Elecraft › [K2]

K2: Debugging low transmit power

Classic

List

Threaded

16 messages Options

omehegan

K2: Debugging low transmit power

Hello -

Two nights ago I completed the assembly of my K2's RF board, and after
a visual inspection verification of resistance checks, I assembled the
radio per the instructions and began the third phase of alignment and
test. I've discovered a problem right off the bat, though. In the 40
meter transmitter alignment section, when I put the K2 into tune mode
at 2 watts power, I'm only reading .3-.4 watts on the onboard
wattmeter (it alternates between the two). This is with a connection
to a 100W dummy load. Varying L1 and L2 has no effect on the power
reading.

I discovered one mistake: I misread the instructions for winding T4's
5-6 and 7-8 windings. I wound these bare wires through BOTH holes in
the core, in a U shape, thus connecting 6-7 and 5-8 (and very possibly
all 4 together with a short in the middle - this is what comes of
staying up too late working on a project!). I corrected this mistake
tonight, but I'm seeing no change on the measured power. Is it
possible that this mistake fried something? I never saw any smoke or
smelled burned silicon, and the thermal pads on Q7 and Q8 look normal.

I have also noticed that if I vary the power adjustment on the front
panel while in tune mode, I see no change on the wattmeter. That seems
significant.

These are the other checks I've done so far, per the troubleshooting
instructions:

* As far as I can tell, my current limit is set to 2.5A. I have not
seen any "high current" warnings.
* I am using a dummy load to test, I do not currently have an antenna
set up that I can use. I have tried two coax jumpers for the dummy
load, both test OK on continuity. No change on power output.
* The bottom cover is correctly installed and there are no shorts to
it that I can see.
* I checked all component values in the RF detector, they are correct.
* I examined transformers T1-T4, all appear to be wound and wired
correctly.
* I checked R50, it is measuring the correct resistance.
* I checked for hot components and found none.
* I checked visually, but not yet with a DMM, for shorts and opens.

I have NOT yet done the following:

* Looked for shorts in the LPF or BPF - visual inspection shows
nothing unusual. How exactly would I check this section with a DMM?
* Checked DC voltages in the transmitter and ALC circuitry - I will
need to re-assemble the K2 somewhat before I can do this.
* Used the RF probe to trace signal flow.

Any further suggestions at this point would be greatly appreciated. I
can get access to an oscilloscope if that would be useful. Suggestions
of the "check x for voltage/resistance y" would be most helpful
(though I realize there is a voltage table in the manual that I need
to go through).

Many thanks in advance!

--
[hidden email] (Owen B. Mehegan)
'You write the lie
You'd like to be
When your life looks like a book you wouldn't read'
--Blake Schwarzenbach

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

Don Wilhelm-4

Re: K2: Debugging low transmit power

Owen,

I am going to assume that you were successful with the Alignmant and
Test Part II and were able to receive 40 meter signals. If not, you
need to backtrack a bit and get receive working first.

The easiest way to troubleshoot a "lack of power output" problem is to
do Transmit Signal Tracing. You can use an RF Probe or an oscilloscope
with a 10X probe.

Turn the manual to the Appendix E section titled Transmit Signal Tracing
and begin there. If using an RF Probe, the correct expected values are
listed, but if using an oscilloscope, multiply those values by 2.8 and
observe the peak to peak voltage of the RF waveform.

An important note - the K2 will normally increase drive in an attempt to
increase the power output if the voltage out of the RF detector is too
small. What that means is that all RF voltages prior to the failing
stage the RF voltages will be higher than the listed expected values.
You will be investigating the transmit chain in the order listed (do not
skip around), searching for the *first* stage where the output is lower
than expected - that is the failing stage which then needs to be
analyzed to determine why it is not operating properly. Measurements
beyond this first failure point are not relevant to that first problem
to be found.
All RF Voltage measurements in the list can be done from the top of the
RF Board. Do not transmit with the heat sink removed. If the PA
transistors do get a bit of drive, will produce output and if the heat
sink is not in place, they will quickly overheat and fry themselves.

