WXforum.net
Weather Station Hardware => Blitzortung => Topic started by: saratogaWX on September 23, 2013, 11:31:18 AM
-
Finally had some time to begin the build process. My wife is away at our daughters helping care for the new twins, and I am on duty for my wife's mother who lives nearby in assisted living. With nobody else to bother with the electronic mess, it's a good time to begin construction :)
Yesterday, I finished the inventory of parts and divided them into Amp and Controller groups.
I set up the assembly area on the kitchen table with anti-static mat, Panavice, bright light, Hakkon soldering station, and trusty magnifier headband.
Wow, the numbers on some of the parts are really small for old eyes, but at 10x, become clear enough to read.
I'm following along with Don's excellent assembly advice/checklist (http://www.wxforum.net/index.php?topic=20037.0) and starting with the Amplifier assembly.
First the SMD op-amps. I found that a self-gripping tweezer with angled head worked best for me to hold the little buggers. What worked best for me was to:
Flux pen the mounting pads on the board
apply a bit of solder to the top-right pad
use the tweezer to position the SMD over the pads
heat the top-right one to melt the solder and hold the part in alignment
Solder the other pins
Resolder the original pin
inspect using the 10x lens to make sure all the joints were 'hot' and no bridging happened.
Took about 2 hrs to get the hang of that.. thank goodness that only 4 SMD op-amp packages were needed.
Then on to the diodes (done) and about half the resistors before bedtime. More work on it today.
-
Glad to hear you are well on your way Ken!
Magnification will become, if it has not already, your best friend during this project.
Obviously, you have patience and appreciate the tips others have provided.
Joe
-
Wow, the numbers on some of the parts are really small for old eyes, but at 10x, become clear enough to read.
Egon did not send Eyeballs MK II with your kit either??? He left them out of my order, also. That's at least twice.
Mike
-
Congratulations Ken!!! I have been too busy to do any further on my build the past few days but hope to get back to it tomorrow or Wednesday. Once I get back in the swing of building again, I'll finish up the step-be-step manual.
Good luck with your build!
-
Alright! Even after 40 years, I can still solder :) :shock: \:D/
I completed the RED Amplifier board and fired it up with a 5V for the hardware test and it worked per Egon's First Hardware Check. \:D/ \:D/ \:D/ 8-)
(http://saratoga-weather.org/images/Blitzortung-RED-Amplifier-sm.jpg)
Thanks Don for making the resistor and ceramic capacitor checklist -- it really simplified the assembly! I used Egon's RED build doc for the rest of the parts. The 'hard' bits I'd encountered:
1) remembering that the longer leads are + for the LEDs and electrolytic caps .. don't cut the leads early :)
2) soldering the tiny lands for the mini-USB power connector (that was a 10x magnify job)
3) straightning the leads on the 5 ICs and getting them carefully in the sockets
Now on to the RED Controller assembly (likely starting Wednesday .. have to clean up the place for dinner Tuesday with son and family here).
-
Looks great. Congrats on getting the amp board done. The trickiest thing on the controller board is the GPS module, once you are past that it should be clear sailing...
Just one item on the capacitors.
remembering that the longer leads are + for the LEDs and electrolytic caps .. don't cut the leads early
There is a negative band or stripe in the shrink wrap of the electrolytic caps. This is contrary to the old positive markings in the past, but is the standard. In addition to the longer positive lead, the wide stripe is the negative. You can see "-" markings in the band.
Greg H.
-
Right you were, Greg. The GPS was a tricky beast to get installed. I did the apply solder to one land (after fluxing), the using the one to hold the module in place while I soldered one side, then the other. Went over the result with the 10x magnifying lens, applied a bit of solder wick to clean up one join, and then was done :)
Thanks for the info on the electrolytic caps .. I'd noticed the band with "-" too. Back in the day, I think the cans were marked with a big "+" for positive, but my mind may misremember that. The nice picture of the assembled Controller in Egon's doc was very helpful as it showed the orientation of the caps also .. I have a big color print of both Amp and Controller boards to remind me how it should look as I assemble the boards.
The controller board now has GPS module and all diodes installed .. then I had to break down the assembly area and clean the house for this afternoon's family dinner/visit. I'll resume construction on Wednesday ..
-
The amp board looks great Ken. Well done! =D>
Don't forget to lay down the two transistors (T40 & T41) and the four 100uF electrolytic capacitors (C44, C48, C55 & C56) so they will be clear of the Discovery board when it is installed. The picture in Egon's document shows them pretty clearly for reference.
-
Thanks Don :)
Today (Wednesday) was to have been a quiet build-it day, but it was interrupted by a broken molar and a trip to the oral surgeon for an extraction and implant .. now being on pain meds, the concentration is somewhat dimmed, so the assembly is proceeding at a slow pace with triple checking before soldering. About half the resistors are now installed .. hope to finish up the controller board tomorrow.
-
Soldering while on Vicodin is definitely contraindicated! :grin:
I went through exactly the same thing today! 4 hours in the dentists's chair...... :shock:
Hope you're feeling better than I am!
Best,
Don
Thanks Don :)
Today (Wednesday) was to have been a quiet build-it day, but it was interrupted by a broken molar and a trip to the oral surgeon for an extraction and implant .. now being on pain meds, the concentration is somewhat dimmed, so the assembly is proceeding at a slow pace with triple checking before soldering. About half the resistors are now installed .. hope to finish up the controller board tomorrow.
-
... a broken molar and a trip to the oral surgeon for an extraction and implant .. now being on pain meds, the concentration is somewhat dimmed, ...
