Web lists-archives.com

Re: Help, windows dummy has bought one. I need it for a job or 100




On Wednesday 08 May 2019 04:55:47 am john doe wrote:

> On 5/8/2019 10:24 AM, Gene Heskett wrote:
> > On Wednesday 08 May 2019 03:49:34 am Joe wrote:
> >> On Tue, 7 May 2019 18:47:50 -0400
> >>
> >> Gene Heskett <gheskett@xxxxxxxxxxx> wrote:
> >>> Greetings all;
> >>>
> >>> First it doesn't have a clue what to do with a wired network.
> >>> It sure wants to hook up to all the neighborhoods wifi, all of
> >>> which are secured.
> >>> Second, its like stretch seems locked to ipv6 but its ipv4 for at
> >>> least a hundred miles in any direction from  my 10-20 in North
> >>> Central WV.
> >>>
> >>> Third, I can't find a place to enter a netmask route or gateway,
> >>> its been sleeping with dhcp for way too long.
> >>>
> >>> I finally find what sort of looks like the old xp network
> >>> configurator but it error beeps at me to the entry of any address
> >>> on my local net that isn't already taken.
> >>>
> >>> So how do I convince this brand new unibody HP to use a static
> >>> wired network setup?
> >>>
> >>> In the FWIW category, it takes winders 10 about 10x longer to boot
> >>> than any of my linux machines. Makes me wonder if they should have
> >>> named it window-0.1 because it is boringly slow.
> >>
> >> Shouldn't. I have a W10 netbook, though I'm not familiar with it,
> >> it had Debian installed within a week. Boot (from definitely off)
> >> is less than thirty seconds. Booting should not be held up by
> >> network issues.
> >>
> >> Open up the properties of the Ethernet adaptor, select TCP/IPv4,
> >> Properties, then untick the automatic options. You should be able
> >> to enter values in the address, mask and gateway boxes, and specify
> >> DNS servers below. It shouldn't need a reboot.
> >
> > There is no place in that sequence to select TCP/IPv4 on this
> > machine. If ipv6 dhcp fails, you are apparently screwed. And they
> > call this an OS? Not where (and when) I went to school.
>
> Why not trying command line (netsh)?
>
I haven't found a command line, yet. But today is another day. Let me get 
one eye open simultaneously first, cold, half a pot of yesterdays coffee 
hasn't kicked in yet.  Usually takes 2 to 3 cups to get me started 
right.

I'm also well pleased that I'm not being chastised for winders here, 
thank you very much for that. I bought this machine to serve as the web 
browser/administrator for a stemlab "redpitaya" because its one of the 
only bits of test gear I could find that can measure the impedance and 
reactance of a radio stations antenna tower and draw whats called a 
smith chart showing how that antenna is tuned.  They all have a matching 
network at the base of the tower, and as other gear gets hung on the 
tower, it goes out of tune. The make has both windows code and linux 
code but the linux code doesn't work, something wrong in the comm 
protocol.  Its taken me 3 weeks to convince the maker that his web page 
serving up this stuff is broken because it feeds a linux user following 
those directions, a windows .exe, which of course does run on linux.

Backing up one directory I found the linux code kit, buckets smaller than 
the windows .exe, an all written in python3. I've apt-get installed all 
the linux stuff, python3 wrapped around matplotlib. It opens its own 
window when run, and draws the basic chart on-screen, but its missing 
the right hand column of buttons and isn't talking to the redpitaya. Its 
actually an stm32 running ubuntu in a cig pack case, and the rf bridge 
in a similar case. But the web server seems disconnected from the app 
even though the ap is running on another workspace of the linux box 
running ff and accessing it.

Take a look at:

<https://redpitaya.readthedocs.io/en/latest/appsFeatures/apps-featured/vna/appVNA.html>

scrolling all the way to the bottom, you'll see a chart of a moderately 
well tuned antenna. Still isn't 100% centered on frequency, but a 
transmitter feeding it wouldn't be getting more than 1 or 2% of power 
reflected back to it.  The tower I'll tune first is reflecting a good 
20% back at the transmitter, making it run hot. The transmitter maker 
has sold the owner a handfull of big toroid cores to install on the 
feedline where it leaves the transmitter, but the ferrites have 
a "curie" temperature that can be below the boiling point of water in 
some formulations, that causes a loss of magnetic properties and makes 
the toroid run hot, hot enough he's shattered 2 of them like hitting 
glass with a hammer. At about $110 a core so busted.

All that will go away if the antenna is "properly tuned", and this is the 
gear that can tell me which way to move coil clips and what have you to 
do this to achieve a chart, hopefully even better than the one at the 
bottom of the above page.  That center dot should be closer to the 
horizontal centerline than it is. Ideally the center dots should be even 
closer to the 50 ohm mark at the center.

This of course changes with the weather, phase of the moon and which side 
of your mouth you've got a chaw of Kentucky Twist parked in. :)

All this is a dying art because the engineers that did do this tuning 
have largely died of old age.  And with the General Radio gear of the 
day, would have to spend hours in the middle of the night logging data, 
and hours the next day drawing a chart from that data, before they could 
make an adjustment, repeat each night and day till it was close enough 
for the girls he went with, something this device can do in ten or less 
minutes per adjustment/measure cycle.

My biggest problem will be in getting from the sma connector marked DUT 
for Device Under Test, on the bridge to the point on the matching 
network where the feed line from the transmitter is bolted into it. 
Those connections will need to be as short is possible.

Thats what this is all about. A ticket to participate in the show will 
cost maybe $1100.  Once.


> --
> John Doe


Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>