Here are some views of the working prototypes. They were hand stitched on a proto board using point-to-point wiring. I used enameled 22AWG magnet wire. The connections are made by "soldering through" the enamel.
The output filter coils were made with the type of same wire 18 AWG forming a low-pass filter. It is not a very good filter but does knock down the some of the out of band harmonics somewhat, especially the big harmonic spike at 54MHz. Although they probably would not pass for a product under FCC Part 15 - it should be fine for testing.
The temperature sensor is a small 10K thermistor connected to the two small white wires on the right-hand side of the board controls the temperature-to-frequency oscillator. This is calibrated to the ensure the oscillation is in the audio frequency range especially at the coldest temperatures expected. The altitude sensor is not yet connected to the circuit, instead the cycle timer rate is set by the small potentiometer on the left-hand side of the board. It will be calibrated to have the longest transmitter key-on time at the highest altitude so the circuits will hopefully stay warmer to hopefully continue to oscillate and we may get more accurate readings. Short key-on time at ground level may help the battery stay alive longer to aid in recovery.
MKIII Proto
MKIII Point-to-point Wires
Altitude Sensor Proto:
I repurposed a small syringe that holds a small amount of air after the cap is removed and replaced at ground level. In theory the air will expand in relation to the decreasing ambient atmosphere as the altitude increases and move the plunger to change the resistance. With such a small movement we may actually need a partial vacuum to cover the highest altitudes. I doubt that it will stand up to the altitude and temperatures would be encountered at altitudes expected of a balloon flight. Maybe some insulation or even a small light bulb may be needed to keep the plunger from sticking or freezing.
Air pressure moves the slide pot.
The transmitter works in AM at 27.145MHz. The thermistor controls the frequency of the audio range oscillator formed by a couple of transistors, and the altitude controls another multivibrator which keys the transmitter on and off at a rate proportional to the position of the slide pot. From these two modifications of the transmitted signal the temperature and altitude may be inferred.
It is Working!
Receivers capable of receiving in the band can be programmed to receive the signal. Here is my old PSR-500 handheld scanner set up to receive the signal.
Receiver settings for Test
Here is a short (really short) video of the transmitter in operation! The audio is from the scanner receiving the signal set up as shown.
Why twenty-seven? It was the project #27 Telemetering Transmitter from one of my favorite books the 1964 book103 Simple Transistor Projects, by Tom Kneitel. (Pages 32, 33) These project books came out in the 1960s and were a loose collection of circuits, mostly radio/CB circuits with little or no construction details or theory of operation. They generally had a short description, schematic, and parts-list. These books also typically had a list of "acceptable transistor substitutions" that were probably a holdover from the tube days. This project was one of my pre-teen self's bucket-list projects. There were at least 3 other projects in this particular book I always wanted to build.
Although I am not exactly sure how I planned to send this device aloft. But I do remember trying to build it up with a CB radio crystal and a few random transistors and some vague idea of several toy helium balloons stuffed in a laundry bag! Probably it would have never worked even if I could have got this circuit to work without any test equipment or knowledge of how the circuit worked.
Information is not even transmitted digitally from this device either. Unlike today we would slap in an Arduino and an off-the-shelf RF data module, like a RFM69HCW Transceiver breakout, and sending tracking with a cheap GPS module, mix it all together with a bunch of libraries to drive it.
The telemetry here is AM modulated RF carrier (it could easily be FM modulated with a varicap diode in the RF oscillator circuit) The frequency of the modulation is controlled by a thermistor. This type of tone telemetry is like the early satellites and rockets, to transmit the telemetry data to earth stations.
The two-page project description mostly describes the package, FCC requirements, along with some circuit description. Here is the entire description:
The original project #27 goes into some detail about the FCC requirements, ca. 1964. He even recommends the transmitter is inspected by a certified person to comply with the then requirements.