Soldering while on Vicodin is definitely contraindicated! :grin:
I went through exactly the same thing today! 4 hours in the dentists's chair...... :shock:
Gentlemen! Relax! Jaw clenching and teeth crunching are not in the build instructions! :roll:
Dikes (side cutters) are the preferred tool for clipping leads... :lol:
Mike
-
Sorry to hear of all the tooth problems. I agree, dikes are the preferred method of trimming the leads. :-P
Anyway, just wanted to let you know I just uploaded another version of my step-by-step instructions. I hope to have the rest of it done in the couple of days as I have to work tomorrow.
My best to both of you and hope you recover quickly.
-
The dentist did ask bout the 24AWG slot in my front teeth. So much easier than digging around the tool box for those wire strippers....auto-adjusting too! :lol:
D.
... a broken molar and a trip to the oral surgeon for an extraction and implant .. now being on pain meds, the concentration is somewhat dimmed, ...
Soldering while on Vicodin is definitely contraindicated! :grin:
I went through exactly the same thing today! 4 hours in the dentists's chair...... :shock:
Gentlemen! Relax! Jaw clenching and teeth crunching are not in the build instructions! :roll:
Dikes (side cutters) are the preferred tool for clipping leads... :lol:
Mike
-
The dentist did ask bout the 24AWG slot in my front teeth. So much easier than digging around the tool box for those wire strippers....auto-adjusting too! :lol:
D.
... a broken molar and a trip to the oral surgeon for an extraction and implant .. now being on pain meds, the concentration is somewhat dimmed, ...
Soldering while on Vicodin is definitely contraindicated! :grin:
I went through exactly the same thing today! 4 hours in the dentists's chair...... :shock:
Gentlemen! Relax! Jaw clenching and teeth crunching are not in the build instructions! :roll:
Dikes (side cutters) are the preferred tool for clipping leads... :lol:
Mike
I agree they do work as the best strippers.... dentist love you using them too! Guarantees you as a future client or repeat customer..... :lol:
-
I agree about the teeth being convenient wire strippers .. I have a small 24gauge notch in my front teeth which was caused by that too...
Don, my oral surgeon only took 1.5 hrs on the one tooth.. one root had to be removed a chunk at a time due to a bulge at the bottom. They had a 3D CAT scanner for the X-Ray before and after .. he'd used it to size/place the implant base with great precision. I'm still doing soft foods, but the healing is progressing -- likely the last day for the pain meds today.
Despite the slight pain-med fog, I did successfully complete the RED controller build last night, and the initial hardware test worked!
This morning, I flashed 4.3 firmware, and used the web interface to flash 4.4 firmware. I did a test setup with all components connected, and have the station connected as ID 802 to user 713 on Blitzortung.org, but it's not showing up yet in the lists (and I don't know why).
(http://saratoga-weather.org/images/Blitzortung-RED-test-setup-sm.jpg)
Two issues:
1) Station 802 isn't showing up in the Blitzortung list quite yet, nor on lightningmap.org -- is there anything else I need to do?
2) The Amber display is illuminated, but showing no data even after reset and pressing of the blue button on the STM board. Is there some test I could do to determine if the display is defective or I've miswired something?
Best regards,
Ken
-
Just did a debug log for the GPS and found this:
2013-09-26 18:52:21 6700 | GPS: ===========================================
2013-09-26 18:52:22 6701 | GPS: 1PPS deviation too high 5522. Reset counter!
2013-09-26 18:52:22 6701 | GPS: 75829582 | PGTOP,11,3
2013-09-26 18:52:22 6701 | GPS: 78255678 | GPGGA,185223.000,3716.4865,N,12201.3762,W,1,7,1.17,126.3,M,-25.9,M,,
2013-09-26 18:52:22 6701 | GPS: 79528286 | GPGSA,A,3,12,17,10,04,25,24,02,,,,,,1.44,1.17,0.84
2013-09-26 18:52:22 6701 | GPS: 81135524 | GPRMC,185223.000,A,3716.4865,N,12201.3762,W,0.15,7.47,260913,,,A
2013-09-26 18:52:22 6701 | GPS: 13-09-26 18:52:23 lat:37.274776 lon:-122.022934 alt:126.3 sat:7 inview:0 fix:3 integr:A pdop:1.4 status:A
2013-09-26 18:52:22 6701 | GPS: ===========================================
2013-09-26 18:52:23 6702 | GPS: 1PPS deviation too high 78826. Reset counter!2
2013-09-26 18:52:25 6704 | GPS: 75256772 | PGTOP,11,3
2013-09-26 18:52:25 6704 | GPS: 77938994 | GPGGA,185226.000,3716.4865,N,12201.3762,W,1,7,1.17,126.3,M,-25.9,M,,
2013-09-26 18:52:25 6704 | GPS: 79210644 | GPGSA,A,3,12,17,10,04,25,24,02,,,,,,1.44,1.17,0.84
2013-09-26 18:52:25 6704 | GPS: 80821028 | GPGSV,3,1,10,02,70,164,20,12,56,324,34,04,55,063,38,24,51,232,22
2013-09-26 18:52:25 6704 | GPS: 82371926 | GPGSV,3,2,10,10,19,113,24,17,18,063,28,25,17,309,24,29,06,258,
2013-09-26 18:52:26 6705 | GPS: 83278600 | GPGSV,3,3,10,05,05,160,,14,01,311,
2013-09-26 18:52:26 6705 | GPS: 1PPS deviation too high 120964. Reset counter!
