Navigation Links

Upcoming Events

Current Events

Social Junk



Quick Search
Advanced Search »

Contact Info

David Robbins, K1TTT



2020 Maintenance and Upgrade Blog

Schematics Index

1/1 Well, the year started with everything covered with ice.

1/3 Most of the ice finally melted off the antennas. Only damage found so far is to the north 160m Inverted L that has a broken feedpoint and a tree down on one radial. I cut the tree off the radial. And some of the 40m 4-Square radials need to be tied back up after the trees lose their ice.

1/5 Walked the dogs around the Beverages, only one big branch down on the NW/SE one, I was able to break off enough to get the wires out from under it but it will take a chainsaw to get rid of it after it thaws out from the ice on the ground. It only took down 2 insulators from the very end of the Beverage.

Found the top fiberglass radome piece from the 180' 2m vertical on the ground. That antenna is only being used for ADSB skimming right now. It will probably slowly degrade as it fills up with moisture.

1/6 Fixed north 160m Inverted L feedpoint.

1/11 Warm windy morning, cleaned up some more broken branches from earlier ice storm.

1/12 All the ice and most of the snow is gone now, temperature still in the 50's and windy. Did more cleanup of some big branches down in the 40m RX Array area.

1/14 Spectrum residential service is being installed in town now. Called their Small Business sales line and could not convince them that there was fiber on my street. Figured I would have to wait for the residential sales rep to knock on the door before they would update it in their network database.

1/16 Got visited by Spectrum Residential sales person... But I had already found that the residential acceptable use policy forbids servers, reserves the right to throttle you if you use too much bandwidth, and can't supply a static ip address, so on to Small Business service again. No, the door knocker rep can't sell me small business service. After 2 days, 4 first line Spectrum Small Business sales people, and 2 supervisors I finally convinced a supervisor that there was fiber on my road and that he could sell me small business service over fiber instead of having to install Spectrum Enterprise fiber at a cost of $10k and $1k/mo. The difference is that Enterprise fiber is dedicated and provides symmetric speeds, Small Business fiber they refer to as cable service, even if it is on a fiber, and it is asymmetric like 200/10mbps, and the prices are more like residential, but their TV packages are horrible... think doctor's office and airport tv style services, or all sports all the time bar channels. I ordered the base internet service at 200/10mbps and one static IP, scheduled to install Sunday 1/19.

1/19 Spectrum Small Business internet service set to be installed between 1-2pm.... they called me after 2pm, tech is sick, rescheduled for Tuesday 10-11am.

1/21 Continuing saga... no call by 11:30am, called the sales rep, he called dispatcher, dispatcher called an hour later said she would try to contact the tech, sales rep called back said tech was still sick, and rescheduled for Thursday. Signed up for Sling TV, activated my Amazon Prime streaming service, and my old Roku free channels, that should keep me in TV for a while along with my over-the-air antenna.

1/23 Will they show or not??? 3rd times a charm isn't it??? Stay tuned, film (if I can upload it) at 11pm. Why do I feel like I'm living an episode of South Park???

Nope, no internet for you! Tech showed but says he can't install business service and static ip's won't work, the network infrastructure is only set up for residential users.

1/24 Doing more research on Spectrum service. Doesn't look like Small Business group is figuring it out so decided to get at least the residential internet so I could get rid of Directv and surf the web faster. Wouldn't you know it, as the Residential salesman was pulling out of the driveway with my order in hand the Small Business salesman supervisor calls back and says they had a conference with the installer supervisor and the engineer that designed the system and figured out how to make it work, and scheduled me for first thing Tuesday morning. So call the Residential salesman back and tell him to hold the older till Tuesday and I'll get back to him... I give them a 50-50 chance of it working when they plug it in, but hopefully with the engineer involved they'll figure out the right router settings to get service here.

1/26 The rain last night seems to have finished off the 180' 2m vertical, at least as far as ADSB on 1090MHz is concerned. It probably got water into the radome and now it is a water filled resistor for 1GHz... the SWR on 2m doesn't seem too badly affected but its probably not as efficient as it should be.

1/27 Have Spectrum scheduled to install tomorrow, this is the 4th appointment, hopefully they can do it this time. However, a Spectrum truck showed up today to run the outside cable. These guys did good, they used my snake to get the cable through the conduit from the garage and did it overhead from their carrier cable to the back of the garage.

1/28 Installer shows up on time, runs cable into the shack, puts box on outside wall for joining the outside and inside cable, does all the connections, hooks it up, it comes on and has internet access... but no static IP address. Takes about an hour to get to this point. 7 hours later, another modem, and 2 more routers, and who knows how many techs back in the offices, he finally finds a tech who knows how to do it.

