The past two mornings, the temperature at the transmitter site has been 44F at 8am.  By 9:30am it had already warmed up to 62F.

I forgot to mention something in out list of unanticipated events yesterday.  About an hour after the morning broadcasts ended, the automation system went ‘berserk’ with lights flashing and bells ringing.  Fortunately we were off the air and broadcasts were not affected.  The problem was traced to low voltage from batteries.  The special circuit inserted in series with the battery charger to prevent overcharging had failed resulting in no charge.  Graham and I decided to use a resistor to replace the special circuit until Hans, the designer of the circuit, can repair it.

About an hour before transmissions ended this morning, the monitor on TX2 went blank.  We installed a replacement monitor, but it too did not work.  I tried the replacement monitor in the other transmitters and it worked in both.  This pointed to a problem with the TX2’s video card.  Since the transmitter still functions without the monitor, we waited until transmissions ended.  Before swapping monitors, Klaus pressed the reset button on the TX2’s computer and the old monitor started working again.  Problem solved!

Next it was decided to test the 10,000 uf capacitor that had given problems on TX2 on Saturday and had ‘returned to life’ on Monday.  We were concerned that it might be electrically leaky, or losing capacitance.  When it was removed from the transmitter, it was charged to 35 volts.  When measured with a digital voltmeter, the voltage was dropping at a rate of about 0.001 volts per second.  This is about right considering the loading effect of the meter.  We then calculated that a 1,000 ohm resistive load would cause a drop from 30 to 10 volts in about 10 seconds.  This proved to be true.  We then checked all the solder connections that could have caused the problems, but could find no bad ones.  The mystery remains, so the spare capacitor with clip leads, ready for a quick connection, will remain for a while, just in case the problem should return.

This afternoon, I calibrated the RF Drive Controllers on both TX1 and TX2.  This involved tuning each transmitter to all the frequencies and antennas that it uses, finding out which one needs the most RF Drive, then calibrating the RF Controller on that one.  Doing this on both transmitters took most of the afternoon.

When I got ‘home’ this evening the temperature was 70F, both inside and outside the house.  We had a supper of leftovers and hope to get to bed earlier tonight.

Today was a day of the unanticipated !

When the  morning broadcasts ended, we decided to remove the capacitor that we put in TX3 on Saturday to analyze what had failed and to be able to explain the reason for the symptoms that we had seen.  Lo and behold, everything was working properly with none of the symptom we had observed for several previous days!  We looked all over for possible bad connections, but found none.  This left us uneasy about the possibility of the return of the problem.  Since the operator on duty is not allowed to open any doors or panels without another person on site, we left a capacitor inside the cabinet with two clip leads at ground potential extending outside.  If the problem reoccurs, the operator has been instructed to connect the two leads together, thus connecting the spare capacitor into the circuit.

Next, we had noticed that one of the door interlock lights on TX1 was on dim.  When the problem door was checked, it was discovered that the interlock switch was not working.  Opening the interlock switch enclosure, it was noticed that one of the connections was corroded as if it had been wet.  The switch and some of the corroded wiring was replaced.  Everything tested OK.

Just as the repair was being completed, Graham felt a drop of water hit his head.  Our next detective work was to find out where the drop was coming from.  Looking from the top provided no clue.  Next, several air filters were removed to allow gain access from the side.  A puddle of water was seen on the top plate of the transmitter and Graham was able, with the light from a ‘torch’ [flashlight], to see that the threaded pipe into the bottom of the heat exchanger [large radiator] was leaking.  Because of air ducting that has been installed on site for cooling, it was necessary to remove all the back doors of the transmitter and the top trim to gain access to that area for repair.  The threaded pipe was not very tight, so we removed it, renewed the Teflon tape on the treads and retightened the pipe, then put everything back together.

In the process of checking out the door interlock light problem, it was noticed that the ‘ground switch down’ light was not working on TX2.  The grounding disc was contacting the bleed resistor but not dropping far enough to contact the ground contact.  The spring on the grounding disc was too hefty, so we located a replacement, but this required some modification.  With a little work, using the lathe and modifying a washer, it was possible to make it work correctly.

While this was being done, I noticed that the optical arc detector light was not on.  Since it is sensitive to light, it should come on when the door is open.  After using several drawings to assist in tracing the wiring , it was found that this detector should have been connected in parallel with a second detector (from the same area) on a terminal block behind the control panel.   When the connection was made, everything worked as it should.

After this discovery, the other two transmitters were checked out.  TX1 was OK, but TX3 had the same problem.  This was also corrected.  I could not determine whether this was left out when the transmitter was assembled in Elkhart, or was removed after it was installed here.