2013-09-26 18:52:26 6705 | GPS: 704650 | GPRMC,185226.000,A,3716.4865,N,12201.3762,W,0.11,7.47,260913,,,A
2013-09-26 18:52:26 6705 | GPS: 13-09-26 18:52:26 lat:37.274776 lon:-122.022934 alt:126.3 sat:7 inview:10 fix:3 integr:A pdop:1.4 status:A
2013-09-26 18:52:26 6705 | GPS: ===========================================
2013-09-26 18:52:26 6705 | GPS: 74913272 | PGTOP,11,3
2013-09-26 18:52:26 6705 | GPS: 77350852 | GPGGA,185227.000,3716.4865,N,12201.3762,W,1,7,1.17,126.3,M,-25.9,M,,
2013-09-26 18:52:26 6705 | GPS: 78623456 | GPGSA,A,3,12,17,10,04,25,24,02,,,,,,1.44,1.17,0.84
2013-09-26 18:52:26 6705 | GPS: 80228232 | GPRMC,185227.000,A,3716.4865,N,12201.3762,W,0.15,7.47,260913,,,A
2013-09-26 18:52:26 6705 | GPS: 13-09-26 18:52:27 lat:37.274776 lon:-122.022934 alt:126.3 sat:7 inview:0 fix:3 integr:A pdop:1.4 status:A
2013-09-26 18:52:26 6705 | GPS: ===========================================
2013-09-26 18:52:27 6706 | GPS: 1PPS deviation too high 134242. Reset counter!
2013-09-26 18:52:27 6706 | GPS: 76325402 | PGTOP,11,3
2013-09-26 18:52:27 6706 | GPS: 78713484 | GPGGA,185228.000,3716.4865,N,12201.3762,W,1,7,1.17,126.3,M,-25.9,M,,
2013-09-26 18:52:27 6706 | GPS: 79986088 | GPGSA,A,3,12,17,10,04,25,24,02,,,,,,1.44,1.17,0.84
2013-09-26 18:52:27 6706 | GPS: 81590864 | GPRMC,185228.000,A,3716.4865,N,12201.3762,W,0.25,7.47,260913,,,A
2013-09-26 18:52:27 6706 | GPS: 13-09-26 18:52:28 lat:37.274776 lon:-122.022934 alt:126.3 sat:7 inview:0 fix:3 integr:A pdop:1.4 status:A
2013-09-26 18:52:27 6706 | GPS: ===========================================
and the GPS info on the status page shows
(http://saratoga-weather.org/2013-09-26_GPSissue.png)
I've tried resetting/cold start the GPS, and moving the antenna to a different location, but no luck.
-
Ken,
As a guess , it may be worth checking the 1pps output pin from the GPS module for dry joint/bridging. It is 3rd pin up from the bottom on the external gps socket side of the module.
Chris
-
From your photograph, it appears the amplifier is not getting control signals from the controller. None of the yellow LEDs should be illuminated while under the controller boards control. The green LED on the amp should be illuminated if the amp is communicating to the controller.
Don
-
Re: Display- You're using Amber display, right? Make sure jumper 1 is open.
Ok.. just thoughts:
Set for manual control under settings, main?
(Settings, all) 'enable potentiometer' is NOT checked?
(settings, all) display 'Contrast' is not turned off?'
You've just done the firmware, also? Unplug your power supply, count 10 and restart then reconnect.
GPS baud rate for highest?
There's no way the Ethernet and amplifier CAT5s are reversed, I suppose...
-
Ken,
As a guess , it may be worth checking the 1pps output pin from the GPS module for dry joint/bridging. It is 3rd pin up from the bottom on the external gps socket side of the module.
Chris
Thanks Chris,
It was very hard to see, but it was a open solder connection on that pin.. go figure. Took a bit of work with a 10x magnifier to see and fix it. Now the GPS log is reporting 2013-09-26 21:51:16 599 | GPS: ===========================================
2013-09-26 21:51:17 600 | GPS: 22224902 | PGTOP,11,3
2013-09-26 21:51:17 600 | GPS: 23768218 | GPGGA,215117.000,3716.4799,N,12201.3775,W,1,7,1.07,114.9,M,-25.9,M,,
2013-09-26 21:51:17 600 | GPS: 24230220 | GPGSA,A,3,29,12,02,10,05,31,25,,,,,,1.35,1.07,0.83
2013-09-26 21:51:17 600 | GPS: 24830098 | GPRMC,215117.000,A,3716.4799,N,12201.3775,W,0.11,22.26,260913,,,A
2013-09-26 21:51:17 600 | GPS: 13-09-26 21:51:17 lat:37.274666 lon:-122.022957 alt:114.9 sat:7 inview:0 fix:3 integr:A pdop:1.4 status:A
2013-09-26 21:51:17 600 | GPS: ===========================================
2013-09-26 21:51:18 601 | GPS: 21547526 | PGTOP,11,3
2013-09-26 21:51:18 601 | GPS: 23105174 | GPGGA,215118.000,3716.4798,N,12201.3774,W,1,7,1.06,114.9,M,-25.9,M,,
2013-09-26 21:51:18 601 | GPS: 23567220 | GPGSA,A,3,29,12,02,10,05,31,25,,,,,,1.35,1.06,0.83
2013-09-26 21:51:18 601 | GPS: 24167050 | GPRMC,215118.000,A,3716.4798,N,12201.3774,W,0.12,22.26,260913,,,A
2013-09-26 21:51:18 601 | GPS: 13-09-26 21:51:18 lat:37.274662 lon:-122.022957 alt:114.9 sat:7 inview:0 fix:3 integr:A pdop:1.4 status:A
2013-09-26 21:51:18 601 | GPS: ===========================================
2013-09-26 21:51:19 602 | GPS: 21881200 | PGTOP,11,3
2013-09-26 21:51:19 602 | GPS: 23422104 | GPGGA,215119.000,3716.4797,N,12201.3774,W,1,7,1.07,114.9,M,-25.9,M,,
2013-09-26 21:51:19 602 | GPS: 23860360 | GPGSA,A,3,29,12,02,10,05,31,25,,,,,,1.35,1.07,0.83
2013-09-26 21:51:19 602 | GPS: 24460188 | GPRMC,215119.000,A,3716.4797,N,12201.3774,W,0.11,22.26,260913,,,A
2013-09-26 21:51:19 602 | GPS: 13-09-26 21:51:19 lat:37.274662 lon:-122.022957 alt:114.9 sat:7 inview:0 fix:3 integr:A pdop:1.4 status:A
2013-09-26 21:51:19 602 | GPS: ===========================================
and the status display shows
(http://saratoga-weather.org/2013-09-26_GPSfixed.png)
so all seems well with the GPS now :D Saratoga (802) station is online with Blitzortung.org station list. \:D/ \:D/ \:D/ 8-)
(http://saratoga-weather.org/2013-09-26_online.png)
The amber display is now working after I found one unsoldered pin on the connector strip #-o .. to make sure, I resoldered them all for the display. :oops:
From your photograph, it appears the amplifier is not getting control signals from the controller. None of the yellow LEDs should be illuminated while under the controller boards control. The green LED on the amp should be illuminated if the amp is communicating to the controller.