1/31 Friday 4:30am internet stream stops. Call at 6am, convince support tech that its brand new service and Saturday is too late to get someone here so he escalates it and says someone will call between 8-9:30am to tell me when tech is coming. 8:30am they call, say they are working on it and will call back in 20 minutes. 9:30am, no callback so I call them, support says they will check on it and call back. 11am no callback, call them again, support escalates to supervisor without me asking. 1pm no callback but a truck shows up. Its a tech supervisor. No signal in shack, none in junction box, but there is on the pole. He cleans pole connection, still nothing in box, so he remakes both connections there and now there is signal in the shack... took all of an hour. As he is leaving I get an automated call saying I'm scheduled for Saturday 10-11am, it takes me about 15 minutes to get them to cancel the appointment because its already fixed!

2/2 Walk dogs around Beverages, everything fine.

2/8 Another ice storm yesterday. This time it didn't warm up before the cold front and winds came through so today everything is still iced over. Antennas all still up in the air, a couple rotors are frozen, and SWR bad on everything. K1MK having to make do with what it is for the WPX RTTY contest. Power was out for a couple hours just before the contest started but the generator was running well.

Took the dogs for a walk around the Beverages in the afternoon. Found big tree down across the NW/SE wires about 100 yards back, this will take a while to get out of the way after the ice melts a bit.


2/14 While getting ready for ARRL DX CW I found a bad jumper from the radio to filter on the 20m station. There is also bad noise from 40m on both 20m and 15m.

2/16 15 minutes before the end of the contest the 40m amp faulted with an ARC FAULT error. It affected all antennas and would reoccur on almost every transmission. With the amp on standby the radio was reporting high SWR, most of the time... but RX was still there and sounded ok.

2/17 Wouldn’t you know it, when I started up 40m Monday morning it was working fine. But after warming it up a bit and pushing and pulling on some cables it started to fault. I went to low power just because I don’t like faulting the amp too much and found the cable from the amp to the SWR bridge would go open if I bent it just the right way at the back of the amp… where it makes a sharp bend against the wall behind the amp and would have been flexed by bumping the table. It would also cause the RX to go deaf after transmitting just like a bad T/R relay which had been reported as a very intermittent failure that I couldn’t find before. I replaced the connector on the amp end and now I can’t make it fault. I did a quick check of the 40m to 20m and 15m noise and it seems to be not as wide as it was, will have to recreate my detailed test and see how it actually compares.

2/18 I hate intermittent faults! 40m/SO2R problem came back today. Wiggling the coax on the amp output, that was just ‘fixed’ caused loss of connection. Took amp down and disassembled it. The SO239 looked OK but I reflowed the solder joint anyway, it still was intermittent so I pulled out the connector. Checked it out, it wasn’t broke anywhere, nothing on it that really could break. Removed all old solder and resoldered it back in place and it tested ok. Put it back in service and it is working fine again.

This is a closeup of the inside of the SO239, when zoomed in you can see cracks and maybe discolored solder in the joint. It would appear the solder didn't wet the coax well enough.


Now if we could just stop with the ice storms!

2/24 Made a quick climb up to 60' on the 20m tower to realign the TIC Ring Rotor after it slipped in one of the ice/wind storms.

3/16 Tried some new tests on the Beverages. I used the XG3 signal generator and a PL259 that has about 3" clip leads from the center pin and shell on it. Took it out to the pole and hung it from the input terminals on the 2 wire Beverage boxes. I did it 2 different ways, first just between the 2 input terminals, this stimulates the terminals out of phase which provides the signal on one of the outputs. Then I switched it and put the shell lead to the ground terminal, shorted the 2 input terminals with another test lead and put the center pin lead to the shorted inputs, that stimulates the other output line. All the boxes seem to be working fine, but I did find a broken wire on the NW/SE box input.

Pulled up some radials on the field side of the 40m RX Array and used the brush mower to cut some of the thick stubble of the buckthorns I cut with George's hedge trimmer in the fall.

3/17 Put the radials back in place.

3/21 Cut apart the big part of a tree that was hanging across the NW/SE Beverage trail.

4/15 Two Beverages (4 directions) sound pretty weak. Run though them with the XG3 signal generator and measure resistances and determine 2 of the transformer boxes at the shack end aren't working right and their reflection transformers are also bad. Lightning storm the other day must have gotten them. Replaced one box with my spare, then repaired it with my home made transformer replacement, and put it out in place of the other bad one.

4/17 Dressed for bear and snow and ventured out into the woods to replace the two reflection transformers on the Beverages.

