text=#ffffff vlink=#ff0000 hlink=#ff0000>

Building Elecraft KX1 ser#1727

~ ~ ~ ~ ~ ~

It arrived today 15 Mar 07. My Elecraft KX1 ser#1727. The ELECRAFT ULTRA PORTABLE XCVR for 40m/30m/20m.
I ordered the 30m mod kit and the auto tuner "KXAT1" for it. Inventory has been completed. Just a 2-56 was screw
missing along with a couple of #4 washers. Good job Christine and Marina.

On the left you can view the top of the PCB. On the right the Bottom.
This is my second KX1 build.
My first was KX1 #1056. I had forgotten
about the low parts count. The PCB is actually smaller than a standard QSL card.
Did you notice the three SMD components that come pre-installed?

BTW I used my scanner on these first pictures

On to Assembly Part I

Pages 14 thru 26

Even though I only install one part at a time.
I somehow managed to install the wrong component
the very first time. Off to a great start.

Visually checking helps C7 was put in backwards

Assembly Part I complete
resistance checks completed infinity readings (although correct )
for banded end of D2 or D3 with S1 on or off does not feel right?
~ ~ ~ ~ ~ ~

Alignment and Test - Part I
Pages 27-29

Here are the voltage checks. ALL checks OK!

~ ~ ~ ~ ~ ~

Assembly Part II
The Receiver
Pages 30 - 42

Page 35.
I'm about half way on Assembly Part II. So far I have manage to correct my mistakes
at the time they happened L5 (5.6uh) broke a quick phone call to Elecraft had a replacement
on it's way and D5 (1sv149) was put in backwards. The flat back on the 1sv149 turned it around.
Thats four assembly errors to date. I need to go slower.

Assembly Part II completed

Resistance checks are okay.

~ ~ ~ ~ ~ ~

Alignment and Test - Part II
Pages 43 - 47

~ ~ ~ ~ ~ ~

All portions of Alignment & Test Part II passed with flying colors.
Power Supply of 13.1 VDC, bat read on KX1 12.7vdc. Peaked the 20m
band using K2 @ 0.1w into DL1 same om 40m. Really enjoyed copying
signalss on both bands. RX current drain with "LED ON" 38.2ma with
"LED OFF" 32.8ma.

~ ~ ~ ~ ~ ~

Assembly - Part III
The Transmitter and LPF
Pages 48 - 57

~ ~ ~ ~ ~ ~

Page 51

All components installed except for the PA.
Prepping the case is next then install the PA and The Final assembply

PA has now been installed and resistance checks made. ALL Okay!

Top cover installed. Looking FB.

Internal battery compartment installed

Alignment and Test - Part III
Pages 58 - 59

All checks passed. Power out on 40m & 20M with a 12.6vdc power supply is slightly over 3 watts. ~ ~ ~ ~ ~ ~

Final assembly has also been completed.

~ ~ ~ ~ ~ ~

Thirty Meter Mod Next

~ ~ ~ ~ ~ ~

Here 's the PCB, both top and bottom views

The 30m Mod went together real easy.
Be careful when installing D1.
It is very easy to put it in the wrong holes.

I won't install the 30m until after the KXAT1 is built.
I don't wish to break this radio down more than needed.

30m mod shown installed

~ ~ ~ ~ ~ ~

Now the KX1 Antenna Tunner

~ ~ ~ ~ ~ ~

Here's the PCB, both top and bottom views

KXAT1 is now fully assembled and ready to be installed in the KXAT1

KXAT1 has been installed.

End on view of the KXAT1 mated to the KX1 PCB. Notice the interlocking clearance.

~ ~ ~ ~ ~ ~

Now a series of email between myself the Elecraft reflector and Don Wilhelm w3fpr

The construction of KX1 #1727  went really smooth all the way through the build. 
ALL the "checks" were well within spec's. It was even okay after adding the 30m mod 
with superb RX  Power out was about 3watts. Then I built and installed the ATU during 
the alignment and test the first several checks on L0-L3, C0-C3, N0-N3 were okay! 
When I tried to calibrate the ATU "Nada". I noticed the PA was "HOT". So was my DL.
The SWR reading was r9.9 Power was "P0.0". Now when checking the RX is dead. 
Audio (white noise)is there and strong so is side tone... Any Ideas where to look?
RC kc5wa 

From: "Don Wilhelm" w3fpr
To: "RC kc5wa" 
Date: Wed, 7 Mar 2007 08:10:02 -0500


It could have been just a co-incedence that the KX1 failure occurred while
you were testing the KXAT1.