Don
Don, I checked the solder joins for the Amplifier and Controller cable connectors and swapped out the cable.. it's still in 'manual' setting mode (using the pot) and the green light is not illuminated (yellow lights are) on the Amplifier board. I don't know where else to look...
THANKS for your help guys.. I guess Cutty Sark Sailor was right about disrecommending assembly while using pain-meds
Best regards,
Ken
-
2) The Amber display is illuminated, but showing no data even after reset and pressing of the blue button on the STM board. Is there some test I could do to determine if the display is defective or I've miswired something?
Ken,
What does the GUI report for the 5V supply voltage? There is an issue with higher voltages and the display. I believe that it has something to do with the ST board running at 3.3V and the display running at 5V. The datasheet is ambiguous about whether the thresholds are 2V or ratiometric off of 5V. I think that some lines might be ratiometic and are not reaching the correct thresholds. Tobi has suggested lower voltages.
I bought a USB supply that runs a little lower, around 4.7V and I have had no display problems since I started using it. (I had some scrambled display problems before)
If you could try a mini-USB supply that runs a little lower it might be worth a try, especially if it is now >5V.
Greg H.
-
Congratulations Ken!
I had a few joints that the old eyes thought they had right. Second time they did.
Enjoy!
Joe
-
Congrats Ken!!! Good to see you on the station list. On the power supply, I am using an Apple iPhone charger that is plugged into my APC UPS to help level out the input voltage. The charger normally puts out between 4.92-4.99 Volts on a consistent basis. I have not had any display issues at all using this setup.
About the only issue I've had is the DHCP issue where in the webpage won't load after a certain time frame. The only way I have been able to fix that is to unplug the controller for 10 seconds and plug it back in. I know Egon and Tobi are aware of this, and hopefully will rectify this with a firmware update.
John
-
Congrats Ken!!! Good to see you on the station list. On the power supply, I am using an Apple iPhone charger that is plugged into my APC UPS to help level out the input voltage. The charger normally puts out between 4.92-4.99 Volts on a consistent basis. I have not had any display issues at all using this setup.
About the only issue I've had is the DHCP issue where in the webpage won't load after a certain time frame. The only way I have been able to fix that is to unplug the controller for 10 seconds and plug it back in. I know Egon and Tobi are aware of this, and hopefully will rectify this with a firmware update.
John
John, I am experiencing the issue. I thought the server was experiencing unusually heavy traffic. Man, that sounds the announcement on virtually every company you call. :lol:
-
My USB supply is the one from Adafruit for the Arduino (same as W3DRM/Don purchased) and puts out 5.2 volts. The system seems to be happy with 3.0V to the processor, and the display is working just fine now that I'd resoldered a bit (and found the one unsoldered lead) :oops:
I found the secret to enable remote control of the Amplifier gain in the German section of wetter-board.de for Blitzortung entitled "Anleitung System - RED : Manuelle Einstellung der Verstärkung (GAIN -Einstellung)"
Paraphrasing what was in there, the Settings panel Remote Control has to be changed from Automatic Mode to Manual Mode.
Then the Amplifier panel will show the gain parameters when the "Enable Potentiometer" check box is UN-ticked.
That turns on the Green LED on the Amplifier and turns out the Yellow LEDs.
Sometimes it helps to be able to suss-out German language instructions :)
Now if I could just understand what 'Idle' status means (or rather, why my test setup doesn't remain 'Running' in the station status on Blitzortung).
Best regards,
Ken
-
Ken,
Great that you found the bad solder joint and that took care of the display issue.
I looked at your signal and it seems that you are running big gain, but have a DC offset. There doesn't seem to be anything actually happening. For as much gain as you are running 32x16, you should see a lot of noise at least.
Can you check the signal chain with an o-scope? Something looks amiss...
UPDATE: Upon looking at the schematics, the amplifier board output is AC coupled. Unless the output 2.2uF was in backwards and leaking current, it is probably not originating from the amp board.
Greg H.
-
I took the Amp offline, checked all the components against the parts list, and looked over all the solder joints (touched up a few) and everything looks like the picture of the Amplifier board, except that instead of MCP6S91 Low Gain programable amplifiers, I have MCP6S21 PGAs .. I checked the data sheets for the two parts, and the pinouts are identical, and electricals are identical.. the MCP6S21 has -40 to 85C rating while the MCP6S91 has -40 to 125C rating, so I don't think the substitution is important.
I can touch the antenna A and B inputs with a small screwdriver and get some response on the Signal page, but even at high gain settings, nothing seems to appear in regular operation. Checked the ferrites connections .. they measure about 4 ohms so there is continuity in the connections. Also tried the Amp 2 connection -- same results.
I don't have an oscilloscope available, so I can't probe the innards to see where the signal may be lost, but it is odd that both channels in the Amp are behaving the same way.
Any more advice???
Puzzled,
Ken
-
Ken, there are pc-based o-scope programs that use the sound card for the inputs. I haven't looked into it myself but that might be something to try in lieu of having a real o'scope. Search for "soundcard oscope" and you'll get quite a few hits.