It was also discovered that the ‘ground switch down’ indicator light was not functioning on TX3, while the switch was actually down.  It looks like the limit switch needs to be adjusted — a job that will have to wait until tomorrow.

TX2 came up for the evening broadcast with no problem, but I did notice that the calibration of the RF drive controller needs to be checked.  This will, also, have to wait as multiple bands will have to be part of that calibration.  TX1 should also be checked out.  TX3 has already been done.

Just before leaving the site, TX2 was manually switched on about 10 minutes early for its second transmission, just to make sure the problem from Saturday had not reappeared.  All was OK.

After I left for work this morning, Virginia learned that we would be having a family over for dinner this evening.   Greg and Ingrid, and their two young children are here for a week.  Greg had been in  Bratislava and in Swaziland at the same time Allan and Joyce were in these places, but Greg was single then.  He is here from TWR’s Africa Regional Office in South Africa to work on the security system in the Manzini office.  Ingrid and children came along, since they are on vacation from school.

This Sunday morning a delegation of three members of a ‘daughter’ church expressed thier thanks to the EBC congregation for the assistance provided to their small church.  The first lady spoke in English and the second lady spoke in SiSwati.  The third person was a young man, who helps with youth.  The started by saying in English, “You’ve heard us speak in English and in SiSwati.  Now I’m going to speak in Spanish.  Do you have an interpreter?”  This got a chuckle from everyone.

Just before his sermon, Pastor Philips sang a duet with Irene.  He said that they had sung it at a funeral on Saturday.  It was a song I have not heard before — “Jesus, You’re the center of my joy.

CHORUS    Jesus, You’re the center of my joy      All that’s good and perfect comes from You        You’re the heart of my contentment        Hope for all I do Jesus,        You’re the center of my joy
VERSE 1     When I’ve lost my direction        You’re the compass for my way        You’re the fire and light        When nights are long and cold        In sadness, You’re my laughter        That shatters all my fears        When I’m all alone,        Your hand is there to hold
CHORUS
VERSE 2     You are why I find pleasure       In the simple things in life       You’re the music in the meadows and the streams       The voices of the children, my family, and home       You’re the source and finish of my highest dreams
[ From: http://www.metrolyrics.com/jesus-youre-the-center-of-my-joy-lyrics-richard-smallwood.html ]

Pastors’s message this morning was from Ephesiana 1:15-23 with an emphasis on Christian knowledge.

After the service, one of the ladies from the ‘daughter’ church came to us and assured us that there would be a multitude of believers in heaven because of the technical work we are doing to broadcast the Gospel.   We do not know her, nor why she singled us out to tell us this, but it was such an encouragement!

Ruth invited us over for dinner after church again.  Like previous times, we provided her transportation home.  No one else was there this time.  Her meal was simpler than the previous ones but very good.  She prepared meat loaf patties and fried potato wedges with a salad.  We then had cake and tea.

We got home about 2pm and spent the afternoon relaxing and doing some blog entries.

We finished the evening with a snack and several rounds of Rummikub.  I then made a Skype call to my mother before turning in for the evening.

I knew that this might not to be the restful Saturday we had planned when my cell phone woke us just before 7:30 this morning.  Klaus was already at the transmitter site trying to sort out multiple problems that were keeping TX2 from  going on the air.  He wanted me to come out to help.  About 10 minutes later and just before I left, he called to say the transmitter was on the air and that we should plan on going out at 2pm this afternoon for further investigation as to what was happening.

We harvested a few more nartjies (mandarins) this morning, but the rest still have some green on them.  These will wait until next week.  Virginia picked a few lemons that are turning yellow.  I watered three small trees, which have been planted fairly recently.  They are less than three feet tall and at least two are citrus trees.  The bunch of bananas that we cut down on Wednesday has ripened only a little bit.  As we have finished up the previous bunch, we hope these are ready in a couple more days.

Klaus and I left for the transmitter site about 2:30.  TX2 came up with no problem for testing, but the metering for RF Exciter Q3&Q4 currents was still showing the drifting up and down like last night.  We investigated further and once again found these currents were actually steady.  All the other metering points on the same card were steady and we could find no other similar anomalies.  I examined the log of faults that occurred this morning and noted that door interlock faults seemed to be the fault most frequently shutting the transmitter down.  Klaus had temporarily bypassed the door interlock this morning to get the transmitter back on the air.

TX2 came up fine at 4pm for its first short Urdu transmission of the evening and was performing well, yet the ‘drifting’ remained.  We continued to ‘scratch our heads’ as to what was happening.  The transmitter was then off for about 30 minutes before the next longer transmission.  So we looked at the computer screen with all the metering information to see if other readings showed a similar drifting.  Only one was found, but it was drifting over a much smaller range and in the opposite direction of the drifting of Q3&Q4 currents.  This one was the RF Exciter +24 volt supply voltage.

