K3 Kit Building
From Zerobeat
Preparing for Assembly
There are very many very tiny parts, easily confused. The manual is excellent with clear text and warnings but sometimes it's just hard to see what you have in your hand. I used an OTT-Lite with 2X then grouped similar parts into clear baggies, taped the rest onto envelopes, and wrote the part information next to it in big letters.
All round head screws go on the bottom, flat head on the top - all teeny tiny and look remarkably the same to me. There are lots of parts that should not be forced and many screws that should not be over torqued. (WB8YQJ)
One of the best pieces of advice gleaned from this reflector prior to assembly was to inventory all the parts, right down to screw level, and store all the different screws, nuts, washers etc, in egg cartons labelled with their specific descriptions. It's much easier to see what's what when you are comparing them with each other, and if you have the wrong amount then that's a clue that one may have been put in the wrong compartment. And having sorted them out it's much quicker to find the correct part by just fishing in the labelled compartment. (G4ILO)
I spent some extra time during inventory because the screws were not always in the packages defined. But they were all there, once all the packages were opened and counted. (KH7T)
Lots of little screws, washers and nuts. You need to keep your head straight to be sure you keep the 3/16 apart from the 1/4. After a while, you can see the difference but in the beginning all the parts look the same. Also, I would have appreciated some larger pictures so I knew where I was supposed to be looking on the board. But, if you take your time, have a bright light and think about what you are doing, this is not difficult.
The only test equipment needed was a VOM to check a couple of continuities. The only tools were a small phillips screwdriver, a 1/2 inch wrench for the encoder knobs and a small diagonal cutter to trim the leads on a few parts. The required allen wrenches came with the kit. (K4IA)
Verify the clipping of any relay leads, if some are left unclipped from the factory, they can short on other metallic parts. This caused the VCO CAL to fail on several bands ... but not ALL. Always check all boards for clearance of leaded components. (K0PP)
Two ice cube trays is about right to hold all the little bits. It's nice to have about 20 sq ft of table space to work on the K3 and to lay components out during assembly. Good lighting is also essential - lots of little desk lamps to illuminate the nooks and crannies. A magnetic Phillips head screwdriver comes in handy, as well as a short "jeweler's" Phillips head (but not too small a head). I also want to pass along an excellent suggestion by WB8YQJ - if you have access to a good color printer, download and print out the Assembly manual to have a color version to work from - it helps locate components on PC boards. Alternately, open the PDF document on your nearby computer to refer to while you work. (N1EU)
RF Board and Chassis
One comment I might make is to temporarily screw down the 3 low power final transistors on the bottom of the RF board during construction. They kept catching on my ESD mat and twisting around during construction. (WA8EBM)
KANT3 or KAT3, Right Side and Rear Panels
The little rubber plugs on the side panel are a bugger to install. Avoid the temptation of using more than fingers. (WB8YQJ) I found using a drop of liquid soap on each plug made the insertion fairly easy. (N1EU)
The manual is good except that it sometimes it leaves the assembler with some doubt. The first occasion is where you have to install the right side panel. You do it but you are told the 3-terminal regulators that should be secured to it will be attached "later". You think, well, later down the page but no. You go on to assemble two other panels so you begin to fret about the unattached regulators. Several pages later the manual instructs you to remove the side panel for another purpose. You secure the regulators when the panel is reattached.
Attaching then removing the side panels has caused the paint in the flat head screw holes to flake off. Now I have to use a black magic market to "hide" the exposed aluminum. Minor, yes, but it is an annoyance. (KC0OV)
Only problem I have encountered so far was inserting the chassis ground thumbscrew through the back panel into the mating connector on the RF board. I had to loosen the rear lower screw on the left-hand side panel temporarily to get that thumbwheel in without cross-threading, and even at that it's a very tight fit when you re-do the side panel screw. The suggestion to use that technique was removed from the instructions by Errata B8 of Rev B1 of the Assembly Manual, and I have sent a note to Elecraft suggesting they put it back in. Cross-threading of the chassis ground screw in the RF board would seem to be rather bad news and to be avoided. (VE1RGB)
I was having difficulty mating J70 on the KANT3 board with P70 on the rf board. Looked at the connector and found that one of the holes was plugged. An email to Eric brought a quick reply that that was supposed to key the connectors but that the corresponding pin on P70 hadn't been clipped. (W6JD)
The bottom uses pan-head screws throughout. The top and sides use countersunk screws. (K6KR)
KIO3/KXV3 Interface
A tip I got from Wayne when experiencing RF feedback on PSK31: Seat the two KIO3 rear panel connector boards (3.5mm fems, and also the RCA / 1/4" phono) so that there is a slight amount of interference with the screw-on panel. This allows the grounds for the jacks to make best contact with the 3.5mm plugs. It takes a little jiggling to get the fit to come out just right, but you need the jack bodies (sleeve is ground) to jut out just a little bit to get proper contact.