Here are a couple links you might want to look at:
- http://zelscope.com/ (http://zelscope.com/)
- http://www.zeitnitz.de/Christian/scope_en (http://www.zeitnitz.de/Christian/scope_en)
- http://www.ledametrix.com/oscope/ (http://www.ledametrix.com/oscope/)
- http://www.ladyada.net/library/equipt/diyaudioprobe.html (http://www.ladyada.net/library/equipt/diyaudioprobe.html)
Hope this helps... I hope to have my system completed early next week. Again, family and other things got in the way of getting it done today and this evening. :-(
-
When I first completed the amp board, I touched the input and looked at the output on the AC scale of my DMM. At big gain you should be able to see output. There is a band-pass filter, so the old "hum" test isn't as valid as it would be with a typical audio amplifier.
Is there an old Walkman or MP3 player that you could get a 10KHz tone recorded on? You could inject this at the input and trace it through the system with a DMM or better, one of the suggested PC scope programs.
Does the DC offset scale with gain? If you turn the gain down to minimum, does the DC offset go closer to zero? That would be a further indication of where the trouble is.
Check that the C15 and C30 electrolytic caps are the correct polarity on the amplifier board. They will leak DC current if they are reversed. Does DC appear at the OUT-A and OUT-B? Assuming you have a meter.
Greg H.
-
The only other possibility that I can think of is that some ADC settings or coefficients in the flash memory are corrupted. I doubt this because it is unlikely that the same issue would be on both channels. It might we worthwhile to try a "reset settings" to clear out the flash memory to defaults. It is a long-shot, but it is an idea to try.
Greg H.
-
Been offline chasing down ideas you've highlighted above .. thanks again for the continuing ideas.
I decided to acquire an oscilloscope and counter/frequency generator -- living close to several Fry's outlets can make you do crazy things like that. I found a Tektronix TDS1002B 60Mhz dual channel digital storage scope (discontinued) as one of two at Fry's in Palo Alto and a counter/function generator. I've now spent more on stuff around the Blitzortung kit than for the kit itself, but it does feel satisfying to have a Tek scope (much better than the monster I'd used back in the '70s).
So, did the DC voltage test with the DMM on the output leads.. Looks like I have about +4.1mV DC on the output, and it does not change with gain (tried 1:1 to 32:32 for gain settings). The C15 and C30 are in the correct polarity (negative stripe facing the test connector pins, which matches the photo of the board from the doc).
I hooked up the new scope and signal generator .. with 20mV P-P 10KHz input the output scales on amp-A and amp-B as you'd expect from the gain setting. At 32:32 the 20mV input signal is 2V P-P (1024 gain).
I'll post some scope pictures as soon as I figure out how to capture them :) Had to find a 256Mb flash drive, update the Tek firmware, and got the two captures (attached) . CH1 is the output pin, CH2 is the antenna pin. First image is Amp-A, second is Amp-B
Short story.. looks like the Amplifier(s) are behaving correctly, and amplified signal does pass through from A and B
So what to check out now?
Best regards,
Ken
-
Wow, that's a nice scope. I hope that you get good use out of it for many years. The closest Fry's to me is in Indy, but I get to go there every so often.
I think that this exonerates the amplifier completely. The generator and scope was the best way to prove it.
I would shift the focus to the controller board now. Using the schematic and ST Discovery board user manual at http://www.st.com/st-web-ui/static/active/en/resource/technical/document/user_manual/DM00039084.pdf?s_searchtype=keyword (http://www.st.com/st-web-ui/static/active/en/resource/technical/document/user_manual/DM00039084.pdf?s_searchtype=keyword) I would connect your generator to the amp as before and trace it to the Discovery board pins associated with the ADCs. If it is getting to the ADC pins on the top side of the Discovery board that will be another check.
Also, see the schematic fragment I have attached. Check that there is 1.5V at the center of the voltage divider from 3V to ground. This is the AC 'pseudo-ground' that the signals swing around. The signals on the ADC input pins should be biased to close to this point. The signals in a single-ended analog circuit swing from ground to the supply (more or less) and should be at the 1.5V bias point with no signal.
The 1.5V point is just DC, you won't see signal there. You should see the signals "riding" on top of the 1.5V bias voltage when you feed them in from the generator via the amplifier board.
Were you able to try a re-flash of the ST Micro? You could try that if the hardware checks out.
I hope that you didn't get a bad ST board, but it is possible that there is a problem with the ADC circuits on it.
Good luck and I hope that you can track it down OK.
Greg H.
-
Another question:
When you scale the gain up/down, does this change at all?
http://www.blitzortung.org/Scripts_php/php_signal_2.php?station=802®ion=3&mode=signal&unit_width=2&unit_height=12 (http://www.blitzortung.org/Scripts_php/php_signal_2.php?station=802®ion=3&mode=signal&unit_width=2&unit_height=12)
Greg H.
-
I've been fiddling a bit with the system.. currently have the gain cranked to 32:32 and have the odd disparity between Chan-A v.s Chan-B
(http://saratoga-weather.org/2013-09-30_gain-issues.png)
With the gain set the same, I'd really expect the output/noise level to be the same .. Should I look for some issue in the Chan-A side?
@Greg: yes, the php_signal_2.php display does change, but only if I use high gain settings 32:16 or 32:32. I have re-flashed (using the web interface) the 4.4 Firmware .. no change in behavior.
I'm starting to suspect that there is something amiss in the Amp-A channel...
Here's another question..
I currently have ' Ferrite rod, SKS120N, 120mm, W. Friese' set for the antenna type .. is that correct for the ferrite rods shipped with the default kit of parts?
