2016 Maintenance and Upgrade Blog

Modified on 2017/01/06 18:19 by Administrator — Categorized as: Uncategorized




1/1 HNY.

1/8 Got some updated software from GreenHeron to try out, this lets the rotor compass display work on the right radio. I also figured out how to make the dual displays so the left and right radios have their own sets of controls. Still need to do some adjusting of sizes and stuff but this is the basic layout. The RX Antennas are common to both radios so those are in the middle whenever either radio needs them. Then the left radio's rotors and switches are on the left and the right radio's are on the right. Also when I changed the colors on the rotor map it put color bars to match them on the rotor numeric display off on the right side so now they are easier to match up.

Image

1/15 Bottom 20m TIC Ring must have water in the motor, when its below freezing it won't turn. Also the indicator is off by about 30 degrees.

1/17 Ice, rain, snow, sleet, wind, cold, yep, its winter. Took another stroll around the Beverages, nothing much going on out there now.

1/22 Argh... 15m intermittent high SWR... tracked to 75 ohm matching section at end of hardline outside the arrester box. There was also some water in the connector. Cleaned out the water and got it working, contest doesn't start till morning anyway and got it figured out about 8pm when temperature was 8f.

1/23 Replaced 4' RG-11 matching section and the T connector for 15m. On the hardline end of the jumper the PL-259 was loose so it probably wasn't soldered inside and was intermittently open. Temperature about 7f, had to warm up the duct seal before taking it out to reseal the connectors.

3/4 Just for fun... LED Bulb deconstruction

3/12 Today I took down (well most of the way) an antenna that hadn't been used in years, and one of the supports for it. Only the most observant visitors may notice it's absence.

3/26 The top 40m rotor controller blew it's fuse again. Replaced it and it blew again, time to troubleshoot now that its broke for sure. When I start working on something like this where I am going to have to disconnect wires I always take a few pictures so when I go to put it back together I can look at them to remind myself how it was connected. This is the inside of the K0XG motor power supply.

Image

In this case I also had to verify the schematic as the documentation contained several variations for different uses. The motor for this rotor is a large DC motor, it runs on about 90vdc which is supplied by a full wave bridge directly across the AC input. The 24vdc control signal from the GreenHeron box switches two small relays which then use DC from a different full wave bridge to control the big relays. The problem was one leg of the full wave bridge for the motor was shorted so as soon as it was turned it blew the fuse. Part is on order.

3/28 K0XG sent me 4 bridges. Put one in, turned it on... click, pop, bridge and fuse blow again. Measure more stuff, disconnect motor cable. Put in another one and another fuse. Everything clicks and clacks the way it should in the control box. Go out and disconnect the MOVs in the lightning arrester box. Use TDR to ring out cables using the now spare potentiometer cables as a reference and everything looks ok. Plug back in motor cable... click, pop, bridge and fuse blow again.

3/29 Wake up to note on remote computer that bottom 40m has high SWR... Go out and take a look, yup, its broke. The front driven element loading support strut is broken off so the linear loading lines are touching the element so that antenna is out of commission for now.

In the evening K0XG suggested trying 12v on the motor. So I pulled a battery out of a UPS and hooked the ohm meter to the pulse detector terminals on the controller and powered it up... YES, we have a heartbeat. It's slow, between 1 and 2 rpm but it seems to turn both directions. So that would imply that the motor itself is working and the cable is connected all the way up the tower. So now what????

4/1 Well, it wasn't raining but it was a bit breezy this morning, but I went up to the 70' 40m Yagi and put a splint on the broken strut so it won't damage anything else from laying on the element and feed point and blowing around in the wind. It should also make the antenna usable again until I can do a permanent replacement.

4/9 Parts came in yesterday so I built a protection circuit for the bridge rectifier in the K0XG power box. I hooked a scope up to it last week and was seeing rather large kickback spikes when the direction control relays were closing. With only 24v from a couple of UPS batteries driving it the spikes were off scale on my Fluke's max/min, and were at least 200v on the scope. My theory is that when the spikes hit the bridge they forward biased a diode that should have been cut off causing a short across the AC line input thus blowing the diode and then the fuse. My circuit should prevent those spikes from reaching the bridge and reduce the size of them with some capacitors.

The circuit is:

Image

There was room to mount it next to the big direction control relays on the wall of the box. The protection circuit is built on a 4 terminal strip on the copperclad board on the right wall. In the middle the big relay is one of the direction control relays, the other one is just off the bottom of the picture. To the left at the top of the white panel is the rectifier that kept blowing, the black and white leads with the pink insulators are the AC from the power relay on the bottom left side. The DC terminals are clear wire with red or black tape to keep the sorted out. The green taped wire makes sure there is a good ground connection to the board and capacitors.

Image

So far it is working. I just don't like how warm the series diodes get. I can't mount them on the circuit board because they have an exposed metal back that is common with one of the terminals, so they would need to be insulated with mica or some other material.

4/16 First nice weekend day in a while. Went up to the 60' 20m antenna and realigned the rotor, it must have slipped when there was ice. Now that it is mostly above freezing the motor should turn most of the time.

4/30 George came over and we took down a couple dead trees by the 80m 4-square.

5/31 Retired.