...and it even had a little certificate tag to sign and attach to the device, or "a reasonable facsimile" proving it was worthy of transmitting within in the limits allowed. Also, to protect it from the weather by carefully wrapping it in plastic bag with its 9-inch No.18 buss wire antenna "naturally" protruding. Putting it all in a small, padded box for attachment to the balloon.
"...Put your name and address on the box"
I suppose if you want to find out where it ended up. There are sites with balloon flight prediction that i suppose will get you close. I think in this day and age it may be better to include something like this statement:
"Scientific Instrument, no explosive or harmful substances contained please notify if found..."
This also makes it look official:
The schematic was strangely drawn, probably more thinking about the layout using:
"...Perforated board as the chassis"
For construction we were also advised to keep it as compact as possible and some details about putting it on 11- or 10-meter bands with appropriate licenses and antenna length.
I thought I had some P-N-P Ge Transistors in my parts box, but it turned out they were mostly old Si devices. I found the original transistors on - line but at a cost that did not make it practical to try or even want to risk these antique transistors to this experiment... Even buying a vintage 3rd overtone crystal on e-bay was somewhat costly. The parts on this list are still available. That strange 4-legged Ge transistor was apparently used often in RF circuits, and apparently...
"...Great for your Rangemaster clone and other fuzzbox / distortion projects, these Mullard made OC71 / CV5712 germanium transistors are getting harder and harder to find"
I found the OC170 for $12.95 +$4.80 Shipping. Also, the other transistors here: 2N107 - e-bay were for $25.00 for a pair, (used) and about another $8 - $10 for a used crystal so basically around $50 USD in parts needed for this original design!
Instead, I created two functional versions, called MkI and MkIII. The Mk1 is pretty much a copy of the original version. The Mk3 may actually be in compliance to operate as a telemetering device on the current 27-MHz band. The Mk3 version also features the same simple Temperature-to-Frequency converter it also now includes a way to send altitude information encoded in the length of the tone burst (Pulse modulated) and even a bright LED beacon also attached to the circuit. (This circuit was loosely based on Project #100)
They are without any Germanium or PNP transistors.
The files for the new versions of the are here in my GitHub
Basically, the first version was a simple overtone crystal oscillator modulated with a multivibrator circuit - just like the original. The crystal was taken from a toy "27MHZ" remote controller. ($free) It turns out the frequency is in the ISM telemetry part of that band.
This is the PWB version, 3D models. Nobody even dreamed, we could do something like this to look at a project before it was built back in 1964? It is roughly the size of a 1.5V AAA cell holder.
FCC Rules. These FCC rules are kind of confusing. I am not sure why Tom was so concerned. Most rules govern manufactured devices. I doubt if the men in the black vans will show up if you are trying to act like the devices authorized on these bands and don't try to manufacture and sell these. A quick search of the FCC rules section 15.205 does not even mention these bands. I did find this section it seems to cover the label requirements for devices for sell:
§ 15.305 Equipment authorization requirement.
PCS devices operating under this subpart shall be certified by the Commission under the procedures in subpart J of part 2 of this chapter before marketing. The application for certification must contain sufficient information to demonstrate compliance with the requirements of this subpart.
I did see some current requirements for Part 15 telemetry transmitting devices that more power is apparently now allowed - but - thought I saw a limit now on the length of time a "tone" can be transmitted, the original transmitter and the Mk1 both transmit a tone continuously so they would probably not meet this. (can't seem to find that section)
But of course, none of these transmitters are FCC type accepted for manufacture -- but for experimental purposes on the ISM band it should be OK. Truely experimental devices can legally still be on the 10-meter band - if you have amateur radio license - there still are some identification requirements. In any case, it should follow the part 15 requirement to not interfere with any other user on the bands.