First check the voltage at pin 8 of U5 and U6 - you should have about 6
volts.  Check the voltage at Q7 base, it should be zero during receive.  If
the voltages are not correct, stop and let me know.

I would proceed to check out the receiver first, then move on to the transmitter.
First use an RF probe (or 'scope) to check for the oscillator injection to
the mixer and Producr Detector.  You must have more than 70 mV RMS (200 mV
p-p) at pin 6 of U5 and U6 to continue.  If you do not have those levels,
you have a problem with the oscillators - check the soldering (look at the
schematic to identify which components are associated with each oscillator.

After you check for the oscillator injection levels, put a capacitor in
series with the center conductor of an antenna and touch it to pin 1 of U6 -
if you hear anything (increase in noise, signals, etc), then the receiver is
alive and the problem is in the front end tuned circuits.

If you do not hear any increase above, touch the capacitor to U5 pin 1.  If
you hear any increase here, the product detector and audio are working and
the problem is in the IF (which consists mainly of the crystals), so look
for a poorly soldered crystal.

Let me know what you find with the above checks.



Date: Thu, 08 Mar 2007 11:16:37 -0600
From: RC kc5wa 
To: Don Wilhelm w3fpr

Here are the first series of tests
U5 pin 8 measures 5.71vdc
U6 pin 8 measures 5.70vdc from the under side
Q7 base measures 0.00vdc
PS 12.99 vdc
Using my K2 RF probe.