6/4 Staked a bunch of Beverage and coax support posts that were flopping in the breeze. This involves cutting 18-24" 2x4s with one end cut at an angled. Those stakes are then driven into the enlarged hole around the base of the posts to keep them from moving as much. Did some on the S, SW, and NW Beverages in the field, the feedline support to the 15m tower, and the feedline support to the 20m tower.

6/9 Several years ago I bought one of those single solar panel emergency power supplies. It is basically a single solar panel on wheels with a long cord, and a box that has three 22ah 12v SLA batteries in parallel, a solar charging controller, and a 120v charge controller and inverter. It can be charged either from the solar panel or from 120vac. The inverter in this one is really bad, its just one step better than square waves so some stuff doesn't like it. A couple years back the batteries stopped taking a charge so I replaced them. When I did that I ran a 12v RigRunner power cord out through a hole I cut in one of the air vents. This was good as it let me run the TS-2000 off it directly... but we have the technology to make it better! Since today was about 50f with 20-25mph winds I decided to operate.

This is what it looks like to start with. The big yellow cord is a short stub that the solar panel plugs into. The AC charging cord comes out the end to the right which is also where my 12v cord initially came out.

Image

First some safety stuff... the 3 batteries are fully charged and contain lots of energy so we have to be really careful to not short them out. First unplug the 120v charger then turn off the controller on top so the inverter is off.

Explode it... this means taking off the two side panels (4 screws each) and the right end cap (4 screws on the sides and 2 underneath). the red&black pair is the one I'm going to move.

Image

Since the cable I'm going to move has PowerPoles on it and is directly connected to the batteries I very carefully cut one wire at a time and taped over the red (hot) wire so it couldn't short against the chassis or anything else while working with it. It would be easier to do initially if it weren't connected to the batteries, but that means MUCH more disassembly of the internal parts.

Image

Then since I had extra wire and some ferrite cores I wound up the extra around a big core to help filter it a bit from noise created by the charger and inverter. Fortunately there is plenty of room in the end compartment.

Image

The cable runs out of the chassis, through a new hole in the side panel (which should have been closer to the existing one to avoid having to bend the wires around to fit. I had hoped to put it through the tube that was already there but it was too tight.

Image

Inside the solar charge controller box the wire goes through a vent slot that was enlarged by removing a rung with wire cutters.

Image

I installed new PowerPoles on the wire. I mounted a RigRunner on the side with 4 short screws backed with lock washers and nuts. The screws have to be kept short so they don't stick out beyond the ribs on the inside of the side panel or it won't go back on properly. This is what the final product looks like.

Image

Now it can power multiple devices if needed and doesn't have a long 12v cable hanging out one end.

6/11 Finished up 6 new chokes for the 20m antennas to replace the ferrite baluns. The left radio on the 40m/SO2R station was doing odd things. It appeared that the DSP modules were dead as any use of DSP for RX or TX blocked the audio. I pulled the pair of DSB boards in their little steel box out of W1TO's dead radio and plugged them in, one of the tests still failed, but some of them worked... a double failure??? To be sure I put the original one back in, and then everything worked. Probably a dirty or loose connection on one of the 3 connectors that it has, probably one of them that supplied the digital part with it's control or timing signals.

Finished up new walk around the north end of the house. Picture is before adding more dirt and seeding, but is pretty close to done.

Image

6/16 First nice day to climb in a while. George came up and watched me replace the broken truss on the bottom 40m Yagi.

7/9 Ok, party is over, things are finally cleaned up. Now back to getting other stuff done. Last Monday, I think it was, the hard drive on the 10m computer crashed. This is one of the 5 identical ones so instead of reinstalling from scratch I decided to try to clone the hard drive from one of the others. New identical drives are something like $40 on Amazon, and an Inateck FD2002 USB docking station and cloner was $25 or so. The Inateck was supposed to clone by itself, but that feature turned out to be slightly less than useful as it requires the new drive to be bigger than the old one, exactly the same size won't work for some reason. So I found some free software that uses the USB docking capability in the FD2002 to clone drives that are the same size. It is EaseUS Todo Backup Free... they do have relatively cheap paid versions but I didn't need any of the added features. I cloned the 80m computer, plugged it in, changed the IP address and computer name. The only other change was that I had to reassign the com port for the Winkeyer in the device manager since it thought it was a new device and gave it the next higher com port.

Also sent in FD report and uploaded the K1T logs to LOTW.

Oh, and in between other stuff over the last two weeks I have moved everything from the old server to a new Dell server with Windows Server 2012R2. Only a few minor changes needed to fix some web site configurations to bring them up to date.

7/10 Installed new server in rack, left it headless.

7/12 Installed new RCS-8V on the 80m stack. The old one contacts were only slightly burnt. Found the hardline connector was not real tight, had a bent prong on the center conductor pin that was close to the shied, but didn't seem burnt. Built a new connector for the hardline after I came own.

7/13 Installed new hardline connector on 40m stack cable. Tried to level the bottom 40m beam, but bolts were too badly rusted to do it.