Normal failures are soldering problems, toroid leads or incorrectly
placed components. Active device failures are way down of the list of
failure suspects unless something has caused it to be abused.

73,
Don W3FPR

On 10/5/2010 2:38 AM, Owen B. Mehegan wrote:

> Hello -
>
> Two nights ago I completed the assembly of my K2's RF board, and after
> a visual inspection verification of resistance checks, I assembled the
> radio per the instructions and began the third phase of alignment and
> test. I've discovered a problem right off the bat, though. In the 40
> meter transmitter alignment section, when I put the K2 into tune mode
> at 2 watts power, I'm only reading .3-.4 watts on the onboard
> wattmeter (it alternates between the two). This is with a connection
> to a 100W dummy load. Varying L1 and L2 has no effect on the power
> reading.
>
> I discovered one mistake: I misread the instructions for winding T4's
> 5-6 and 7-8 windings. I wound these bare wires through BOTH holes in
> the core, in a U shape, thus connecting 6-7 and 5-8 (and very possibly
> all 4 together with a short in the middle - this is what comes of
> staying up too late working on a project!). I corrected this mistake
> tonight, but I'm seeing no change on the measured power. Is it
> possible that this mistake fried something? I never saw any smoke or
> smelled burned silicon, and the thermal pads on Q7 and Q8 look normal.
>
> I have also noticed that if I vary the power adjustment on the front
> panel while in tune mode, I see no change on the wattmeter. That seems
> significant.
>
> These are the other checks I've done so far, per the troubleshooting
> instructions:
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

bdenley

Re: K2: Debugging low transmit power

?Don:
If Owen also has no transmit power on the other bands then are his symptoms
identical to mine?
Brian Denley
http://home.comcast.net/~b.denley/index.html

----- Original Message -----
From: "Don Wilhelm" <[hidden email]>
To: "Owen B. Mehegan" <[hidden email]>
Cc: <[hidden email]>
Sent: Tuesday, October 05, 2010 10:01 AM
Subject: Re: [Elecraft] K2: Debugging low transmit power

> Owen,
>
> I am going to assume that you were successful with the Alignmant and
> Test Part II and were able to receive 40 meter signals. If not, you
> need to backtrack a bit and get receive working first.
>
> The easiest way to troubleshoot a "lack of power output" problem is to
> do Transmit Signal Tracing. You can use an RF Probe or an oscilloscope
> with a 10X probe.
>
> Turn the manual to the Appendix E section titled Transmit Signal Tracing
> and begin there. If using an RF Probe, the correct expected values are
> listed, but if using an oscilloscope, multiply those values by 2.8 and
> observe the peak to peak voltage of the RF waveform.
>
> An important note - the K2 will normally increase drive in an attempt to
> increase the power output if the voltage out of the RF detector is too
> small. What that means is that all RF voltages prior to the failing
> stage the RF voltages will be higher than the listed expected values.
> You will be investigating the transmit chain in the order listed (do not
> skip around), searching for the *first* stage where the output is lower
> than expected - that is the failing stage which then needs to be
> analyzed to determine why it is not operating properly. Measurements
> beyond this first failure point are not relevant to that first problem
> to be found.
> All RF Voltage measurements in the list can be done from the top of the
> RF Board. Do not transmit with the heat sink removed. If the PA
> transistors do get a bit of drive, will produce output and if the heat
> sink is not in place, they will quickly overheat and fry themselves.
>
> Normal failures are soldering problems, toroid leads or incorrectly
> placed components. Active device failures are way down of the list of
> failure suspects unless something has caused it to be abused.
>
> 73,
> Don W3FPR
>
> On 10/5/2010 2:38 AM, Owen B. Mehegan wrote:
>> Hello -
>>
>> Two nights ago I completed the assembly of my K2's RF board, and after
>> a visual inspection verification of resistance checks, I assembled the
>> radio per the instructions and began the third phase of alignment and
>> test. I've discovered a problem right off the bat, though. In the 40
>> meter transmitter alignment section, when I put the K2 into tune mode
>> at 2 watts power, I'm only reading .3-.4 watts on the onboard
>> wattmeter (it alternates between the two). This is with a connection
>> to a 100W dummy load. Varying L1 and L2 has no effect on the power
>> reading.
>>
>> I discovered one mistake: I misread the instructions for winding T4's
>> 5-6 and 7-8 windings. I wound these bare wires through BOTH holes in
>> the core, in a U shape, thus connecting 6-7 and 5-8 (and very possibly
>> all 4 together with a short in the middle - this is what comes of
>> staying up too late working on a project!). I corrected this mistake
>> tonight, but I'm seeing no change on the measured power. Is it
>> possible that this mistake fried something? I never saw any smoke or
>> smelled burned silicon, and the thermal pads on Q7 and Q8 look normal.
>>
>> I have also noticed that if I vary the power adjustment on the front
>> panel while in tune mode, I see no change on the wattmeter. That seems
>> significant.
>>
>> These are the other checks I've done so far, per the troubleshooting
>> instructions:
>>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html