Best regards,
Ken
-
Ken, what happens when you swap the ferrite inputs to the amplifier board? That should tell you if the problem is in antenna or the amps. If it's the antenna, then the relative waveforms should move to the other amp. If it is the amp then you will see little of no change to the waveforms.
-
Good idea, Don. Here's the captures before/after swapping the two ferrite antennas:
(http://saratoga-weather.org/2013-09-30_gain-issues.png)
(http://saratoga-weather.org/2013-09-30_gain-issues2.png)
It looks like A-channel has about 1/4th the gain at 32:32.
-
I'd suspect a problem with the Amp A circuitry. Going all the obvious things;
- Have you checked all of the solder connections for bad or missing solder joints?
- Any solder bridges?
- Since you have a scope do you see any input at the input pins?
- Then, looking at the output pins do you see any increase (amplification) in the signal output?
- Voltage comparisons between the two amps may also help find the problem.
Best of luck...
-
I've been meaning to mention this for a while. I've had trouble, even on my little VP2 interface board, with SMDs and solder joints. I finally started using a 10x loupe to check them and found some "dry" joints. Some would even work OK 'til the boards were coated, and then fail.
Inspect closely, and reflow if in doubt.
Loupes are available from Digi-Key and Mouser among others, and are cheap. I use the monocle style, but a hand held one like rock hounds and stamp collectors use would be OK.
- Have you checked all of the solder connections for bad or missing solder joints?
- Any solder bridges?
-
I thought that everything was hunky-dory with the signal generator and you were able to reach about 2V pk-pk on the amp outputs.
This would tend to lead to the conclusion that the amp channels are OK.
Are you putting the antenna wires directly into the screw terminals? You should solder the magnet wire to a larger stranded wire for the screw terminals.
Also, the insulator on the coil wires isn't the solder-strip type. It needs to be carefully scraped off to be soldered to the wires. Can you measure the continuity at the screw terminals and see the DC resistance of the antennas at the screws? Maybe it isn't making good contact.
Greg H.
-
I thought that everything was hunky-dory with the signal generator and you were able to reach about 2V pk-pk on the amp outputs.
This would tend to lead to the conclusion that the amp channels are OK.
Are you putting the antenna wires directly into the screw terminals? You should solder the magnet wire to a larger stranded wire for the screw terminals.
Also, the insulator on the coil wires isn't the solder-strip type. It needs to be carefully scraped off to be soldered to the wires. Can you measure the continuity at the screw terminals and see the DC resistance of the antennas at the screws? Maybe it isn't making good contact.
Greg H.
Putting my old Field Engineer hat on:
Ken has already switched the antenna wire inputs and the result shows a similar output. Since the problem did not move, the circuitry associated with Channel A (including the wire connections as suggested by Greg H.) has some kind of problem. Ken needs to follow the signal with his scope from the input, through the op-amp stages and all the way until he finds where the signal is getting attenuated or not passed correctly. It could also be on the Controller board.
IDEA: BEFORE tracing signals with the scope and, to further isolate the problem, do the following first:
- Move the shielded cable coming from the Amplifier board to the other RJ45 input on the Controller board. This will help isolate individual channels on the Controller board if the problem is with the controller board itself.
The other thing we haven't eliminated is the cable itself between the amp and controller board.
- Move the cable to the second controller input. If this doesn't doesn't change anything,
- use another cable to eliminate that.
If the problem still exists, you will just have to follow the signals from component to component until you find where it goes bad.
Just be very careful with your scope probe that you don't short-out any adjacent pins.
-
Having had the same problem as Ken.... I'd check the solder joints, very closely, on the channel that is in question and re-seat the IC's for that channel as well. Once I did both, my problem went away.
Joe
-
Ken,
Please oscilloscope probe the pins shown below at the ST-Discovery board with a generator test signal at the amplifier input. They feed the ADCs on the STM processor.
You can ignore channel 3 for now.
You should have about 2/3 (100/147) of the signal as at the amplifier outputs test header, except biased up to 1.5V.
If there is no signal at these Channel1 and Channel2 points, there is a problem on the controller board from the inputs to the Discovery board pins.
Greg H.
-
Having had the same problem as Ken.... I'd check the solder joints, very closely, on the channel that is in question and re-seat the IC's for that channel as well. Once I did both, my problem went away.
Joe
I decided to trace the signal through the paths on the amp(s) checking corresponding levels when the gain was set to 4:4 for both amps. I used a bit of clipped-off resistor wire in the probe so I could carefully touch pins on the ICs (including the SMDs) without shorting adjacent pins.
Chan-B wasn't working because Pin 3 on IC6a (the input to the first filter amp) wasn't connected (despite it looking connected)
(http://saratoga-weather.org/2013-10-05_no-signal.png)
A re-solder of that connection caused Chan-B to produce the same output as Chan-A for all the gain settings (with a 20mV 10KHz sine input). So, it looks like the amps are both working now.
With the 120mm ferrite antennas connected, I'm now seeing an odd difference between the two:
(http://saratoga-weather.org/2013-10-05_odd-noise.png)
One of the ferrites is picking up a repetitive 60Hz signal that is associated with that one ferrite, and the signal moves to the other amp if the connection is swapped between the antennas. I'm trying to figure out where the signal is coming from, but no luck so far. I did apply E shielding (foil tape) to the antennas (with a 1/8in gap along the axis), and tied the foils together to the antenna ground pins at the amplifier -- seems to have no effect on the periodic signal.
@Greg: I think the ADCs are working ok since the Chan-B Amplifier is now working.
@Don: thanks for the suggestions .. I have tried them all, and the 'trace the signal' was the resolving diagnostic test :)
Still slogging through the debug, and I'm building a ball-loop antenna to try out also.
Best regards,
Ken
-
That's great. Looks like you are making progress.
It seemed that when you tested it with the generator it was passing the signal, so I thought that the controller board/ADC was the remaining possibility.