7/15 Got new motor and indicator from TIC yesterday. This morning before it got too hot George came up to press buttons and play with Pablo while I went up to replace the one that kept freezing last winter. It took me 2 trips to 60' because there was a problem with fitting it to the triangle support frame. Here is a view from above showing the supplied bracket up against the welded corner bracket of the frame. Note that the drive gear (under the orange safety cover, is not engaged with the ring gear.

Image

A view of one of the other corners that shows the same corner bracket.

Image

The third corner is worse because it is a longer bolted joint.

Image

The fix required me to come down and get some longer bolts and extra washers. This shows the modification to push the motor out to engage the ring. The angle piece is supposed to be up against the channel, now it is spaced about 3/4" out from the channel on the corner end.

Image

The other end of the motor only got a single extra plate, that moved it enough to make the drive gear engage the ring with a bit of room for adjustment.

Image

The pot enclosure bracket was supposed to be used as the backing plate for that end of the motor, but that put the slide for it directly on the Rohn 45 diagonal brace, so it was moved over and mounted separately.

Image

There was a tiny Allen head set screw in the bottom of the indicator gear. They supplied a wrench for it which I assumed was for height adjustment. When I went to adjust it up some I could not move the gear. So instead it got a couple washers to tilt it up and give better engagement with the ring gear.

Image

Some more 'after' shots. Top of the ring showing brackets and spacers.

Image

Relationship of motor bracket and pot enclosure. The tape in the middle is holding the motor power connector that goes between the pot enclosure and the motor.

Image

7/17 Bought a Red Pitaya to try out to see how it would be as a Skimmer SDR. I set it up with the software that makes it look like a Hermes radio. The results are not so great in first tests. Running CW Skimmer Server it was generating many fewer spots and the ones it did generate that matched up with the QS1R Skimmer Server were consistently showing 3-8db lower SNR values. This morning I got both set up to use cwsl_tee to split the data between CW Skimmer Server and HDSDR. I put my XG2 on 1uv on 40m and ran it through a DX Engineering splitter to the receivers. Below, the QS1R is on the left and the RP on the right. Note that the noise floor is about 5db higher on the RP. This agrees with the Skimmer Server SNR reports from signals decoded by both of them. Someone with a more controlled environment needs to repeat this though. Some possible causes:


Image

7/30 Disassembled the old motor from the 20m 60' TIC Ring Rotor to see if it could be used as a spare. This one obviously had some water in the motor/gearing because when the temperature was below freezing it wouldn't turn, but when it warmed up it was fine.

This is the outside of the case showing the drive gear. First part of disassembly is to remove the spring clip on the end of the shaft and also push out the roll pin through the gear. There is a hole in the back of the surround to make it easy to remove the roll pin. If the motor was damaged so it couldn't turn that could be a problem. Also remove the 2 slotted hex head screws on each end and the 6 machine screws on the bottom.

Image

Then it should come apart like this. The motor is on the right and is mounted to a cast aluminum gearbox on the bottom. The output shaft from the gearbox comes up and goes through a steel gear that has a set screw in it and to a press fit brass gear that drives the pot. The output shaft is just in a hole in the mounting plate so it is kind of floppy at this point. Removing the shaft would require taking off another spring clip that is hard to get at under the mounting plate. The small circuit board taped up under the pot is a noise filter for the motor.

Image

To get to the gear train remove the 6 torx head screws from the bottom of the case and gently pop it off. This is the inside of the bottom cover. Note all the red rust deposits in the cavities, it was dry at this point but obviously had been wet for a while.

Image

This is the gear train. The motor input is on the left side, the small gear that is partially hidden. The output shaft is on the right. I popped off the spring clip on the gear but could not get it off the shaft, it is keyed to it and probably frozen from rust as both the gear and shaft are steel. I wanted to see the bushing under it so I loosened the set screw on the other end of the shaft and could push the shaft out of the bushing with only some light persuasion. The brass gear that drives the pot just stayed on the shaft and fit through the gear.

Image

I cleaned up some of the old grease and rust deposits on the bottom cover and drilled two drain holes in the plate, one near the motor end and the other near the output shaft. I also drilled holes at the same locations on the outside plate. Then I put in some new grease and slapped it back together. This is the bottom of the aluminum case with holes. Note that the holes had to be in those locations because of how the outer steel plate mounts to the bottom of the aluminum case, there is no room on the long edges.

Image

This is the steel outside plate with matching holes.

Image

7/31 Back to the Red Pitaya noise...

Hmmmm, nice rainy morning so played some more with the Red Pitaya. I think I found the problem. On the input lines there are two sets of jumpers… I didn't quickly find a final schematic, apparently they are keeping those private, but I did find a 'development' schematic, excerpt shown here:

Image

So jp1 sets the input to one of two attenuators, and jp2 sets the output. I thought I would try the HV setting, which attenuated my 1uv to below the noise level. But I noticed when I pulled the jp2 jumper off the output noise went well below the display level I had set on the HDSDR. So after playing around injecting my signal to pin 5 of jp2 through capacitors and jumpers and generally making a lot more noise, I just put one of their jumpers from pin 2 to pin 5 which bypasses the two input attenuators and voila, now the two CW Skimmers are generating almost identical spots. In fact some of them from the Red Pitaya are 1db higher SNR, and a few are a bit lower, but that is bound to happen just from the skimmer software timing of when it generates the SNR value, one may sample before or after the other and get a noise spike or not.