On 11 meters, data links/telemetry can operate between 26957 - 27283 kHz
Part 95 of the FCC rules allow use of the six 26 MHz/27 MHz RC frequencies for high power control, supervision, data link, telemetry networks and "attention getting" devices. Mean (average) power limit is 4 watts on 26.995 MHz, 27.045 MHz, 27.095 MHz, 27.145 MHz, 27.195 MHz and 25 watts on 27.255 MHz The majority of high-power systems use 27.255 MHz - however some do transmit 4 watts on 26.995 MHz
The Mk3 version. It attempts to limit the time that the tone is transmitted (saves on battery life too) and to attempt to clean up the output signal to hopefully better comply with the rules. The original transmitter seems to rely on low power and the antenna length to limit the spurious emissions surely caused by a 3rd overtone crystal oscillator.
The timing circuit was implemented as another two-transistor multivibrator. Since it a lot like: "#100 Flashing Light" project -- I also added a high efficiency LED that produces a bright strobe beacon as well so it can be tracked and possibly help find it. Since the timing was arbitrarily set, I realized that that could be, in fact, controlled and a second channel of information could be transmitted by this pulse length instead of a fixed resistor or potentiometer.
Oh, and yes,Anything worth doing is worth over doing! I added 4 additional transistors. The second 2 transistor multivibrator (drawn in a more recognizable form) one transistor as a keying switch, the 4th transistor is a modulator. Also added are some simple low pass and trap filter to clean up the output.
I then came up with a 1960's technology version of an altimeter to control the tone burst length based on air pressure. I would consist of simple sealed tube of air with a plunger attached to a small slide potentiometer. I will add details of this device later. (The Mk2 Transmitter was a never completed high(er) power version.)
Mk3 Transmitter (top)
The original was intended to be powered only by 1.5V cell. Three volts should be OK to meet power output requirements. I think a modern LiPo would probably survive the extreme cold this balloon-borne radiosonde would encounter. There is now plenty of room for two cells as shown on the bottom view. Of course, weight is a prime consideration for balloon payloads as well so LiPo may be best, with no holder. Here is a 3D peek at the bottom to show scale - that is a dual AAA battery holder.
Mk3 (Bottom)
Calibration. To get any useful data out of this device we need to know what frequency will be modulated onto the carrier at any given frequency. The instructions were simple:
"...Calibrating them with a known temperature on the ground before sending them aloft"
The thermistor used was not the original Philco 33-1343-3 or GLOBAR 22-922 specified. I got new 20 ea. 10K thermistors, new for a few $ online. The nice thing is they are small and react to temperature changes very rapidly.
Selection of the capacitor sets the frequency range. f = 1/1.38*Rth*C Where: C= C1 = C3, Rth = R5. In this design, select the value of C such that the frequency is well in the audio range at lower temperatures to allow the modulating frequency to pass thru audio circuits of the receiver.
OK now that we know all about what it is it is time to build and test one. Part 2 will be the construction details and some testing. Maybe there will even be a Part 3 if there is ever a flight test?!
On this day many, many summers ago in 1969, I sat transfixed to my family's B&W TV. There was no actual video from the space craft landing on the moon it simply was not available. The TV Screen showed an animation and a digital countdown timer was all that their was to see. Even at my young age I understood there was a drama unfolding a quarter of a million miles away...
As they descended to the lunar surface, mysterious alarms were being called out "1202, 1201, 30 seconds" of fuel left...a real nail biter..
102:38:26 Armstrong: (With the slightest touch of urgency) Program Alarm 102:38:28 Duke: It's looking good to us. Over.
102:38:30 Armstrong: (To Houston) It's a 1202.
102:38:32 Aldrin: 1202. (Pause)
...1202 Alarm! what the hell is that? Oh no! this can't be good.... 102:38:53 Duke: Roger. We got you...(With some urgency in his voice)
We're Go on that alarm.
I am pretty sure i said "oh no!" at this... I knew that anything called an "ALARM" can't be good....two ALARMs is really bad. Here is how the rest went...
102:44:04 Aldrin: I got the shadow out there. 102:44:07 Aldrin: 250 (feet altitude), down at 2 1/2, 19 forward. (Pause) 102:44:13 Aldrin: Altitude (and) velocity lights (on). 102:44:16 Aldrin: 3 1/2 down, 220 feet, 13 forward. (Pause) 102:45:17 Aldrin: 40 feet, down 2 1/2. Picking up
some dust. 102:45:21 Aldrin: 30 feet, 2 1/2 down. (Garbled) shadow 102:45:25 Aldrin: 4 forward. 4 forward. Drifting to the right a little.
