-channel!), and the coupler is unidirectional, not 
bidirectional, so it'll have to be switched around to perform the two 
different measurements.  A little sprinkling of GPIB control software 
and these instruments should take a few hours to perform a complete 
sweep of each antenna, measuring:

1) Incident power
2) Reflected power
3) Received power
4) Frequency

This will characterize the SWR of each antenna and their claims of 16dBi 
maximum gain.

The next antenna experiment will be to measure radiation patterns, as 
the manufacturer specs are not trustworthy.  HamWAN Labs has bought a 
rotor for this purpose and it needs to be rigged with a custom 
controller circuit to step it accurately.  It also needs to be subject 
to computer control.

We'll need to figure out a way to control elevation as well as azimuth 
to produce 3D models.  Given the short transmission ranges involved, a 
typical Az-El rotor setup will introduce excessive sideways motion of 
the antenna under test and will skew the results.  I'm open to 
suggestions.  The TX and RX antennas are spread about 20ft apart.

Finally, and this is the stretch goal, it would be good to measure 
radiation patterns as they vary with frequency.

Once the antennas are fully characterized, inter-antenna coupling can be 
modelled and interference can be minimized.  The coupling should also be 
directly measurable by setting up the antennas as though they were part 
of a distribution node (3-sector cell site), injecting a known signal, 
and measuring the RX.  This type of measurement will just verify the 
antenna models are correct, and that the measurement methods are sound.

I'm also trying to calibrate the Boonton 4200 power meter and the power 
probes.  To this end, I'll be trying to construct a copy of a Boonton 
2500 DC calibrator unit.  The ones on eBay are prohibitively expensive 
for what amounts to a 10VDC supply in a box with a bunch of resistors.  
I don't have a good AC calibration source (50MHz from +20dBm to -20dbm 
for one sensor, and down to -60dBm for another), so I'll be faking it 
with signal generators and an oscilloscope I have on-hand.  If you 
happen to have an accurate and calibrated signal source @ 50MHz, do let 
me know!


*Sites*

On the site front, Cougar is still in the works and taking longer than 
expected.  Please do talk to people you know and help secure sites for 
the project.  There's a map of desired sites here 
<https://www.hamwan.org/t/tiki-index.php?page=Puget+Sound+Data+Ring&structure=HamWAN>, 
towards the bottom of the page.  Gold Mtn, Blyn Mtn, Capitol Peak, 
Hatfield's Daughter, Sugarloaf Mtn, Graham Hill, Cougar Mtn, Tiger Mtns 
and that one Unknown one is apparently Cultas Mtn...these all need to be 
landed.  Paine Field and Mt.Constitution are secured. Lyman Hill is 
securable with money.


*Speaking of Money
*

Just to file for 501(c)(3) charity status is a $400 application fee, so 
please, people, donate!  And spread the word to other hammy-compu-nerds!

Thanks,

--Bart

PS: Real-time project discussion / coordination is on irc.freenode.org 
in #HamWAN.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.hamwan.net/pipermail/psdr/attachments/20121026/9ed13e28/attachment.html>