So my guess is that somehow the input attenuator is coupling noise from the board ground into the input signal going to the amp/filter block. But it would take someone who knows more about the low level design of the amp to figure it out. I guess I would have expected to see the inputs isolated from the digital board ground and going into a differential amp, but I can't tell that from the schematic I found. It could also just be that the attenuation of the input before the amp dropped the signal closer to the amp noise floor and all I'm doing is boosting it a bit, but whatever the mechanism it seems to have helped a lot.

8/9 Nice calm day so I figured I would replace the first of the 20m M2 Baluns. George came up and I pulled up the lifting rope so he could help with positioning the antenna from the ground. I started with the south one which is about 10' above the 30' guy wires, I had to unhook the boom truss on the driven element end to get room under the 60' guy wires, otherwise it went fairly easily. As soon as I got the balun off I could tell it was bad, they should not sound like a rattle! I had the Mini-VNA on it to make sure it was connected properly before I came down, I thought about doing screen shots before and after but with the stubs and 75 ohm hardline the results are not intuitively obvious, but it did look better afterwards... but any change just shows that something was different, not exactly why... It could be that it was different just because it added a few feet to the coax before the switchbox and transition to 75 ohm hardline.

This is what the stack of 3 binocular ferrites look like:

Image

This is one from 40m done last year, this one shows more heating in the foam: 40m Balun Dissection

8/11 Replaced the M2 balun on the 60' 20m antenna on the TIC Ring Rotor, at least this one doesn't rattle. Also Tom noticed the rotor indicator was about 90 degrees off so we reset the pot and recalibrated it.

This is the new shed that was put in last week. It is giving me actual room to work in the old shed and taking some of the seasonal stuff and general junk out of the garage. So far the only radio stuff in it are the two big footlocker type things that the big lifting ropes are stored in, they really got in the way in the old shed.

Image

8/15 Replacement indicator pot for 60' 20m TIC Ring is bad, they are going to send a new one this week.

8/18 Tried to replace more baluns on 20m but the wind got stronger instead of dropping off so just did some other little things. Re-taped lots of feedlines on the tower legs, replaced the rope for the 120' standoff spacer, tightened up the 120' boom-mast plate a bit, straightened up the boom support struts on the 60' ring rotor, and inspected a bunch of stuff. Maybe tomorrow will be better.

8/19 George and Tom W1TO came over again and we did the 90' 20m balun.

8/20 George and I did the 120' 20m balun with K1MK reading the meters in the shack.

8/21 Friday night the 40m station was losing power output quickly when transmitting RTTY. In the period of a CQ it would drop from 1500w to less than 1000w. As a quick fix I switched them from the left radio to the right radio. Saturday I tested it again and found it was the BandPasser-2 40m section that was detuning as it heated up. I could see the SWR on the radio increase as the power out of the filter went down. Today I swapped BandPassers from the right to the left radio so that one was back operating properly. I check the filter tuning and it looked fine when cold so I removed the cover, hooked it up, and ran 100w through it until I found a hot capacitor. I replaced the capacitor and now it is stable again.

8/24 Thanks to George and W1TO I got the last of the 20m baluns replaced today. The noise from 20m to 6m and other bands seems to be much better now, almost no crackles on 6m where before it was pretty annoying. I also found an intermittent in the 30m hardline connection that I must have wiggled when I did the 120' balun. I took it apart and cleaned it up then redesigned the matching network inside to get the SWR better in the 30m band. The wires must be about 5" too long as the natural resonance is just below 10MHz.

I have now replaced two 40m and five 20m baluns from MSquared with coax chokes. I thought it might be interesting to see just how much damaged they were. I have only dissected 2 of them, the 40m 70' one and the 20m south one. Both of them rattled badly and displayed internal damage when opened up.

This was the 40m 180' one, it was not in the air more than a year probably and was the first one replaced when I started to suspect them. It was replaced probably before failure because the antenna was on the ground for other repairs.

Image

This was the 40m 70' one. It was obviously bad as seen here: 40m Balun Dissection

Image

This was the 20m South one and rattled when taken down, picture of the cracked ferrites is above.

Image

These two I forgot to label so they are either the 20m 60' or 90' ones.

Image

Image

This is the 20m 120' one.

Image

And the last one replaced the 20m 150' one.

Image

8/29 The old Trendnet webcams have been having problems recently. One of them seems fine, but the outside one and two of the others keep hanging up or resetting themselves randomly. It doesn't seem to be related to RF in the shack as it will happen when nothing is on the air. I bought a new one for outside, its probably good that I did it now. When I took the dome off and touched the camera to disconnect it the camera fell off the base. The same thing has happened with 2 of the inside ones, when they hang upside down the plastic that the base screws onto gives up and they fall off it. For the inside ones I just turned the mounting bracket over and set them on top of it, but I can't do that outside. So now the new one is outside and I'm continuing to test the others, I may just get rid of any that are having problems still as I don't think they are being used much. I turned the external controls and streaming off a couple weeks ago to see if it was something from outside causing it and there hasn't been any big outcry from users. Now only the periodic still pics update on the web site and the Weather Underground site is grabbing the two cameras that look outside.