20 feet, down a half. 102:45:31 Duke: 30 seconds (until the 'Bingo' call).
...Now they are running out of Fuel!
102:45:32 Aldrin: Drifting forward just a little bit;
that's good. (Garbled) (Pause) 102:45:40 Aldrin: Contact Light. 102:45:44 Aldrin: Okay. Engine Stop. 102:45:47 Aldrin: Mode Control, both Auto. Descent
Engine Command Override, Off. Engine Arm, Off. 413 is in. 102:45:57 Duke: We copy you down, Eagle.
They did it?
102:45:58 Armstrong (onboard): Engine arm is off. (Pause) (Now on
voice-activated comm) Houston, Tranquility
Base here. The Eagle has landed.
...Yes!
102:46:06 Duke: (Momentarily tongue-tied) Roger, Twan...(correcting
himself) Tranquility. We copy you on the ground. You got a bunch of guys about
to turn blue. We're breathing again. Thanks a lot.
102:46:16 Aldrin: Thank you.
You can google and read the entire Apollo 11 Landing Transcripts But this is not the real point of this post... it is really about:
The Apollo Guidance Computer
How did they do it? What was this mysterious box producing these cryptic alarm numbers? How in the hell can you navigate and land on the moon with a computer with less "computing power" than a digital watch?
First of all, it is not less computing power. After some research (yes, I'are a Google Scholar) it was a pretty cool bunch of transistors and probably the first digital display in any aerospace machine till that date. And...
I want to build one!
Spock: Captain, I must have some platinum. A small block would be sufficient, five or six pounds. By passing certain circuits through there to be used as a duo-dynetic field core...
Capt. Kirk: [interrupting] Uh, Mr. Spock, I've brought you some assorted vegetables, baloney and a hard roll for myself, and I've spent the other nine tenths of our combined salaries for the last three days on filling this order for you. Mr. Spock, this bag does not contain platinum, silver or gold, nor is it likely to in the near future.
Spock: Captain, you're asking me to work with equipment which is hardly very far ahead of stone knives and bearskins.
For my stone tools, the challenge is to make a 1960s vintage machine from about the same vintage (ca. 1979) parts. Impossible you say? Yes, well, almost. Here is my stone tool:
This is a RCA COSMAC 1802 microprocessor I built, actually the 3rd one but using parts from the first two (yes, in my parents basement) ca. 1977 - 1979. I know this because the notebook it is sitting on has the dates and parts orders, plus notes and other useful information.
This old machine had been sitting in my garage for at least the last 15 years and other storage for the last 4 decades. After a little cleaning, replacement of the little red push button that had corroded, some wires, and one LED...it actually still worked!
Reality Bites...
Being able to re-write the entire AGC software is an undertaking that would take years, and require understanding of all the equations and NASA documentation. Not going to happen as a past time.
The second option is: can this machine run the actual code? Well, maybe. But a quick mental calculation; it will take perhaps 10 - 100 COSMAC instructions to make each AGC instruction. These two machines run at approximately the same speed ( 2MHz) so the result will be 10 - 100x slower and take that much more memory.
OK so there is a third option...how about just the display? It would make a nifty looking clock.
The DSKY!
Beautiful, is it not? The epitome of 1960's computing user interface technology.
...Is it just me - or do these displays look like a couple of smart phones?
I wont put all the details here, they are in my GITHUB. repository. You an go there and study the details if you like. It is a work (?) in progress but it really is just a fun diversion project. There is no real need for us old assembly language programmers anymore - i still do enjoy low-level programming, just something to keep my old brain's spark plugs going a while longer - kind of the geek equivalent to a crossword puzzle. Here is an early version...it really is just a VIP Clock with the digits re arranged to look like V16N65 The mission elapsed time on the DSKY, the current version is a working proto.:
I may add some more here as i plan to make a second version that does not only time, but navigation ( tied to an old Garmin 12 GPS) and some inertial navigation system measurements (Gyroscope). This will be a cool addition to my 1960's vintage Dune Buggy eh?