Glad you found it. As you said, on to the antennas...
One of the ferrites is picking up a repetitive 60Hz signal that is associated with that one ferrite, and the signal moves to the other amp if the connection is swapped between the antennas. I'm trying to figure out where the signal is coming from, but no luck so far. I did apply E shielding (foil tape) to the antennas (with a 1/8in gap along the axis), and tied the foils together to the antenna ground pins at the amplifier -- seems to have no effect on the periodic signal.
Do the suspect ferrite's antenna wires have continuity with an ohm-meter? Maybe it is acting like a "wick" and the ground side is open? Is there a solder connection between the ferrite magnet wire and your hookup wire? That magnet wire was a bear to solder to as I recall.
Greg H.
-
Having had the same problem as Ken.... I'd check the solder joints, very closely, on the channel that is in question and re-seat the IC's for that channel as well. Once I did both, my problem went away.
Joe
I decided to trace the signal through the paths on the amp(s) checking corresponding levels when the gain was set to 4:4 for both amps. I used a bit of clipped-off resistor wire in the probe so I could carefully touch pins on the ICs (including the SMDs) without shorting adjacent pins.
Chan-B wasn't working because Pin 3 on IC6a (the input to the first filter amp) wasn't connected (despite it looking connected)
(http://saratoga-weather.org/2013-10-05_no-signal.png)
A re-solder of that connection caused Chan-B to produce the same output as Chan-A for all the gain settings (with a 20mV 10KHz sine input). So, it looks like the amps are both working now.
With the 120mm ferrite antennas connected, I'm now seeing an odd difference between the two:
(http://saratoga-weather.org/2013-10-05_odd-noise.png)
One of the ferrites is picking up a repetitive 60Hz signal that is associated with that one ferrite, and the signal moves to the other amp if the connection is swapped between the antennas. I'm trying to figure out where the signal is coming from, but no luck so far. I did apply E shielding (foil tape) to the antennas (with a 1/8in gap along the axis), and tied the foils together to the antenna ground pins at the amplifier -- seems to have no effect on the periodic signal.
@Greg: I think the ADCs are working ok since the Chan-B Amplifier is now working.
@Don: thanks for the suggestions .. I have tried them all, and the 'trace the signal' was the resolving diagnostic test :)
Still slogging through the debug, and I'm building a ball-loop antenna to try out also.
Best regards,
Ken
Ken, glad to see you are making progress. I am glad to see you haven't pulled out the sawed off to solve the problem. No doubt, I have less hair now then when I started my kit.
I am ecstatic with the success of my ball antenna. Just finished relocating it to the attic of my garage, another reminder that I'm no kid anymore, to get it away from all noise issues. Immediately, I had to crank down my gain. Another day or two to get the right settings.
Best of luck!
Joe
-
Ken,
Congrats on finding the bad connection! =D>
Since the problem moves when you swap the antenna inputs, I wonder if there is some kind of wiring problem with one of the ferrites? Have you ohmed them out to see if they compare in readings? The readings will be quite small. Another thought - have you scoped each of the ferrites (not connected to anything but your scope lead) to see if they both have the same output traces?
It sounds like you are picking up some kind on signal on one of them and not the other.
Keep going, you are certainly getting closer to be operational.
-
That's great. Looks like you are making progress.
It seemed that when you tested it with the generator it was passing the signal, so I thought that the controller board/ADC was the remaining possibility.
Glad you found it. As you said, on to the antennas...
Do the suspect ferrite's antenna wires have continuity with an ohm-meter? Maybe it is acting like a "wick" and the ground side is open? Is there a solder connection between the ferrite magnet wire and your hookup wire? That magnet wire was a bear to solder to as I recall.
Greg H.
I checked the continuity of both ferrites to their terminals (3.2Ohms) -- yes, had to carefully scrape the varnish off, tin the wire, then solder it to some Amphenol pins (gold plated) to use for the connections to the antenna block on the amplifier (also shrink-wrapped the connections).
A question: what does the knot on the lead to the ferrite mean? Is it the start of the wrap or the return? Should the knotted wire be used for the signal input or connected to the common ground on the antenna terminals? Or, does it really matter as long as both ferrites connect the same way (knot to signal or not)?
I've connected the Controller and Amp with a 50ft shielded Cat-6 cable so I could walk the antenna to various locations. The interference is stronger towards the kitchen area (even with RE1000 repeater, florescent lights and house phone unplugged, and family room TV/lights off). Signal is weaker in master bathroom at other end of main all, but still unable to determine where the interference is coming from. Grrr...
As I rotate the antenna in the horizontal plane, I can see the interference peak on one channel, then the other, so my prior statement about it not moving was incorrect. The signal is strongest on the ferrite pointing along the long axis of the rod. Does that mean the longitudinal axis is pointing in the direction of the signal, or is the perpendicular the direction of the signal?
It's been too long since EE school to remember that. #-o :oops: :roll:
Edit: found that answer The ferrite rod antenna operates using the high permeability of the ferrite material and in its basic form this may be thought of as "concentrating" the magnetic component of the radio waves. This is brought about by the high permeability μ of the ferrite.
The fact that this RF antenna uses the magnetic component of the radio signals in this way means that the antenna is directive. It operates best only when the magnetic lines of force fall in line with the antenna. This occurs when it is at right angles to the direction of the transmitter. This means that the antenna has a null position where the signal level is at a minimum when the antenna is in line with the direction of the transmitter.
So the perpendicular points to the signal source (and 180 from it). Search is your friend :)
-
This chart is very helpful. I think that RichoAnd posted it on the main Blitzortung forum.
On the ferrite rod antennas, the pattern is perpendicular to the rod. On electrically small loops the pattern is in the plane of the loop.
I had to make myself a cheat sheet for this (and spent a lot of time looking it up, but Richo's chart is better).