8/30 Friday afternoon I noticed that power was off to the SO2R/Remote station. The Digital Logger power controller was offline. It appears the lower voltage supply to the control board was dead. Ordered new version from EBay, it arrived Monday, put in in today (Tuesday), everything fine. New one has some extra featured like a local display of status and control buttons for local control. It also doesn't have a fan so its quieter... and it fit on the same mounting screws as the old one, uses the same default IP address so the access point I was using for the other one worked just fine. I took some extra time to sort out the wires hanging from under the table, but that will still need some more work when the weather is worse.

Replaced 4 old 2x4 posts on the 10m tower and 40m 4-Square and RX Array feedlines with new 4x4 posts and straightened up some others.

8/31 Re-taped a bunch of feedlines and control cables this morning. Just one of those things that has to be done periodically.

9/9 I think I found one source of lockups on the GreenHeron Wireless Base used for controlling antenna switches for the SO2R station. They use XBee units in the 2.4GHz band that overlap WIFI frequencies. What I found was that the Wireless Base was set to channel 'C' which corresponds to channel 12 on the Digi Frequency Table. That channel overlaps WIFI channel 1 which my GuestNet access point was set to. When I started a VNC session from my laptop on GuestNet to one of the SkimmerServer machines and ran HDSDR with a frequently updated waterfall it saturated WIFI channel 1 and caused the Wireless Base to loose its connection to the remote boxes and eventually lock up. I changed GuestNet to channel 6 and disabled the WEBCAMS access point that used to be there and now there is almost no interference.

This weekend W1A is using the remote from the ARRL New England Division Convention, after that is over I'll try the PTT lockout again and see if that works ok with the channel conflict resolved.

9/11 Took a stroll around the Beverages. Some trees down but no real damage. All 3 could use a visit from the chain saw and mower.

Haven't done this since 2011 before the lightning arresters were installed. I know the lightning arresters distort the TDR traces on the Beverages (and other antennas and feedlines) but I hadn't quantified it. So today after checking that the Beverages were all up and connected ok I decided to create new TDR traces through the arresters. However I wasn't sure that what I was seeing was really correct, so the first step was to bypass the 6 arresters... lets see, 6 arresters means 12 connectors to take off, then to put onto barrel connectors, then take pictures, then take the 12 connectors off the barrels and put back onto the arresters, all while standing on a short ladder and working just above head height in the lightning arrester box... if you weren't counting that's 12 off, 12 on, 12 off, 12 on, or 48 UHF connectors screwed one way or the other. And for some reason these seem to be a good tight fit and have very long threads.

This is what the TDR traces looked like in 2011: 2011 Good Beverage TDR traces

I am using basically the same settings now. .5usec/div and 1v/div

Lets compare one of them in detail. Take the first one, pulse into the north and return from the south. This is what the 2011 trace looked like.

Image

And this is the new one through the lightning arrester. Note that because the lightning arrester has a series capacitor and a drain coil it forms a high pass filter so all the normally flat (DC) sections of the trace become exponential charging or discharge shapes. While this distorts the longer section of the trace you can still see discontinuities as they should appear.

Pulse into North, return from South:

Image

Pulse into North East, return from South West:

Image

Pulse into South East, return from North West:

Image

Pulse into South, return from North:

Image

Pulse into South West, return from North East:

Image

Pulse into North West, return from South East:

Image

9/12 Feels like fall, time to work on Beverages. Ran the DR Mower up and down the NW and SW paths. That was a good morning workout.

9/15 The continuing saga of the original DR Mower... I have been keeping this thing going since the spring of 1998. Since DR redesigned it some of the parts are no longer available for this old model, but so far that hasn't stopped it. The engine, gas tank, and muffler are probably the only original parts left at this point. Even the frame and handle bars have been replaced, the frame because I bent the original one, and the handle bars as a safety upgrade I think.

In any case, yesterday I finished mowing the South Beverage path and decided to take it through the badly overgrown connecting path between the South and North West paths. I had been noticing the chain slipping so tightened it up and set out on the long connecting path. It was going ok, but the chain started slipping again part way through, so I slowed down and tried to just keep going forward as it seemed worse when switching from forward to reverse. I lost the path at one point so I stopped cutting and just went up hill till I hit the North West Beverage path then went back down till I got to where I lost it, turned back around and did a second pass on really bad section where I lost it. It was still working ok so I was going to do the whole path the other way so went back down the little hill and then the chain came off the gear on the differential. Oh yes, the battery is dead so I can't shut it off to work on it and all I have is a Leatherman on my belt. Fortunately I have been here before when the chain stretched and popped off, so I reach under and get the chain part way on then go around front and pull it forward a bit and the chain wraps around the gear. I tighten the adjuster as much as I can with the Leatherman and head toward home, now uphill all the way. I get about 5' and the chain pops off again with a weird noise. So I repeat the process, slow down some more and get a few more feet up the hill. Lather, Rinse, Repeat, probably another 10-12 times, now I'm only getting a foot or two at a time, but I made it up to the main Beverage trail. At this point I'm hot and tired so I shut it off and walk back to the house. I call George to come help retrieve it. After I cool off and refuel I get the longest of the tag line ropes which are pretty old polypropylene but about 3/4" diameter and still in good shape, I think this one is 500-600' long. And I drive the JD Mower over to where the NW Beverage goes into the woods. I start walking and uncoiling the rope as I go and just reach the DR Mower. I tie it off to the mower deck spindle as that is the best spot on the front of the mower and walk back to the JD Mower and just have enough to tie it onto the hitch. Then I go have lunch while waiting for George. When George shows up after lunch, he was out mowing with his newer mower, I have him drive the JD Mower while I guide the DR Mower over roots, rocks, stumps, etc all the way out to the old shed. We flip it on it's side and find that 2 of the bolts on the differential holding it to the main drive gear are broke, one is completely missing, the other is about half there. The partial one was probably catching on the frame at times which made the weird noises and would pop the chain off. The other two bolts were also loose. And part of the differential housing is missing. I tighten the remaining bolts and go through my spare parts stock and find 2 bolts exactly the right size and put them in. We flip it back up and jump start it and it is running nice and smooth again with no more chain slipping and odd noises.