Fortunately we decided to stay for TX2 to start the second transmission.  Every time the high voltage vacuum contactor switched on, the door interlock light flashed and one or more of the optical arc detector lights came on.  These were the same symptoms that the transmitter had this morning.  We noted that +24 volts supplies the interlock string.  Klaus quickly bypassed the door interlock like he did this morning, but the problem persisted.  The arc detector supplies are +15 volt.

Klaus decided to measure the voltage on the RF exciter supply when the contactor switched on and noticed a significant momentary drop in voltage.  I suggested paralleling an additional capacitor across the one in the supply, so Klaus quickly found a 230 uf capacitor that he pressed against the terminals of the one that was already there.  The transmitter switched on!  Now we had another problem — could Klaus hold the capacitor like that for at least another 20 minutes!  I managed to find some clip leads and we were able to replace Klaus by some strategic connecting of the clip leads.

We discovered that the capacitor that was bad was 10,000 uf and the one paralleling it was only 230 uf, so we searched for a bigger one in the stockroom.  The largest one that could be found was 5,000 uf, but we figured if 230 uf could help get the transmitter on the air that the larger one would work until a replacement could be ordered.  Using clip leads, we connected the 5,ooo uf capacitor which promptly tripped the transmitter off, but we turned it back on immediately.  We should have put a resistor in series with the capacitor to charge it up before connecting it.  Not doing so, the voltage dropped momentarily when the capacitor initially charged.  Maybe next time we’ll remember!

We finally left the site at about 5:30pm with confidence that the problems had been solved, but I am still ‘scratching my head’ trying to figure out why the Q3&Q4 metering currents were drifting, while their actual currents were steady.  Also, why were the optical arc detectors also triggering on, as they were on a 15 volt supply?  These mysteries will have to wait until Monday, when the schematics can be reviewed some more.

This evening we were able to watch Samuel and Henri playing on the beach via Skype and Angelique’s smart phone.  They were at South Haven, Michigan for the day.  Oh, the wonders of today’s technology!  Later we got to talk to, and see, Allan and their children.  We learned that Audrey’s Upward Soccer team was undefeated this year.  Allan, Joyce and Mark were her team’s coaches.

When we arrived at the transmitter site, the temperature was 44F.  No problem keeping the transmitters cool this morning!

I led devotions again and Sibonginkosi translated again.  My next week to lead the devotions will be the last full week that we are here.

After devotions, I checked my email and found that our reservation request at Kruger had been declined, so I quickly checked the on-line reservations and discovered another opening (apparently there had been a cancellation).  Amazingly, this time the reservation was accepted, so now we have a bungalow reserved for July 6.

As soon as the morning broadcasts were over, an attempt was  made to align the Driver Grid inductor of TX3 so that it could tune the 16m (17MHz) band and still retain the capability to tune the important 90m (3MHz) band.  By adjusting this inductor one turn toward the higher frequencies, it was now possible to tune not only the 16m band, but also the 13m (21 MHz) band, and still tune 90m.  We considered trying for the 11m (25MHZ) but decided against it, because we did not want to jeopardize the much more important use of the 90m band.  Also, the transmitter, as configured, can not tune the final stages at 25 MHz anyway.

After this adjustment, it was now necessary to retune the transmitter on all 27 different bands to program the revised tuning numbers into the transmitter’s on-board computer.  This took just over two hours to accomplish.  The transmitter was then quickly tested at high power on all its currently used frequencies to verify there should be no problems tonight and in the morning.  The next step was to download all these new preset numbers into a backup file so they can be retrieved in case of loss of the transmitter memory.

Last night we noticed an anomaly  on TX2 and decided to check it out this afternoon.  The RF exciter Q3&4 (output transistors) current readings on the transmitter monitor kept drifting up and down from 0.06 amp to 0.55 amp.  The trip point is 0.60 amp, so this is disconcerting.  We measured the actual current readings, which were steady with no drifting.  This indicated that there is something wrong in the metering system, but we did not have time to further investigate as the transmitter was about to come up for the afternoon transmissions.  We left the site about 30 minutes late as a result of checking out and reviewing this problem.

This evening there was a carry-in dinner at Klaus’ home as a thank you, and send off, for Gottfried.  He returns to South Africa tomorrow after looking over the various computer networks and internet services utilized here at TWR Swaziland.  We have seen a great improvement and trust that it will continue this way.  Klaus built a fire in their fireplace, which felt nice on this cool evening.

It’s been a busy, tiring week, so I turned off out alarm clock so we can sleep in tomorrow.