4/23 I noticed a couple days ago after a big wind that the 60' high 20m Yagi was pointing the wrong way again. Usually in high winds if it is pointing West a strong North-West wind will make it slowly drift South and often past the South stop, and the indicator would stay calibrated... This time it should have been pointing East and was instead pointing South but the indicator was still saying East. This normally only happens if someone tries to turn it when there is ice buildup on the ring so the drive gear slips, but I didn't think that had happened. Anyway, today was a bit breezy but not too cool so I went up to realign it, about a 20 minute job... but when I was reengaging the drive gear I found that it was slipping teeth internally only in one direction, so I locked it down and retreated to the spare parts shed. I pulled out the universal spare that I had bought for just such an occurrence, carried it up, and installed it in place of the old one. A half hour later I had George rotate it around both ways and I came back down for the second time. Autopsy results later.

Well, easy to find cause... the setscrew either loosened or gouged out the flat part of the shaft it is on so it wasn't really slipping it was turning a tooth's width then sticking... still bad. I can tighten it and it seems to stop slipping but if the shaft is damaged will it hold or just start eating away at it some more. Unfortunately the bracket above it is welded to the plate below it so the only way to really take it off is to take the plate off the gearbox and pull the gear up over the smaller gear at the same time... or I could just take the set screw out, add some Loctite, and call it done.


Well, took the set screw out and was able to inspect the flat part of the shaft, but it seemed too buggered up to make taking the gear off the shaft reasonably easy... So I broke out the Loctite and put it back together. I do need a new pin for the main drive gear on the top of the output shaft, it was half way sheared, but otherwise it seems to be in reasonable shape. After I get a new pin for the main drive gear I'll run it with as much load as I can put on it to make sure nothing else is slipping.

5/21 Took a trip up the 150' tower. Tied the 120' 20m European Yagi back down, it was moving about 15 degrees in the wind when it shifted. Fixed windy intermittent in the 150' 80m Inverted-V. Taped up and inspected everything on the way down. The indicator gear for the 60' TIC Rotor must have slipped, had to move it's stops to 120 degrees or it would complain about the pot being out of range... that will be another trip up there for it.

5/29 The bottom TIC Ring on the 10m tower started acting like it was stuck. Went up to it and found the indicator gear had slipped down and disengaged from the main ring, so it was trying to turn but the GreenHeron controller would not see it moving and stop it.

6/1 K1MK was here doing WPX CW. Sunday morning PTT on the left radio became intermittent. I tried pulling cables and checking what I could but could not figure out what was going on. We put the radio in Break-In mode and turned the delay up so it wouldn't chatter the amp relays and he finished like that. Today I started pulling things apart. There were 2 failures... first the ground on an RCA Y cable being used to bring CW keying from a separate WinKeyer lost a ground connection, I think that may have been that way for a while, but it was being masked by a shared ground on the PTT line. From the YCCC SO2R box the CW and PTT lines share a ground pin so losing one or the other was not fatal, but what I think was happening is the impedance got high enough on the PTT ground so that it could not key CW AND do PTT... but there was enough sinking current for it to key the CW, but not enough to reliably do PTT. I took off both RCA connectors, took a 3' RCA jumper (commercially made with molded connectors) and cut it in half, and spliced those 2 pieces onto the cables from the SO2R box which were a little bit short anyway. So now the resistance is back to normal and I have longer cables to the left radio to make future service easier.

6/15 Went up and realigned the 60' 20m TIC rotor again. Found part of the rotor loop had cracked insulation and taped it up.

6/28 The big UPS that protects the ACOM 2000A amps was warning of batteries reaching end of life and had been running it's fan because the batteries were getting warm... well, it was right, not long before the end of Field Day it started beeping about battery failure and disabled itself. I took it off line and pulled the batteries out, the metal cases holding them were too hot to hold! While I have it out I popped the cover and vacuumed it out. This is what it looks like inside:


These are the battery packs. In each one are 8 12v SLA batteries all in series. The batteries in hot swapable holders sell for something like $600, but you can buy the 16 individual cells for a total of $200 with free shipping from Amazon.


7/1 Cells in the UPS battery packs came in and were replaced. The CP-1250 cells were an exact fit. I had to replace one terminal connector that was corroded from a cell leaking but otherwise they just slid in and connected right up.

7/8 The touch screen monitor on the SO2R station used for rotor and antenna switch control has been getting flakey. It was a Planar PXL2230MW which uses an optical touch system, that means it sends infrared beams across the front of the monitor and detects when they get interrupted. The problem is that dust and dirt can get into the grooves where the IR LEDs and detectors are causing false detections, even a gnat landing on the screen can trigger it. I replaced it with a real capacitive sensed Planar PCT2235 monitor... it is physically smaller because it has a very thin bezel but has the same display size. It seems to be maybe a little slower responding to touches but is very reliable so far, also the display seems a bit brighter than the old one.