This is the missing part of the differential housing with the new bolts in place. Note the torn metal on the right side is the destroyed axel lock mechanism that hasn't worked for years.

Image

This is one of my older modifications to the deck, the piece that used to go down and just act as a skid under the outlet chute kept catching on roots and stuff when backing up, eventually it broke off. I put an angle plate under the front corner, welded a piece of ground rod to it and bent it down, back, and then up and used the big cable clamp to connect it to the back of the chute so there is nothing to catch roots when backing up. That has been rewelded a few times but I think I have enough extra welds on it now as it hasn't broken in a while. You can also see that the tree pusher has been rewelded a few times, as have some of the braces on the spindle. One you can't see is that the pipe that lets the deck swivel on the frame has been rewelded to the spindle several times also.

Image

This shows one of the bigger structural changes I made. The bent tubes gave up a few years ago, the started bending and breaking at the frame joint. I straightened them back out to get the right angle and then welded on those gray plates. Note the fancy bungee cord and rope replacement for the spring that got lost at some point on the belt tensioner.

Image

9/21 Put back up the refurbished and drain holed TIC Ring motor on the 60' 20m rotor. Also say hi to the new 160m RTTY operator, Hank. Picked him up from Paws New England yesterday. He is about a 3 year old Treeing Walker Coonhound.

Image

10/4 A few days ago we noticed some noise from 7150khz on 50125khz, and even more from 3600khz to 28800khz which was about s9 while on low power on the low 80m Inv-V. With W1BX at the controls I went up to the feedpoint of the low 80m Inv-V and untaped the feedpoint connections. Both split bolts were loose, tightened them up and the noise dropped to maybe s1 on the harmonic. It can be heard with the ssb filters but is much lower than it was. I also tightened up the low 40m Inv-V but neither one affected the 40m to 6m noise which is pretty weak and intermittent any way.

10/5 A much nicer day than yesterday to climb, so went up the 15m tower. Re-taped lots of the hardline and other cables and pointed the 90' European Yagi back where it belonged, it was about 30 degrees too far South.

10/13 Went looking for a possible noise on the 20m tower. Replaced the RCS-8V, though the old one didn't look too bad. Noise may be a bit better but still there. I thought it was just from the south antenna, but its just because that one was pointed best at the 10m tower. It may also be related to rain, and things have been dry again so will have to see if it gets worse the next time it rains.

10/17 Took Hank out for his first Beverage check, just up the NNW one. Added more supports, trimmed some branches, and found a pine tree down across it about 2/3's of the way out so will have to drag a chainsaw up there soon.

10/18 Took chainsaw up NNW Beverage and cleared the downed tree and a bunch of other junk. Added tension to it about 2/3's of the way out and at the end. Cut off about 10' of extra wire now that it is not sagging as much. Found resistor is damaged, probably from lightning so will have to replace that.

10/19 Another trip to the end of the NNW Beverage. Replaced the termination resistor. The resistor was open when I tested it.

10/23 First snow of the year, followed up by a day of 50-60mph winds. Early morning already shows one tree down across the SW/NE Beverage, will probably find more tomorrow.

Found tree broken off at the ground that is holding one of the radials from the North 160m Inv-L, temporarily fixed that. Found a tree down on one of the South 160m Inv-L radials, took it off and then found the tree that it was tied off to was broken also so moved that to a different tree.

10/24 Windstorm pretty well over, expecting 30-40mph later today but fairly steady at 15-20mph now so it was safe to go back in the woods to check some of the Beverage damage. The N/S one is ok, but there are 2 big trees down on the NE/SW one:

This tree broke off:

Image

And landed square on the Beverage:

Image

And farther back by the stream this smaller one also came down on the Beverage:

Image

Time to fire up the chainsaw and take another walk with a pocket full of spare insulators. Just a note, for all my Beverages I use 10 ga copperweld that has a heavy PVC(or is it HDPE? too early, not enough caffeine yet.) insulation. The Beverages will still work even with the trees down, just maybe not quite as well. The electric fence insulators are only used to hold the wires to the fiberglass electric fence rods that are stuck through those 2x4's. In the event of trees down like this the fiberglass bends and the insulators slide off thus protecting the 2x4 and fiberglass. Most of the time you can find the insulators nearby, but sometimes they fly off never to be found, and occasionally they actually break and must be replaced.