omehegan

Re: K2: Debugging low transmit power

Hi Brian -

I haven't tried transmitting on any other band yet, but your
description of only getting about .5 watts out when you tune, and L1 and
L2 having no effect on this, mirrors my own.

Tonight I will begin tracing the circuit with the RF probe. I also have
access to some more sophisticated test equipment at a friend's house, so
if I can't make any progress on my own I'll enlist his help. I am
concerned that my cheap-o Harbor Freight DMM will not work with the RF
probe, but my friend has a Fluke which should be more than adequate.
Hopefully I won't have the same frustrations you've had Brian. I'll be
sure to post an update on my progress tonight or tomorrow.

Thanks for your reply Don!

On Tue, 5 Oct 2010 19:06:53 -0400, "Brian Denley"
<[hidden email]> wrote:
> ?Don:
> If Owen also has no transmit power on the other bands then are his
> symptoms identical to mine?
> Brian Denley
> http://home.comcast.net/~b.denley/index.html

--
Owen B. Mehegan ([hidden email])
"He is a dangerous mixture of sophistication and recklessness which
makes one anxious about his influence on other boys."
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

Don Wilhelm-4

Re: K2: Debugging low transmit power

In reply to this post by bdenley

Brian,

Even though Owen's outward symptoms are similar to yours, the failure
point is likely quite different.
That is why Transmit Signal Tracing is important. There are 5 active
devices in the transmit chain as well as the bandpass and low pass
filters, and the oscillators (VFO and BFO) can cause low output as
well. The only way to know which one is the culprit is to move forward
with the transmit signal tracing. There is no one single answer, each
case is different.

Every solder connection and each toroid lead is a potential failure
point, not to mention those components that jump out of the correct
holes and end up in the wrong places just before soldering.

73,
Don W3FPR

On 10/5/2010 7:06 PM, Brian Denley wrote:
> ?Don:
> If Owen also has no transmit power on the other bands then are his symptoms
> identical to mine?
> Brian Denley
> http://home.comcast.net/~b.denley/index.html
>
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

omehegan

Re: K2: Debugging low transmit power

In reply to this post by Don Wilhelm-4

Thanks again for the suggestions Don. Tonight I started doing some
signal tracing, and my RF probe seems to work fine with my DMM.
However, I ran up against a different obstacle: I don't have a signal
generator handy, and the LPF/BPF/T-R Switch tests seem to require one.
I can get access to one but it won't be until tomorrow at least.

I did find one other data point for my symptoms. On SOME bands, I get
the expected power out when I put the K2 into tune mode. Here are the
results of my tests:

80m sometimes shows "high cur" the first time I test, then I quickly
turn off tune, then try again and I see a normal reading
40m always too low
30m same as 80m
20m OK
17m OK
15m OK
12m too low
10m too low

I checked continuity on all my toroids, from "pad to pad" instead of
lead to lead, as far as I was able. Everything seems to check out
there. I'm not sure I can do anything else without a signal generator,
but I'm open to suggestions.