As for running the actual code? Oh - I think the current ARM boards such as Raspberry Pi could pull it off.
As we like to say on the Rad Ranchero:
~Stay Tuned! ~
...and happy birthday pop - where ever you are! we miss you down here on earth.
Yeah, I know, Steve get a new car. The 2001 Ford Escape that me and my assistant replaced the head lights on. Was that really 3 years ago? We had to pull out the tools and see to fixing the driver side turn signal lamp.
No big deal? Well answer this; how many tools does it take to replace a turn signal lamp on a 2001 Ford escape?
Yeah, the only light on the car that is likely to get you a ticket, the owner's manual states; "...bring to qualified service technician or dealer to replace" So i guess now that I am a qualified technician?
Side nut...exposed
The painful thing about this vehicle is some of the lamps nuts and bolts are behind those plastic fenders held on with the push clips. You can't simply un-screw them, you have to pry them out.
For this new effort, i decided to forgo the screw driver and needle nose method to breaking them off and this time got a new tool! The new red handled tool is: S & G Tool - Upholstery Clip Removal Tool Part # 87810 There are some fancy pliers types, they just did not have in stock so I got this on and it worked like a champ.
Lower Nut ... is here
The whole head lamp assembly must be puled out just to reach the sockets!
Almost There!
I did not break any this time but went ahead and replaced all 6. I had to buy a 6 pack when i stopped by the auto parts store so I would have a spare just in case. Since I would just loose them or get a new car before i have to do this again, i replaced all 6 with new clips! ( BTW - all the broken ones from before had already been replaced)
Target located!
The old bulbs had quite a bit of the enamel paint peeling off so i would have had fancy white turn signals shortly...
Fuzzy old bulb
New and shiny...
New!
Plastic things were easy as:
One...
...Two...
...Three!
Back together with a push ( x6) and good as new. I never got a picture of the tool in action, it went so fast, just slip behind and pry out less than 10 seconds they were pretty much done. It took at least a minute with the pliers and I broke the clips anyway. So are these mundane project blog stories pointless? Well, I guess the point of this story is;
Buy the tool, and plastic fasteners, when you pick up the turn signal bulbs.
..and the second point: there are two bulbs in a pack so replace them! I did both.
My assistant was a little tired after all the bunny guard duty he performed this morning but will be ready for the next project!
Tired Boy
~ ~ ~
BTW I got behind on blogging some projects, work and business travel to China got in the way... and i never can remember which phone/camera i documented the progress with. I'll release more drafts when I clean them up a bit!
We have a remote desert camp that go to when the weather is nice and play with radios. It is a pretty quiet spot both solitude and radio -wise. ( when the dune buggies are not running around)
Several years, at least a decade ago, On a camping trip here I convinced my nephew Nick there may be gold in the ground at about this spot. We proceeded to dig a 6 foot-deep hole and when we got to the bottom, I put in the 8 foot pole that is the support for this antenna and put all the sand and rocks back in it. I got the Hustler antenna from a ham club swap meet for $20 years earlier. It had been in service since the late '70s. Finally we dug some trenches and strung some tuned radials under the sand in all directions. They have been here ever since that day.
I used this antenna for many years. I never really took it down as this spot is so remote and nobody ever bothered it ...until last summer.
[bent antenna]
Mother nature had some fun one summer afternoon. I am not sure what kind of storm it was, the desert tends to get some violent thunderstorms in the summer as a monsoonal flow sets up from the warm waters of the Sea of Cortez inside of the Baja peninsula. In any case this storm left lots of sand sticking to the sides of the buildings poles and trailers so it guess that the force of the wind plus the mass of the rain and sand was too much for this 35-year old antenna. At least the bottom section.