All fixed, an hour or so with a chainsaw cleaned them right up. didn't even lose any insulators. This time of year they are easy to find as the leafs have come off the trees and brush and turned brown so the bright yellow insulators stand out nicely.

11/2 Yesterday and today I have been out cleaning up dead or leaning trees. This is about the best time of year to do it. It is nice and cool, all the leafs are down so you can see the whole tree, and the first wind storm has broken off or dropped the really bad ones. Yesterday I cut about a dozen of the dead scotch pines around the 40m RX array. Today I took down a leaning birch that was hanging over the NNW Beverage and ski trail. I also took down a couple dead pines near the NW anchor point of the 150' tower.

11/5 More removing of dead trees today. There are more to go also, lots of the scotch pines in the 40m RX array area are dead, as are most of them around the 80m 4-square.

11/7 More dead trees cut down today around the 40m 4-square and RX array.

11/8 Even more dead trees cut down today. First near the 80m 4-square, then on the stream side of the 40m 4-square. Found one of the guy ropes had broken on one of the 40m 4-square verticals, tied what was left to a different tree... the verticals are still self supporting even when the ground isn't frozen, that's a good thing to know.

11/14 Friday before the WAE RTTY contest the 70' 40m rotor wouldn't turn. Went up today to see why. That is a TIC ring with 2 drive motors. First, one of the motor's had sheared the roll pin that holds the drive gear to the shaft so it wasn't doing anything. The other motor was turning but the ring was so low it was under the gear. The plastic rollers had worn so much that the ring was sitting on the top of the motor case which was probably the final thing that stopped it from windmilling.

Also cleaned up a bunch of weeds and a branch that were on the 40m 4-square radials that were causing noise when transmitting on it.

11/16 Replaced the right radio on the SO2R station with the spare. It had developed intermittent T/R relay and also A/B antenna relay. In checking out the SO2R system afterwards I found some horrible clipping of audio when using Dual Rx on either radio or one radio in each ear through the SO2R box. Swapped the SO2R box twice and finally tracked it down to the cable going to the Pyle noise filter for the remote operation. That cable had an adapter that combined the two channels into one so remote ops could use the Dual function to listen to split operations. I removed the Y cable that had split the audio from the SO2R box for the remote use and instead put it straight to the left radio headphone jack.

11/18 Started removing undergrowth and small trees just west of the 150' tower. There are several trees in that area that are getting too big and need to be removed.

11/26 Finally, got second Acom amp for the SO2R position.

Image

11/27 Took quick walk around the Beverages, only one small branch down on the trail between them. Shotgun season for deer starts tomorrow so Sundays are the safest time to do that this time of year.

11/28 Took down some more trees near the 150' tower and up the old North Beverage path.

11/29 Took the streaming webcam pages off the Wiki as no one has asked about them not being available for a while. Also cleaned up some other stuff on the remaining webcam pages.

12/2 Rained and warmed up a couple days back so went back to cutting more of the small trees and dead stuff west of the 150' tower.

12/6 Only and inch or so of snow on the ground today, temperature is above freezing and there is no wind... so took down one of the 160m Inverted-L radials and cut some trees out of the way of other trees that really need to be cut down along the 40m receive array near the 10m tower.

12/11 Another fairly calm day without much snow left on the ground so cut a couple more trees west of the 150' tower.



12/18 This one is fun… There was a RTTY contest that overlapped part of the 160m Stew Perry contest... 160m had no problem with the RTTY, but after that was over I reset the SO2R station to be used remotely and N1RR came on to play. When he started using CW on 40m it created broadband noise S9 across 160m. What I found this morning was… With 160m RX on the inverted V and the 40m TX on the top beam (10’ away) there was no noise. This is good, it shows that the noise is not some kind of arcing outside on 40m. The noise was only there on the 160m RX antennas. But it’s even more complex, because the noise depended on which Beverage was selected on BOTH 40m and 160m… when both were on South it was the worst, but some other combinations also caused it. And if the preamp on 160m was off it was much lower level than with it on, more than I would attribute to the preamp gain so maybe that preamp was overloaded by the in-band noise. Then I found that it went away completely with the 40m/SO2R station preamps turned off. Note that when setup for SO2R or remote use I remove the 40m RX filter so the 40m/SO2R station can use the RX antennas on 30/40/80/160m. Put the 40m bandpass filter in place and the noise went away (even when transmitting on 40m). It was really interesting that with the SDR-IQ on the 160m RX line the noise was not flat across the band, it had many peaks that were very unstable. My conclusion is that the SO2R station preamps must have been overloaded and were oscillating making trash all over 160m (and probably elsewhere), then they were feeding back into the RX switching matrix (because the bandpass filter was missing) and out to the DXE Bidirectional Beverage boxes since sometimes I could get the noise on both the same and the opposite direction Beverage. This could potentially be a source of noise during M/M if I forget to put the 40m RX filter back in line… maybe some of the otherwise unexplained inter-band noise we have had. I could also make it happen when TX on 80m on the SO2R station, but not as bad and only on a couple of Beverage combinations.