--
[hidden email] (Owen B. Mehegan)
'This above all: to thine own self be true,
And it must follow, as the night the day,
Thou canst not then be false to any man.'
--William Shakespeare

On Oct 5, 2010, at 7:01 AM, Don Wilhelm wrote:

> Owen,
>
> I am going to assume that you were successful with the Alignmant and
> Test Part II and were able to receive 40 meter signals. If not, you
> need to backtrack a bit and get receive working first.
>
> The easiest way to troubleshoot a "lack of power output" problem is
> to do Transmit Signal Tracing. You can use an RF Probe or an
> oscilloscope with a 10X probe.
>
> Turn the manual to the Appendix E section titled Transmit Signal
> Tracing and begin there. If using an RF Probe, the correct expected
> values are listed, but if using an oscilloscope, multiply those
> values by 2.8 and observe the peak to peak voltage of the RF waveform.
>
> An important note - the K2 will normally increase drive in an
> attempt to increase the power output if the voltage out of the RF
> detector is too small. What that means is that all RF voltages
> prior to the failing stage the RF voltages will be higher than the
> listed expected values. You will be investigating the transmit
> chain in the order listed (do not skip around), searching for the
> *first* stage where the output is lower than expected - that is the
> failing stage which then needs to be analyzed to determine why it is
> not operating properly. Measurements beyond this first failure
> point are not relevant to that first problem to be found.
> All RF Voltage measurements in the list can be done from the top of
> the RF Board. Do not transmit with the heat sink removed. If the
> PA transistors do get a bit of drive, will produce output and if the
> heat sink is not in place, they will quickly overheat and fry
> themselves.
>
> Normal failures are soldering problems, toroid leads or incorrectly
> placed components. Active device failures are way down of the list
> of failure suspects unless something has caused it to be abused.
>
> 73,
> Don W3FPR
>
> On 10/5/2010 2:38 AM, Owen B. Mehegan wrote:
>> Hello -
>>
>> Two nights ago I completed the assembly of my K2's RF board, and
>> after
>> a visual inspection verification of resistance checks, I assembled
>> the
>> radio per the instructions and began the third phase of alignment and
>> test. I've discovered a problem right off the bat, though. In the 40
>> meter transmitter alignment section, when I put the K2 into tune mode
>> at 2 watts power, I'm only reading .3-.4 watts on the onboard
>> wattmeter (it alternates between the two). This is with a connection
>> to a 100W dummy load. Varying L1 and L2 has no effect on the power
>> reading.
>>
>> I discovered one mistake: I misread the instructions for winding T4's
>> 5-6 and 7-8 windings. I wound these bare wires through BOTH holes in
>> the core, in a U shape, thus connecting 6-7 and 5-8 (and very
>> possibly
>> all 4 together with a short in the middle - this is what comes of
>> staying up too late working on a project!). I corrected this mistake
>> tonight, but I'm seeing no change on the measured power. Is it
>> possible that this mistake fried something? I never saw any smoke or
>> smelled burned silicon, and the thermal pads on Q7 and Q8 look
>> normal.
>>
>> I have also noticed that if I vary the power adjustment on the front
>> panel while in tune mode, I see no change on the wattmeter. That
>> seems
>> significant.
>>
>> These are the other checks I've done so far, per the troubleshooting
>> instructions:
>>
>

Don Wilhelm-4

Re: K2: Debugging low transmit power

Owen,

You should not need a signal generator for Transmit Signal Tracing - I
believe you are in the Receive Signal Tracing area - skip that and go to
the Transmit Signal Tracing. The K2 itself is the signal generator for
this TX signal tracing.

Try increasing the CAL CUR setting to 3.50 Amps rather than the manual
suggested setting of 2.50 amps.
Often the 80 meter output will be well above the 15 watt level, and the
current needed for that higher power output will be in excess of the
2.50 Amp setting.

If the Hi-Cur messages persist, then look to the Low Pass filter for the
answer to your problems. Count the turns on the toroids carefully. A
problem with Hi-Cur coupled with low power output is normally an
indication of a problem with either T4 or the Low Pass filter.