Fortunately, DXE sells replacement parts "...allow you to keep your Hustler vertical performing at its peak. Whether for maintenance or repair due to unfortunate damage, Hustler offers the replacement antenna parts that you need." So I bought and replaced the lower mast section (P/N 4087-11) I think with shipping it was twice the cost i paid for the original used antenna but she is standing tall now:
The base of the antenna was in pretty good shape considering that it was basically in this desert environment for 11+ years. When I decided to fix the bent antenna I dismantled the entire antenna and took it back to the Radio Ranch repair labs. In the process I had to cut the radials to remove it as the bolts were badly rusted.
ANTENNA RADIALS
The new installation would not be able to attach the radials ( 2 per band) to the original tie points as they were slightly too short now. There would be no way to pull the radials back to make up for this with out digging them up - not an option with out a Nephew in site. ( he is now all grown up with two kids of his own to dig holes with) Instead, I fabricated a radial plate out of 4 Stanley angle mending brackets:
I simply formed a square ( 2 under, 2 over ) and fastened with #8 SS hardware. The bracket was bolted to antenna base with a salvaged TV antenna "U" bolt that happened to come complete with a mast clamp ( a second bracket that prevents crushing the square antenna tube:
But instead of preventing crushes, I drilled a couple of well-placed holes near the "top" edge of the clamp and used it as the bracket:
These were fastened with 1/2" and only one nut and lock washer as the space is tight between the mast clamp. Finally the original radial tie points screws were replaced with stainless screws, washers and nuts. Replacing the original "stack of nuts" that corroded before. I attached these points, in a bit of overkill, with two ground straps made up of a length of braid salvaged from an old RG-8 cable, they look pretty impressive anyway.
At the end of the day, the bracket is mostly buried in the sand anyway. The first 6" of the insulation of the original radials as you can see - is now long gone due to exposure to the elements, are now safely buried under sand and these small stones. The connection point and UHF connector are now wrapped in tape and a large tube of heat shrink as well.
THE WORKS.
BTW we made some PSK contacts with 5 watts on 15 and 20 meters and a couple of RTTY and - of course - CW. The contacts ranged from Cuba in the east to Japan and Russia in the west. All bands but 10 meters was tried but i think that band was not too active the day I tested it.
I know this is on a slow track but progress is progress.
90% Complete Power Amp
It has been a while since I posted any progress on this build. The main build took place, mostly on weekends between 9/1/2013 through 11/09/2013. The board has a couple of interesting quirks...
TIP: I used a Red Sharpie marker to mark uninstalled (DNI) components:
Note: C6 and C31 are Marked DNI Parts
TIP: Two soldering irons and a solder gun were also very useful in this build. Use two irons as heat sources for removing mistakes (caps and resistors) to heat both pads at the same time. I used the 100W gun to solder to the substantial ground plane.
RF Connection Mods.
First, the antenna jacks and RF layout is a bit difficult to figure out. I put shorting jumpers on the path to allow one antenna for TX and RX. This seemed the most straight forward for operations.
RF In Connector
Then I had issues with the 0.1" headers for RF connections. I used an Xacto knife to scrape a couple pads for mounting the jack. I tinned the pads and flowed a generous fillet of solder to the connector
RF In End-Launch SMA connector...
Antenna Connector
The antenna output is a vertical launch SMA. I could not find a suitable connector. While an end launch location would have been more convenient, but I could not figure out an RF friendly placement. There is a trace on the left-hand side of the board near the RF coils.
Instead, I modified a SMA connector by cutting off the mounting tabs. I then used a file to machine them flush (brass works easily) and used the existing antenna pad for the center pin. The ground connection was made by scraping the blue solder mask, tinning the copper exposed, and "sweating" a fillet with a generous amount of solder. The ground connection is probably adequate.
The modified SMA connector is seen mounted in this picture. It is pretty substantial and there is enough room for the connector, which will be a right-angle SMA plug:
New RF Connector near the White AXICOM Relay
If I would have planned ahead a little better, I would have done this before mounting the Relays and potentiometers. I did not want to remove or melt them by trying to solder the left-hand side, and had to be extra careful not to melt the plastic on the pot.
