V7 Chapter 54
Its a blog - Read from the bottom up
306 - Reconfiguration Blues 2014-07-09
It all seemed so simple. Tear it all down, change it, put it back together. What could be simpler? Well this:
- Archive hard drive dead - Rebuilding from a back up cost all the SETI data collected between 2014-06-17 to now. Its up and running again.
- Port Forwarding - Seems wrong to me but this is what works:
- InMotion overload - My hosting service, InMotion, cannot seem to handle more than one or two pixels posting every second or so - it goes into overload. What I have to do is buffer the pixel software to send say 10 events at a time or send all of them once an hour or something like that.
- Multiple GPS Antenna - I still cannot seem to get the system running unless each pixel has its own GPS hockey puck hanging out the window. The repeating antenna does not seem to do the job. Why. I still have no good answer for this
- Clock - I have screwed up my Trimble Lady Heather clock somehow. I must get this running. Late word - I have sorted out the clock (simple cold solder joint) and Lady Heather is back to snarling and whipping the local oscillator back into shape. Its now phase locked as God intended.
305 - Station Reconfiguration 2014-07-01
Lately I have felt that the station is being dragged down by the way it is mechanically laid out. Computers here and there, a cable cabinet that is *way* out of control, data links running everywhere. It needs to be reorganized. This happens once every twenty years or so...
So I pulled all the electronics down, moved all the desks and work benches out and am putting down a new floor. After that all the stuff goes back in, but this time with some forethought.
I will post on Facebook and Twitter when operation restarts.
304 - Pixel Coincidence Detector 2014-06-24
I am interested in detecting when two (or more) pixels detect a gamma ray event. So I am working on the software to display when it happens. It will show up on this page of the web site.
I have come to realize that the hosting provider I use (InMotion) can be overwhelmed by just two pixels posting every second or so. They start to yell at me and start making suggestions that I need to find a hosting company more 'in line' with my needs.
So what I have to do is combine the posting of each pixel to a batch of events rather than one at a time. I think this might help.
303 - Back To Basics 2014-06-23
Sometimes I simply flounder. My eyes roll to the top side of my head and I lay at the bottom of the ocean and get nothing done for days on end. Other times I founder when I can't get my feet to work right. Either condition is serious.
When it happens I start a ToDo list to clear my head:
- Get my GPS disciplined oscillator running right using Lady Heather my station disciplinarian. Lady Heather
- Figure out how the damn GPS antenna farm I have created can be simplified
- Sort out the serial port interface that causes problems with my workbench RF mouse and keyboard
- Lock the 1.4 GHz LO with the 10 MHz output from the disciplined oscillator
10 MHz GPS Disciplined Oscillator
thats a start.
Jumping Mouse - It appears that when the Trimbel GPS receiver is trying to access the GPS satellites it 'triggers' the serial port it is connected to. This causes all the other serial devices to 'jump'. One of those devices is an RF dongal that runs my keyboard and mouse for the workbench computer CoCo.
Its running fine now that the receiver is connected to an outside antenna all to itself.
I managed to get the GPS locking system running. Not really sure how but I think its the GPS antenna itself that is the problem. This box needs a 5 Volt active antenna and the rest of the stuff in the shack needs 3 V antennas.
302 - Errors in remote MySQL server 2014-06-13
My hosting provider (InMotion out of L.A.) moved my web site to a different server. This morning I got a lot of errors like this:
Warning: mysql_close() expects parameter 1 to be resource, null given in /home/setine5/public_html/operation/hits_skymap/hits_skymap.php on line 43
I remembered that I had to change the .htaccess file to contain the following:
#Alternate default index pages
ErrorDocument 400 /400.php
ErrorDocument 403 /403.php
ErrorDocument 404 /404.php
ErrorDocument 500 /500.php
I also remembered that I had a file on the remote called: remote.htaccess and that all I had to do was to rename it and joy would prevail.
There is a file on the remote named Xphp.ini. That needs to be renamed php.ini. This clears the warnings from the remote MySQL server.
The file must remain named Xphp.ini on the localhost because otherwise the local MySQL server will not work.
Next time this is a reminder...
302 - Thinking out loud 2014-06-09 to present
I have been thinking about the ERGO pixels some more. This is what I know (or what I think I know).
- Cosmic-rays strike the Earths atmosphere at about 10 km high. About 90% of these are protons
- These strikes produce air showers of daughter particles
- The shows strike the Earth over several square km
- The ERGO pixels I have running generate a huge number of 'events' but most of these are noise created by the detector itself
I think what I need is to distribute the two pixels I have over a short distance (one I'll keep and the second about
It seems as though what is needed is a coincident detector...
301 - Spectrum Analyzer Repair 2014-06-05 to 2014-06-12
While working on getting two pixels running at the same time I realized that I had to have a better GPS antenna system. As it is each GPS receiver, and I have many, requires its own outside antenna I have a forest of GPS antennas hanging and dangling. I decided to use just one good quality antenna and a distribution system for the rest.
That lead me to try to use my HP Spectrum Analyzer. I quickly found that it had crapped out. This picture shows the problem:
H.P. 180 Oscilloscope Main Frame
The image only fills about a 1/4 of the display. I opened it up and thought I'd take a chance and simply change the Horizontal Output amplifiers. That was a pain but I finally finished. Trouble is that the problem remains.
The Horizontal Amplifier
Output Stage Close up
This picture shows the NTE 128 and NTE 129 output amplifier and how they are thermally connected by a heat sink.
So now that the shotgun approach has failed I have to apply some actual troubleshooting.
When a CRT display 'blooms' it usually because the High Voltage power supply is not putting out enough voltage. Low HV allows the focusing anodes to pull the beam to much as it paints the phosphor on the screen and causes it to widen. Conversely a display that is not wide enough and at the same time not tall enough is usually caused by a HV power supply that is putting out to much juice.
I broke out my old Heath Kit High Voltage probe and will attempt to make some measurements.
OK - That's not the problem either. The HV is 3120 Volts just as it should be. At least I got to use the probe for the second time in the last 40 years...
I give up - After spending an inordinate amount of time on this problem I decided it was best to simply go to eBay and order a working display. That's what I did. Should be here early next week.
Update: The replacement display section came in today (2014-06-12). It was *very* rough, dirty, cobwebby etc. A bit of cleaning and looks good and works fine.
300 - Display of Scans 2014-05-25 to present
I have been developing some tools to help with the search. If you look at Operations/Search Planning you can see what is available. This helps a lot but what is needed now is a chart that will show how the scans are progressing through the 100 MHz of the Waterhole that SETI Net is scanning. Something like this:
The trouble is that where this chart has 12 Y axis entries in the series I would have 370 (one for each target). But its a good starting point.
299 - Moving the raw SETI Data 2014-05-18 to 2014-05-20
My Hosting company decided that I was filling up there server to fast with JPGs and WAVs from the search so they moved me to a different server. I am sure this one is located in an outhouse in the back of InMotion and is reserved for miscreants like me that think just because they were sold a hosting service that has 'unlimited storage' that I could actually store 'unlimited' stuff on it - I was wrong of course. This Windows 3.1 server (with a 386 and 256 Meg memory I'm sure) was way to slow and kept crapping out so I had to change something.
I decided to move the 25 GBytes or so of JPGs and WAVs to my own server. Then when a user of the web site drills down to the raw data and clicks on a file it will be delivered directly from my own server. All the other normal web site functions still come from InMotion.