73,
Don W3FPR

On 10/6/2010 1:42 AM, Owen B. Mehegan wrote:

> Thanks again for the suggestions Don. Tonight I started doing some
> signal tracing, and my RF probe seems to work fine with my DMM.
> However, I ran up against a different obstacle: I don't have a signal
> generator handy, and the LPF/BPF/T-R Switch tests seem to require one.
> I can get access to one but it won't be until tomorrow at least.
>
> I did find one other data point for my symptoms. On SOME bands, I get
> the expected power out when I put the K2 into tune mode. Here are the
> results of my tests:
>
> 80m sometimes shows "high cur" the first time I test, then I quickly
> turn off tune, then try again and I see a normal reading
> 40m always too low
> 30m same as 80m
> 20m OK
> 17m OK
> 15m OK
> 12m too low
> 10m too low
>
> I checked continuity on all my toroids, from "pad to pad" instead of
> lead to lead, as far as I was able. Everything seems to check out
> there. I'm not sure I can do anything else without a signal generator,
> but I'm open to suggestions.
>
> --
> [hidden email] (Owen B. Mehegan)
> 'This above all: to thine own self be true,
> And it must follow, as the night the day,
> Thou canst not then be false to any man.'
> --William Shakespeare
>
> On Oct 5, 2010, at 7:01 AM, Don Wilhelm wrote:
>
>> Owen,
>>
>> I am going to assume that you were successful with the Alignmant and
>> Test Part II and were able to receive 40 meter signals. If not, you
>> need to backtrack a bit and get receive working first.
>>
>> The easiest way to troubleshoot a "lack of power output" problem is
>> to do Transmit Signal Tracing. You can use an RF Probe or an
>> oscilloscope with a 10X probe.
>>
>> Turn the manual to the Appendix E section titled Transmit Signal
>> Tracing and begin there. If using an RF Probe, the correct expected
>> values are listed, but if using an oscilloscope, multiply those
>> values by 2.8 and observe the peak to peak voltage of the RF waveform.
>>
>> An important note - the K2 will normally increase drive in an
>> attempt to increase the power output if the voltage out of the RF
>> detector is too small. What that means is that all RF voltages
>> prior to the failing stage the RF voltages will be higher than the
>> listed expected values. You will be investigating the transmit
>> chain in the order listed (do not skip around), searching for the
>> *first* stage where the output is lower than expected - that is the
>> failing stage which then needs to be analyzed to determine why it is
>> not operating properly. Measurements beyond this first failure
>> point are not relevant to that first problem to be found.
>> All RF Voltage measurements in the list can be done from the top of
>> the RF Board. Do not transmit with the heat sink removed. If the
>> PA transistors do get a bit of drive, will produce output and if the
>> heat sink is not in place, they will quickly overheat and fry
>> themselves.
>>
>> Normal failures are soldering problems, toroid leads or incorrectly
>> placed components. Active device failures are way down of the list
>> of failure suspects unless something has caused it to be abused.
>>
>> 73,
>> Don W3FPR
>>
>> On 10/5/2010 2:38 AM, Owen B. Mehegan wrote:
>>> Hello -
>>>
>>> Two nights ago I completed the assembly of my K2's RF board, and
>>> after
>>> a visual inspection verification of resistance checks, I assembled
>>> the
>>> radio per the instructions and began the third phase of alignment and
>>> test. I've discovered a problem right off the bat, though. In the 40
>>> meter transmitter alignment section, when I put the K2 into tune mode
>>> at 2 watts power, I'm only reading .3-.4 watts on the onboard
>>> wattmeter (it alternates between the two). This is with a connection
>>> to a 100W dummy load. Varying L1 and L2 has no effect on the power
>>> reading.
>>>
>>> I discovered one mistake: I misread the instructions for winding T4's
>>> 5-6 and 7-8 windings. I wound these bare wires through BOTH holes in
>>> the core, in a U shape, thus connecting 6-7 and 5-8 (and very
>>> possibly
>>> all 4 together with a short in the middle - this is what comes of
>>> staying up too late working on a project!). I corrected this mistake
>>> tonight, but I'm seeing no change on the measured power. Is it
>>> possible that this mistake fried something? I never saw any smoke or
>>> smelled burned silicon, and the thermal pads on Q7 and Q8 look
>>> normal.
>>>
>>> I have also noticed that if I vary the power adjustment on the front
>>> panel while in tune mode, I see no change on the wattmeter. That
>>> seems
>>> significant.
>>>
>>> These are the other checks I've done so far, per the troubleshooting
>>> instructions:
>>>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>