The Transformers and Coils...
Winding school.
Count the turns on the inside of the toroid cores. Scrape and tin the wires. The instructions are from the Soft Rock 50, plans you need to translate the part designators between the two different boards.
Coils (L7, L8, L9, L10 & L11)
The five toroid coils were pretty straight forward, 3 with 7 turns of AWG #22 and 2 with 5 turns on T-50-6 (Yellow) cores. The red enamel stained the cores a little when tinning the wire. NOTE: L7 is a little tricky to mount, be careful there is a large ground via close to the pads, don't solder to that hole! Here are the winding specs. from the BOM;
I did a manual sweep of the two filters. I am not too happy about the 10 meter filter. I need to check the values of the caps etc. It seems shifted to favor 12 Meters..
12M - 10M Filter Response
The plot of the 50 Meter pass filter looks pretty reasonable. Very much as expected.
6M Filter Response
Transformers (T1) and (T2)
These two were just as described:
"...[T1] on the PA board two loops of #18 wire through a BN43-202 Core. The wires are
cut to 3" (75mm) each and stripped so that the insulation is 1 5/8"
(65mm) long.'
(Seen on right side with Red and White #18 wire);
"...[T2] 3"
(75mm) length of RG-316, remove the outer jacket from one end and expose
0.75" (20mm) of the shield... Remove the outer jacked from the other end of the wire so that
the insulation length is 1.5" (40mm)....Loop
the wire through the BN43-202 core, unbraid the remaining shield wires
making sure that the shield on both ends of the wire point in the same
direction. Strip the dielectric from the two center conductors. Tin and trim
both shields and center conductors."
Seen on left side of the following picture. (basically, just a short length of RG-316 coax thru binocular core):
T1, T2 and L7 (Note via near L7)
Transformer (T3)
The mounting pad for T3 was not good. I applied a little too much heat and the pad lifted;
After close examination it appeared that there was no electrical connection to the pad so I attached the rather substantial transformer to the board with a dab of 2 part epoxy.
Transformer (T4)
The instructions are pretty clear for this transformer:
"Step 1: Wind 2 windings each using 18" of #30 Kynar wire, (Green) 10 turns, one winding
through each hole in a BN43-202 core, ( I wound in same direction)..
Step 2: Cut 2
pieces of #22 Teflon wire, (I used Red and White) Strip the ends so that the insulation
remaining is 7/8” long, the length of the bare wires is not important just so
it’s long enough to go through the PCB. Tin the ends:"
T4 on DSP-610 PA
I did not really follow the mounting instructions, I just pulled the wires thru marked and stripped the lengths to match, For the #22 wires I just measured the insulation and simply bent them to feed the leads thru the circuit board pads.
Transformer (L6)
The description for L6 was pretty vague... (T1 = L6)
"..[L6]
has 2 windings on a FT37-43 toroid: the primary consists of 2 turns of #26
enameled wire, stripped and tinned; the secondary is 13T of #26 enameled wire.
It's best to wind the secondary first then wind the primary over the secondary.
Try to form the leads to match the pads in the PCB..."
This was the most confusing part on the board for me. First a transformer labeled as an inductor? Anyway...The winding instructions are pretty straight forward but not a picture to back up the description. Here is what I came up with...
L6 as Wound
The best way I see is to run "two" turns is to loop it (green wire) so it passes through the torrid twice with the most wire through the toroid, exiting on the mounting side. Forming the leads is also a little tricky. To make it match the schematic, after some careful study of the PCB I ended up with this:
L6 with leads Formed
The First Part Was...
Oh, yeah. the first soldered part on the KD6VKF DSP-610 project was L1 on 9/1/13:
L1 Installed
The next installments; Transistors, DC Bias adjustments, and compete the testing the board. Please Stay tuned! (The KD6VKF DSP610 will eventually get done, really ☺)