I obtained the following readings U5-6 .129 vdc U6-6 .135 vdc using a capacitor on the antenna and touching U6-1 the volume increased sufficiently that the AF had to be turned down ( RF was @ max, filer full wide, AF was @ max). Same results when touching U5-1. RC ***** From: "Don Wilhelm" w3fpr To: "RC kc5wa" RC, All that looks good, you have a good receiver from the mixer on through the audio. Next step is to touch the antenna to the RX ANT point that you will find on the schematic just to the left of the low pass filter (lower right corner). If you have a signal there, but not at the antenna jack, then there is something wrong with the low pass filter - if you have no more signal at that point than you have at the antenna, there is something awry with the tuned circuits in the front of the receiver. If you do fine it is the LPF, check the toroids and caps there carefully for soldering. It may be that you simply bumped something when you were adding the KXAT1 and its connection was broken. As always, check soldering and toroid leads first - there are no active components in the receiver until you get to the mixer pin 1 (U6), and you found things good there. 73, Don W3FPR ***** Date: Tue, 13 Mar 2007 19:18:53 -0500 From: RC kc5wa After checking for a signal increase at "RX ANT" using a capacitor on the antenna lead between C44 and L1 at C45. I do not get a signal increase like I did at U6-1 or U5-1. I do think a heard a slight swbc signal, not much though.. RC kc5wa ***** From: "Don Wilhelm" w3fpr To: "RC kc5wa" Date: Wed, 14 Mar 2007 00:34:55 -0500 RC, Take a look at the schematic in the KX1 and KXB30 manual which indicates the components in the receiver front end tuning circuits. Somewhere in there, you have something wrong - perhaps a poor solder connection, perhaps something tuned incorrectly. What is certain is that the receiver 'works' if you can get a signal to the mixer. The tuned circuits have to pass the selected band signal to the mixer. Try peaking things again - in order - on 20 meters, peak the 20B cap, then the 20A cap. After that, peak the 40 meter cap, and last peak the 30 meter cap. You must use an insulated tuning tool. Attempts to adjust these caps with an uninsulated tool is possible by the 'set and try' method, but IMHO is an exercise in creating frustration. If you cannot find a peak, check and recheck the soldering at all of the components involved. If things are tuned right, you should get a signal path to the mixer (which you know is working). 73, Don W3FPR ***** Date: Wed, 14 Mar 2007 13:30:30 -0500 From: RC kc5wa Well Don I don't really know how it happened. I started by peaking 20m then 40m next 30m then started backtracking from U6-1 using a capacitor on the antenna looking for a break (a non increase of noise) in the signal train. I reflowed each solder joint and went on to the next component. I turned it back on to see if it was better tuning the VFO looking signals. On 20m I copied a rag chew with W7CI 14.165 then checked 40m heard signal at 7.040 and the AM carrier (both sides)on 10.000mhz and the time hack when nulled. So it seems I might have stumbled on the issue and fixed it. Later on this evening when signals are more plentiful on the bands I'll repeak the bands for max signal strength. The sticky on the pads for the 30m mod is no good it won't stick. Will foam tape be alright? I'll attempt to install the ATU later. RC kc5wa ***** RC, It certainly sounds like you have it fixed. I would say the most likely was a marginal solder connection. Yes, that sticky on the rubber pads is not the best. Foam tape tape should work just fine as long as it is not double sided sticky! I usually put a drop of my wife's hobby glue (Goop or something like that) on the original pad. 'Shoe Goo' also works fine. 73, Don W3FPR ***** From: "Don Wilhelm" GE Don I have repeaked the 20m 40m & 30m bands. Even installed the ATU. RX FB.With ATU in circuit readings are with R4 fully clockwise and dummy load on the antenna "P 0.0" "r1.0"...............With ATU out of circuit R4 fully clock wise and a dummy load readings are 16.53 vdc about 3 watts 12,6 PS. I might need to go over the ATU again. Alignment test on ATU "C9" null show 0.5vdc on 200mvdc scale. RC kc5wa ***** From: "Don Wilhelm" Date: Wed, 14 Mar 2007 22:28:20 -0500 RC, Well if you are getting actual power out and the KX1 display shows 0.0, then there is something wrong with the wattmeter circuit. If it did not work before, check the soldering, particularly the leads of the toroidal bifilar wound transformer. Make certain the correct color leads are in the correct holes. If it has worked properly before, still suspect the soldering, but you will be looking for an intermittent connection. If soldering is not the problem, then you can suspect the diodes. That wattmeter null is not deep enough - you should be able to achieve a null of less than 10 mV if all is well with the tuner. An unsoldered relay pin would do the deed of leaving excess residual inductance or capacitance on the tuner path even though it was set to CAL S or CAL P menu options. 73, W3FPR ***** Date: Thu, 15 Mar 2007 11:47:09 -0500 From: RC kc5wa Don I found the problem with the KXAT1..... Boy do I feel dumb.I built it late one evening. I reversed the #1 #3 leads when I installed T1. RC ***** From: "Don Wilhelm" w3fpr To: "RC kc5wa" RC, You are not the first one nor will you be the last to do that. Congratulations on finding it. 73, Don W3FPR **********

Now! I need to rebuild a T1 for the KXAT1 ATU and install it "CORRECTLY" 
then do the ATU alignment over!


I am using an external power supply of 12.01 vdc.
The voltage readings I believe are RMS voltages read off the dummy load with a DMM.

Using the graph the reading correspond to about 2 watts of output power a small per cent.

The frequency is where I made the readings:

The "P60.3" and the "r1.0" were read from the KX1 display.

7.056 mhz P6.3 r1.0 DMM reading 13.29 vdc. Turn it down until the indicated power is equal to the power computed from the RF Voltage - P=V2/R. In other words, the 40 meter output is 3.53 watts for a measured RF Voltage of 13.29 Vrms.

10.133 mhz p6.3 r1.0 DMM reading 15.00 vdc. the 30 meter output is 4.5 watts for a measured RF Voltage of 15.00 Vrms.

14.102 mhz p6.3 r1.0 DMM reading 14.45 vdc. the 20 meter output is 4.1 watts for a measured RF Voltage of 14.45 Vrms.

I suspect that I did not calibrate it properly and should redo that.

RC kc5wa Now to the difficulty with the bumper pads on the back of the 30m mod.

I trimmed the scrap from a pack of radio shack cushion feet and placed in on the back of the 30M PCB. This should prevent shorting to the case and it doesn't fall off

~ ~ ~ ~ ~ ~

Shown here with Morse Express 2004 Camelback Chrismas Key and
the American Morse KK1 key kit
KX1 #1727 is now buttoned up and operational and on the air.

72/71 de "rc" kc5wa