omehegan

Re: K2: Debugging low transmit power

Hi Don -

OK, I have gone through the transmit signal tracing steps and found a
couple of anomalies. I'm referencing the specific section name and
step number in the troubleshooting here.

ALC:
3. Pin 2 of U8 (on the control board) shows 4.97VDC instead of the
expected .7 - 2.5VDC. The manual tells me to check the component
values in the RF detector, which I have. They are all correct.

Transmit Mixer, Buffer, etc
2. Pin 4 of U10 shows .228 Vrms instead of the expected .016.
3. Pin 6 of U9 shows 1.148 Vrms instead of .2.
4. W6 shows 0.
5. D1 shows 0.

So those seem _totally_ out of whack. These checks were done on 40m,
which is where I'm having problems. Now what?

--
[hidden email] (Owen B. Mehegan)
'Nothing fixes a thing so intensely in the memory as the wish to
forget it.'
--Michel de Montaigne

On Oct 5, 2010, at 11:33 PM, Don Wilhelm wrote:

> Owen,
>
> You should not need a signal generator for Transmit Signal Tracing -
> I believe you are in the Receive Signal Tracing area - skip that and
> go to the Transmit Signal Tracing. The K2 itself is the signal
> generator for this TX signal tracing.
>
> Try increasing the CAL CUR setting to 3.50 Amps rather than the
> manual suggested setting of 2.50 amps.
> Often the 80 meter output will be well above the 15 watt level, and
> the current needed for that higher power output will be in excess of
> the 2.50 Amp setting.
>
> If the Hi-Cur messages persist, then look to the Low Pass filter for
> the answer to your problems. Count the turns on the toroids
> carefully. A problem with Hi-Cur coupled with low power output is
> normally an indication of a problem with either T4 or the Low Pass
> filter.
>
> 73,
> Don W3FPR
>
> On 10/6/2010 1:42 AM, Owen B. Mehegan wrote:
>> Thanks again for the suggestions Don. Tonight I started doing some
>> signal tracing, and my RF probe seems to work fine with my DMM.
>> However, I ran up against a different obstacle: I don't have a signal
>> generator handy, and the LPF/BPF/T-R Switch tests seem to require
>> one.
>> I can get access to one but it won't be until tomorrow at least.
>>
>> I did find one other data point for my symptoms. On SOME bands, I get
>> the expected power out when I put the K2 into tune mode. Here are the
>> results of my tests:
>>
>> 80m sometimes shows "high cur" the first time I test, then I quickly
>> turn off tune, then try again and I see a normal reading
>> 40m always too low
>> 30m same as 80m
>> 20m OK
>> 17m OK
>> 15m OK
>> 12m too low
>> 10m too low
>>
>> I checked continuity on all my toroids, from "pad to pad" instead of
>> lead to lead, as far as I was able. Everything seems to check out
>> there. I'm not sure I can do anything else without a signal
>> generator,
>> but I'm open to suggestions.
>>
>> --
>> [hidden email] (Owen B. Mehegan)
>> 'This above all: to thine own self be true,
>> And it must follow, as the night the day,
>> Thou canst not then be false to any man.'
>> --William Shakespeare
>>
>> On Oct 5, 2010, at 7:01 AM, Don Wilhelm wrote:
>>
>>> Owen,
>>>
>>> I am going to assume that you were successful with the Alignmant and
>>> Test Part II and were able to receive 40 meter signals. If not, you
>>> need to backtrack a bit and get receive working first.
>>>
>>> The easiest way to troubleshoot a "lack of power output" problem is
>>> to do Transmit Signal Tracing. You can use an RF Probe or an
>>> oscilloscope with a 10X probe.
>>>
>>> Turn the manual to the Appendix E section titled Transmit Signal
>>> Tracing and begin there. If using an RF Probe, the correct expected
>>> values are listed, but if using an oscilloscope, multiply those
>>> values by 2.8 and observe the peak to peak voltage of the RF
>>> waveform.
>>>
>>> An important note - the K2 will normally increase drive in an