With the KIO3 pushed in as far as it would go the bracket of the accessory socket was still about half a millimeter too far out. Because only two screws are used to secure the faceplate there is a slight gap at the other corner because the flange of this socket pushes against it. (G4ILO)
Front Panel and DSP
The K3 front panel has 4 connectors to mate, all at the same time, which is nearly impossible with the 2D fasteners mounted up front. Save some time and remove these two, perform the mate, then slide them in there and tighten them down. This takes all the frustration out of it.
The front panel has 4 main standoffs, all 5/16ths long but 2 are drilled at 2-56 and 2 are 4-40. I can't see the difference between those sizes myself but you need to take the time to know that there are 4 in the process, compare them against each other, then mount them where they go - specifically. (WB8YQJ)
I want to expand on the above point about the 2D fasteners. At first I didn't remove any 2D fasteners because it looked to be a bear to screw them back in after the Front Panel eventually was in place. But what I found after struggling to get the FP board mated was that on my kit, it was actually IMPOSSIBLE to have ever plugged in the FP because the inside 2-D screw near the mic jack abutted the front panel pc board and prevented the adjacent pin connector from engaging. I missed the tip in Figure 51, which would have saved me troubleshooting time. That was the only 2-D screw I had to remove. I had no problem getting the screw back in (jeweler's screwdriver) after the FP was in place. (N1EU)
Fitting the front panel was the hardest job. Having eyesight that has deteriorated with age, I could not focus close enough peer through the small gap to see if the innermost connectors had mated. A magnifying visor limits the distance range that you can see in focus too much. Eventually I managed it but the help of a keen-eyed youngster would be invaluable at this point. (G4ILO)
A slight misalignment of the rubber belt finger grip on the main tuning knob makes it appear to wobble slightly while spinning the knob. (KH7T)
Be sure you don't use two of the short 2-56 screws where the longer ones go for the VFO plate. (W0YK)
The siting of the front panel board on the panel itself took me 45 mins I think I was a little timid it did not seem to want to go all the way at the mic socket end. the LEDS were lining up OK but after a while it just slipped into place everything looking perfect. It is a tight fit but then it has to be. Be patient here! (G4APO)
While attaching the front panel to the front panel board it was virtually impossible to get the mic connector through the corresponding hole and keep the rest of the board aligned. The instructions make mention of this possibility and suggest "wiggling" the board and panel until they mate. In this case there just plain wasn't enough wiggle room so I used a sharp pocket knife to slightly increase the diameter of the hole with no harm to the panel. (W6JD) I also had to do a little scraping with a small pocket knife and it worked like a charm to allow the mic connector to seat. (N1EU)
Power Supply Cable
It's at this point that the power supply cable needs to be wired. Supposedly Elecraft recommends soldering the Anderson PowerPole connector pins to the cable, although the instructions don't say so. You definitely should solder the pins unless you have a suitable crimping tool that doesn't just crush (and distort) the pins, like the cheap tools do. Otherwise, the pins won't seat properly in the connector. (N1EU)
KREF3
Be sure to clip all the longish leads on the Ksyn3 board, view with magnification and light before and after installing the board. I had one nearly invisible relay lead that I clipped but it did not fall away. The calibration sequence fails when it gets to the lead that is touching the nearby shield. (WB8YQJ)
The three TMP cables connecting the KREF3/RF BOARD/ KSYN3 boards were difficult to insert in their respective sockets mine were a very tight fit 1 cable had been crimped over the plug end itself but I managed to reshape this using pliers and insert it without damage to the cable or the socket. you have to get them perpendicular to get them in easily and applying an even pressure. (G4APO)
KPA3 Shield
It went enjoyably as usual, but all the PA connectors take quite a bit of physical force and you really need to support the RF board carefully to keep it from an over flex condition.
Press -real- hard, support the board from the bottom, and don't zap it with static while doing it. ;-)
I pressed from the bottom with a folded up anti static bag under my fingers.
If you follow the instructions, there is a structural bar that you are supposed to mount on the K3 lid, dismount, then mount again in the K3. I mounted this in the K3 and left it there, was able to work around it just fine while installing the PA.