Deep in the Blitzortung PDF, there is a mention of the "knot". The convention is to make that the positive (or signal input for Red). Richo also mentions that it really doesn't matter on the posting Antenna question over there on the main forum.
-
To further define the meaning of "direction sensitivity", are we referring to "magnetic" or "electrical" field? And, when one is trying to find a source of interference, would one rotate the ferrite antenna so it was broadside to this source. For example, if I have a source of interference that is due North of my location would I find it by rotating the ferrite antenna so it pointed towards the north or would I point it so the antenna is broadside (90 degrees) to the source of interference?
Hope I'm making sense here but it will help all of us to correctly orient our ferrite antennas so we can identify sources of interference.
-
I decided to build one of the Ball loop antennas using a 10in solid foam ball from Michaels' craft supply store. The antennas are shielded by foil tape and consist of 30 turns of 20gauge magnet wire. I've not noticed much difference in reception from the default 120mm ferrites that Egon sent. Also tried terminating the antenna leads at the amp end with 100ohm termination resistors.. no change to speak of.
(http://saratoga-weather.org/LoopAntenna-sm.jpg)
Here's the 60Hz noise I can't seem to locate:
(http://saratoga-weather.org/2013-10-08_loop-noise.png)
By rotating the ball in the horizontal plane just so I can minimize the noise spikes and crank up the gain to 10:4 currently. So here's a better look, and you can just see the interference showing up at a minimum.
(http://saratoga-weather.org/2013-10-08_loop-reduced-noise.png)
I'll have to port the antenna/AMP around the house some more to attempt to pinpoint the source of the interference. Thanks for the education about directionality of both the loop and ferrite antennas -- that is very helpful!
Best regards,
Ken
-
What gain are you running at? If you're running at 10/4, under the current weather situation you're not going to get much.
When I first started my test on the 12 inch (30 turn, 18ga wire)...I started off at 10/10 and ended going up from there.
I just built an 8 inch (45 turn, 18ga wire) ball antenna on Friday to see how low I wanted to go in regards to the size. The starting gain was much higher...around 16/16.
I'd probably start at 16/10 with yours and go from there.
Lance
Huntingtown, MD #724
-
Hi Lance,
Thanks for the gain advice -- I will crank up the gain as soon as I can isolate where the strange 60Hz spikes are coming from. I've tried running higher 10:4, but currently, it only moves the controller into interference mode due to the false-positives provided by the spikes :(
Best regards,
Ken
-
Ken,
Based on what I've experienced in the ham radio arena, sometimes, when attempting to locate an interference source, you will get better results by rotating the antenna to get a null in the readings. Nulls tend to be sharper (narrower due to them being lower in amplitude) and thus easier to find than a peak in signal.
I would look for a wall-wart (transformer used for all kinds of devices today - ie, laptop power supply, cell-phone charger, etc, etc), some kind of light dimmer or even the door-bell transformer to be your problem source. Do you have lamp in your door-bell button outside? All of these things could be sources of interference. Is this 60Hz signal present day and night or just at some particular time of day?
You may also want to begin turning off the circuit breakers for different areas in your house. That would help narrow down the location of the circuits causing the problem. Sorry, I know that means resetting your digital clocks but...
Good luck finding this elusive signal...
-
I would try using an AM radio to see if you can find anything. I have had mixed results with this, sometimes the radio finds noises that are not related to the lightning detector response.
I have used several different "sniffers" and have been able to eliminate some noises. I have an infrequent severe noise of unknown origin.
I noticed that my "green" detector responded and went into interference mode when the wife was running the vacuum cleaner. It could be a motor. Cutty has night-time outside light problems. Unfortunately, we live immersed in RFI from a lot of sources. Best of luck tracking them down, W3DRM has some good advice there and should give you some ideas.
Greg
-
I found the best sniffer for noise detection is the amp itself. Try connecting the monitor pins on the amp to a stereo headphone jack and plug the amp into a high-impedance input audio amplifier and listen to the signal. It is very easy to distinguish normal low-level 60 Hz background hum from noise from motors, 60 Hz harmonics from a bad "wall wart" power cube or other sources. With stereo earphones and each antenna connected to left and right channels of the amp, the nulls from the antennas are a very good way to trace the direction and type of interference. I used a simple "CMOY" headphone amplifier for the purpose. The CMOY is a simple single-chip headphone amp. Try Googling for the schematic. A PC soundcard would work, but then you are dealing with processor noise from the computer. An old set of PC powered speakers may work well.
Don
I would try using an AM radio to see if you can find anything. I have had mixed results with this, sometimes the radio finds noises that are not related to the lightning detector response.
I have used several different "sniffers" and have been able to eliminate some noises. I have an infrequent severe noise of unknown origin.
I noticed that my "green" detector responded and went into interference mode when the wife was running the vacuum cleaner. It could be a motor. Cutty has night-time outside light problems. Unfortunately, we live immersed in RFI from a lot of sources. Best of luck tracking them down, W3DRM has some good advice there and should give you some ideas.
Greg
-
Here's a link to the CMOY article. This site includes a complete, and detailed, description of the CMOY amp and a full listing of where to get the parts. It appears to be well done and should be easy for anyone to build.
- http://tangentsoft.net/audio/cmoy-tutorial/ (http://tangentsoft.net/audio/cmoy-tutorial/)
-
That circuit has a gain of only 11. (1+(10k/1k)). Just less than 21dB.
Is that enough to hear anything?
It might be for a strong enough noise source. If the amp remains stable, you could increase the 10K feedback resistor to get more gain.
Greg
-
This thread ended without a resolution. What did Ken finally find to remedy the problem?
-
Oh, this was 7 years ago.. I don't rightly remember what I did to get it all working again. The controller box was relocated (and dual ferrite antennas relocated) a bit within the garage and that seemed to fix the 60Hz issue I was having.