12/24 Ok, first step to fix the noise problem with Beverage filtering and switching was done yesterday. I figured out an approach to allow the SO2R radios to each select the correct filtered RX antenna line for their bands. Second step, figure out which pieces I have and what else I need... Found 2 GreenHeron remote boards, 2 RCS-8V's, but had to order some RG-6 and control cable to wire them into the band decoders. I think I have enough connectors and T's to make up the connections. The only thing that will be different is that for SO1R to use the RX antennas on 30m I still have to figure out a way to bypass the filter. Third step, wired the 2 GreenHeron remote boards into the 80m and 160m RX Antenna control boxes. I'm missing a screw terminal header for the 160m one, they said they would send me one after the holidays. Cable should be in Wednesday next week so I can start hooking it up.

12/25 This is the 'before' rewiring of the receive antennas for SO2R filtering. Green wires are control, black are coax.

Image

And this is the planned 'after' wiring. This adds remote control to the 80m and 160m RX antenna controllers, the SO2R station will use those to select the RX antennas for those bands instead of using the 40m controller for all bands. This also adds two RCS-8V switch boxes controlled from the left and right radio band decoders. Three inputs are used, one from after the preamps and filters from 40m, 80m, and 160m. The RCS-8V boxes will be controlled from the existing left and right radio band decoders, hopefully they can handle another relay load.

The leads from the preamps to the RCS-8V boxes must be kept short so they do not act as stubs and affect the 80m and 160m stations during M/M operation... This could be done by going from the preamp to one RCS-8V, then to the other RCS-8V, then on to the radio, that adds an extra T connector but might be better then splitting before the first RCS-8V, or not. On third thought, the longer cable from the 80m and 160m preamps to the individual station will have to be disconnected to prevent that from acting as a stub and affecting the SO2R reception. A manual switch or disconnect might be needed from those two preamps to select either M/M or SO2R as is done on each station for the transmit antennas now... still thinking.

Also unresolved is how to use RX antennas on 30m.

This should also allow removing the RX Protector from before the 40m preamps as the input to them will always be filtered... or should I add protection on the other two? No, probably not needed as they have survived for years now without protectors. Maybe the 40m one was never really needed either.

The SO2R GreenHeron setup shown at the start of this year will get more complex also as there will be separate control panes for 40m, 80m, and 160m RX antennas instead of using the single one for both radios. These will change with the band selected for each radio and will be able to select different directions for each radio which might be handy.

Image

12/28 Ok, made a few changes to the 'after' design above... will document later after I make sure it all works. Part of it I did today, first I changed the preamps so 40m/SO2R/Remote only has a single preamp. Cascading two of them was really just raising the noise level and wasn't useful. The extra preamp will now be used for 30m. To split 40m from 30m at the RX antenna switches I converted an old ICE 40m Bandpass filter to 30m and upgraded it to be a 2 section filter more like the W3LPL designs. This leaves just a restriction that 30m and 40m have to listen to the same RX antenna. This won't be a problem for remote use, or for SO1R, or for M/M, it would only come up if doing SO2R with 40m on the right radio and 30m on the left radio, which is not a normal contest thing of course so I'm not going to try to get around that... it could be done but would require another RCS-10 and 8 more cables and another T on each of the 8 junctions and another RX control box, nope, just not going to do it.

12/30 Used clip leads to connect 160m GreenHeron remote control in RX antenna controller so I could integrate the rest of the stuff and test it out. Everything seems to work for both SO1R and SO2R. Still need to close up the new RCS-8V's and mount them up on the wall. Found a problem with the WARC Bandpasser that was doing weird things like getting stuck on 30m when other bands were selected causing no output power and high SWR, bypassed it with a barrel connector for now.

This is what the final switching system looks like. From the last iteration I have removed the cascaded preamps and made one of them into a 30m dedicated preamp following a 30m bandpass filter. This meant bringing over the 30m band selected wire from the band decoder and more changes to the GreenHeron software configuration.

Image

Ok, now the pictures... This is after getting started. At the top are 3 RCS-10 switches, the left one is for 160m, the middle for 80m, and the right one for 40m. Under each box are the preamps, on top of each box are the bandpass filters. In this picture the two filters under the 40m box have been separated and the rx protector removed. Also on top of the 40m box there is the new 30m filter. So at this point the output of the RCS-10 for 40m goes to the 30m filter then T's to the 40m filter, each filter output goes to one of the preamps under the box. On the right between the cable duct are the two new manual switches, one for 80m and one for 160m. The output of the preamp for each band goes to the common port, then one output goes to the single band station and the other output goes to the SO2R station switching.

Image

These are the RCS-8V switch boxes for selecting which band goes to the SO2R Left and Right stations. The left box is on the left and has the outputs from the 4 preamps (via the manual switches for 80m and 160m). Then the 40m, 80m, and 160m inputs are T'ed to go to the right box.

Image

These are all the boxes mounted on the wall, but before doing some wire management work.

Image

This is what is being replaced, what used to be the 40m RX antenna cable came in and was split to the two radios.

Image

And this is the rats nest that makes it all possible... this contains outputs from 2 band decoders that then drive bandpass filters, TX antenna switches, and now the RX band selection switches... all very carefully labeled and color codes documented on fancy file cards.

Image

Top