You are requested to pre-test fan operation before mounting the actual PA, the leads are VERY short for fan power, just mount the fan assembly with a screw for strain relief before connecting to power. When you go to enable the fans in the setup, don't panic, they probably won't work. Exit the setup, power up, power down and they should go. (WB8YQJ)
Despite having the correct number I found I was two of the longer pan-headed black screws short when it came to fixing the driver and PA transistors to the bottom cover. I used the identical length silver screws, of which I had several spares. Installing the bottom cover I found that the transistors were slightly too proud, preventing the bottom cover from fitting properly. The low power PA board was actually flush with the RF board when installed, so I lifted it a bit to raise the transistors and allow the bottom panel to fit flush. (G4ILO)
KNB3 Noise Blanker
I was a little concerned after installation of the Noise blanker board as it did not sit level to the RF board. I thought that the standoff I had installed was a little short. Measuring it it was correct as per the manual . Removing the board got me a little worried as there appeared to be a pin missing from J77 and the hole in the socket was filled!
I thought that somehow the pin had broken off.
Phone call to Elecraft Tech Help relieved my worries here. Not all the pins are present and used on a lot of the connectors of the boards in the kit. The reason it was not sitting as it should non the socket was that the unused pin had been cut and left a little long. Filing this down cured the problem. (G4APO)
Like G4APO, I was also concerned that after installation, the Noise blanker board didn't sit level to the RF board. I added a third split ring washer to the assembly so there are now two on top side of the board, raising the spacer just enough to level it out. (WA6VNN)
Top Cover
I managed to chip off a bit of paint on the top edge of the top cover. A black marker made the 1/8 inch line of exposed aluminum nearly invisible, but I know it is there. I also noticed that tightening the cover screws tends to peel some of the paint in the countersink area for the screw heads. (KH7T)
KPA3
Just about the very last assembly step is dropping in the amplifier assembly. There are two pin connectors that must seat completely. It is unbelievable how much force you have to exert (while supporting RF board from below) to seat these connectors and likewise unbelievable how suddenly it pops in with a loud snap. Eat your Wheaties before doing this. (N1EU)
Test and Calibration
I'm getting the ERR PL1 error on 28MHz after the synthesizer calibration. Other bands seem OK & is otherwise working fine.
M0XDF replied: "ERR PL1/2: VPLL out of range on band change (to view actual PLL voltage, set CONFIG: TECH MD to ON, then tap [DISP] and use VFO B to locate the PLL1 and PLL2 voltage displays). Verify that the oscillator can on the KREF3 is fully plugged in and is not in backwards. Make sure all internal cables are plugged in, specifically the cables between the KREF3 and KSYN3 modules (synthesizers). Try re-calibrating the applicable VCO (CONFIG:VCO MD) (tap SUB within the menu entry if you saw ERR PL2, to make sure you¹re calibrating the sub receiver¹s synthesizer). If this doesn¹t work, try removing the 2nd synthesizer (for the sub receiver), and set CONFIG:KRX3 to NOT INST. If this eliminates the error, the sub synth may be defective. You can also try swapping it with the main synth to see if it can be calibrated in this slot."
While doing the initial synthesizer calibration I keep getting an error message. I get "Error VCO" and the
error code is "00003". When I turn the rid on I get the "Error PL1" message. I did check jumpers and the
crystal as suggested and they are correct. Any ideas?
AD5X replied: "I had the same problem. It is probably the same issue - the KSYN3 card VCO inductor is slightly offset. Elecraft offered to replace the card, or I could try to adjust the range by tweaking the VCO coil. Per their instructions, I removed the sleeve around the coil (unscrew the nylon nut) and slightly expanded the turns on the coil, and then reassembled the sleeve (they told me I should expand the turns slightly). I went too far and got a different VCO error message. So I tried again and this time compressed the turns slightly in between where they had been and my first attempt. That fixed the problem - my synthesizer calibrated fine. If you do this, just very slightly expand the turns on the torroid. You can also see your PLL voltages on each band using the DISP function when you are in the TECH MD mode. The PLL voltages on each band should be between 1-6.7VDC. Probably the best thing to do is to record these voltages (and you'll probably see where the problem is). Tweak the VCO coil and look at the voltages again and see where they shifted."
If you are getting an ERR TXG error message when attempting to calibrate the power on 60 meters, try doing the calibration on one of the 60M channels. Try 5403.5 kHz (AB7R)
