WB4SON

It’s been about 5 days since I did any work on the K1.  This time of year it’s important to keep the family happy, and that means lots and lots of Christmas decorations.  I hate the thought of landfilling the many sets of lights we use each year, so I insist on repairing them.  This year I replaced several hundred lamps, which occupied my time for several days.  But the lights are up, the house is festive, and my time is my own again.

Today I began the assembly of the K1 RF board.  I estimate that this one card is about half the total assembly effort — lots of parts.  But I surprised myself today by spending about 70 minutes to install 40 components; under 2 min/part.  When I did the inventory of the RF board parts, the tiny caps were a devil to read, so after accounting for each one, I put them into individually labeled bags.  I still verify them, but at least I don’t have to look through dozens of identical looking caps to find the one I need.  And I discovered that the resistors, which had been sequenced on tape as if for automatic assembly, were in the order that they are installed.

I’d say this is about 1/6 of the work (2/6 for the Front Panel, and 3/6 for the RF board remaining).  It took a total of 255 minutes to complete the installation of 64 component, or about 4 minutes per part (that would put the total assembly time for the K1 at about 25 hours).  That includes all of the final assembly tests (resistance, shorts, opens, correct placement, etc.) as well as quite a bit of time spent with the toroids (especially the incorrect diameter parts).

Things were going swimmingly with the K1 Filter Board assembly, still at a rate of 1 component per 3 minutes (I follow a 7-step process:  check value, check value, install, check value, solder, check value, check solder).  Then I hit a snag…

When you order the K1, you select which two bands you would like.  I picked 30m and 20m, two of my favorite CW bands.  I had done a careful inventory of all parts and found them all there.  However, the Bill of Material didn’t specify the toroid size, just the color.  I had two T50-6 cores with the 30m kit, and two T37-6 cores with the 20m kit.

When I got to the assembly instructions for the 30m L9 and L10 toroids, I found they should have been on a T37-6 core too — the cores I had were a bit larger.  The desired inductance (1 uH) was known, specified at 18 turns on a T37-6 core.  I verified that, then reran the calculations using a T50-6 core, and found it would require 16 turns to maintain the 1 uH value.  I then called Richard at Elecraft and asked his opinion — “no problem, use the T50 cores and just take a turn off”, was his advice — so 17 Turns it will be.

The only down-side I see is that the T50 cores, being a bit larger, put L9/L10 a bit closer, and it is possible there might be an issue with the automatic antenna tuner since the toroid stands tall.  However it is exactly the same height as the slug-tuned inductors in their metal cans, so surely the board spacing will be adequate to clear both.  Besides, at the moment I have no plan on using the internal tunger since I have several external ones.

Needless to say, I verified the diameter of every other toroid in the kit.  Fortunately they appear to be correct.