7/23 I had figured out how to use the 'What you hear' or 'Stereo mix' from the sound cards for simplifying the audio routing, especially on the SO2R/Remote station... but today the SO2R station has been plagued with audio issues. N1MM+ reporting it can't open the recording device for SSB messages, got that fixed, then WSJT-X can't open it's audio input, fixed that, then Mumble has no audio. Pull everything apart and put it back together and suddenly its all working. Then overnight nothing is working.

I did find a way to digitize my existing Daiwa power meters. Add on a $13 USB 10 channel 12 bit ADC and a little bit of programming.

I just connected two channels of the ADC to the Daiwa meter sensor output and write a little app to read the data it streams, and away it goes.



7/28 Did some more playing with voltage divider resistors to keep input voltages in range of the ADC chip (0-3.3v). Bought a resistor assortment box, almost 5000 resistors, I need 20 or so of them.


Here are 4 of the USB ADC boards ready to get wired into the meters. I decided to add a temperature sensor also, so am waiting on delivery of some resistive temperature detectors(RTD). On the bottom of the ADC boards I have two 51kohm resistors going to ground as one end of the power input voltage dividers, the 2 resistors on the terminal strip are 10kohm for the input end. The other resistor on the bottom is 1kohm connecting input 2 to +3.3v for powering the RTD sensor. The RTD sensor will go from input 2 to ground on the terminal strip.


The calibration is done with an Excel spreadsheet and some extra outputs on the app. With the meter connected normally to a dummy load I run the power up from 0 to 1500w and record the voltages the ADC measures. Then I reverse the meter connections and go from 0 to 200w and record them. The graph shows a quadratic curve fit equation which I put into the app setup screen for forward and reverse power. The temperature curve requires the constant part of the equation and is just calculated from the RTD characteristics for now.


This is the app and it's setup form showing all the controls and outputs needed for calibration. In use it would be shrunk so only the three bar graphs will show.



7/29 I love it when a plan comes together... and thanks to DigiKey the RTDs showed up a couple days early today. I attached the RTDs to the cables and installed the completed boards on the 80m, 10m, 15m, and 20m stations. The calibration from the first one was close enough for the others, so now I can monitor the amp output, SWR, and temperature on all 6 stations. Those 4 won't auto tune like the ACOM 2000a's on the 40m and 160m stations but at least remote ops can monitor the health and call for retuning if needed.

8/9 Last week's tropical storm finally broke off some of the dead branches from the last big ice storm. I had to take my long reachy pole down to the 80m 4-Square to pull some of the out of the trees and off the Beverages and radials.

8/13 Spent about 2 hours up the 180' tower removing the broken 2m antenna and replacing it with a new Discone.

8/30 Nice cool morning, Hank is off his restrictions from the knee operation so took a walk around the Beverages for the first time since spring. There are 2 big deadfalls across the North-South path near the 80m 4-Square but otherwise everything looks good.

8/31 Calm and cool morning, decided to fly the drone around low to see if I could find where the weird 20/15m harmonic mixing is occurring. Well, I need more practice at low level flying around here. The drone lost a fight with a guy wire that resulted in 2 props and minor case damage on the Raspberry Pi. Then after that it also lost a fight with a tree that broke a landing strut and cracked a couple of other things... total of about $100 damage for all the parts. Data from flights wasn't very interesting, but it looks like the mixing product is near the shack.

This mixing product is a weird one... it is either 2*20m-15m or 2*15m-20m... so 14010*2-21010=7010 or 21010*2-14010=28010. Both products are fairly clean, we actually decoded FT8 from the non-second harmonic station on the resultant a couple times, but you can see the second harmonic widen the audio in the WSJT-X waterfall.

Now that was fun... when the drone hit a tree it landed hard on a pile of logs I put over a stone wall to protect against ricochets on my rifle range. One of the front landing struts broke off and I couldn't find it during the day. But just at sunset I put on a high intensity LED headlamp and went back to the crash site. First I found a piece of clear plastic from the PI case, then about 6" from that I round the broken piece of landing strut. It looks like it could be welded back together, so that will likely be a spare part after the replacement shows up.

9/5 Cut down the big rotten stuff that was over the N-S Beverage wires in two places near the 80m 4-Square.

9/8 Parts came in for the drone and Raspberry Pi case repair. Spent a couple hours replacing both rear propeller arms and one front landing strut. Then replaced the UPS hat on the RP and rebuilt the top of the case that was cracked.

All better:


Speaking of 'all better', after almost 3 months of spending hours on the phone with Spectrum tech support, working my way up past 2 layers of 'leadership teams', and finally getting a manager(they wouldn't call him a manager but I know he is because the first thing he did was to call a meeting to figure out who was to blame) involved, they finally fixed a weird routing problem that was preventing local Spectrum residential users from getting to my web site (and other services). Turns out they needed to update their Adtran head end equipment.

9/11 Borrowed George's saw on a stick and started trimming branches hanging over radials around the 80m 4-Square.