>>> attempt to increase the power output if the voltage out of the RF
>>> detector is too small. What that means is that all RF voltages
>>> prior to the failing stage the RF voltages will be higher than the
>>> listed expected values. You will be investigating the transmit
>>> chain in the order listed (do not skip around), searching for the
>>> *first* stage where the output is lower than expected - that is the
>>> failing stage which then needs to be analyzed to determine why it is
>>> not operating properly. Measurements beyond this first failure
>>> point are not relevant to that first problem to be found.
>>> All RF Voltage measurements in the list can be done from the top of
>>> the RF Board. Do not transmit with the heat sink removed. If the
>>> PA transistors do get a bit of drive, will produce output and if the
>>> heat sink is not in place, they will quickly overheat and fry
>>> themselves.
>>>
>>> Normal failures are soldering problems, toroid leads or incorrectly
>>> placed components. Active device failures are way down of the list
>>> of failure suspects unless something has caused it to be abused.
>>>
>>> 73,
>>> Don W3FPR
>>>
>>> On 10/5/2010 2:38 AM, Owen B. Mehegan wrote:
>>>> Hello -
>>>>
>>>> Two nights ago I completed the assembly of my K2's RF board, and
>>>> after
>>>> a visual inspection verification of resistance checks, I assembled
>>>> the
>>>> radio per the instructions and began the third phase of alignment
>>>> and
>>>> test. I've discovered a problem right off the bat, though. In the
>>>> 40
>>>> meter transmitter alignment section, when I put the K2 into tune
>>>> mode
>>>> at 2 watts power, I'm only reading .3-.4 watts on the onboard
>>>> wattmeter (it alternates between the two). This is with a
>>>> connection
>>>> to a 100W dummy load. Varying L1 and L2 has no effect on the power
>>>> reading.
>>>>
>>>> I discovered one mistake: I misread the instructions for winding
>>>> T4's
>>>> 5-6 and 7-8 windings. I wound these bare wires through BOTH holes
>>>> in
>>>> the core, in a U shape, thus connecting 6-7 and 5-8 (and very
>>>> possibly
>>>> all 4 together with a short in the middle - this is what comes of
>>>> staying up too late working on a project!). I corrected this
>>>> mistake
>>>> tonight, but I'm seeing no change on the measured power. Is it
>>>> possible that this mistake fried something? I never saw any smoke
>>>> or
>>>> smelled burned silicon, and the thermal pads on Q7 and Q8 look
>>>> normal.
>>>>
>>>> I have also noticed that if I vary the power adjustment on the
>>>> front
>>>> panel while in tune mode, I see no change on the wattmeter. That
>>>> seems
>>>> significant.
>>>>
>>>> These are the other checks I've done so far, per the
>>>> troubleshooting
>>>> instructions:
>>>>
>> ______________________________________________________________
>> Elecraft mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:[hidden email]
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>>
>

Don Wilhelm-4

Re: K2: Debugging low transmit power

Owen,

I tried to caution you not to get excited about RF voltage values that
were higher than the listed expected values.
That is normal for a K2 that is not detecting any output - the K2
increases the drive in an effort to obtain more output. Since you have
no output, the RF voltages on the working stages will be at the maximum
value.
The listed "expected" values are within 10% of what is to be expected
for a properly working K2 at 5 watts.

The transmit buffer has output and there is none at W6. Look at the
schematic so you can identify the components between those two points.
Basically it is the D6/D7 switch and the bandpass filter.

Is there any chance you put C6 in the wrong holes? In the set of 3
holes at the J15 location, the center hole should be empty. C6 goes in
the holes for pins 1 and 3.

73,
Don W3FPR

On 10/6/2010 11:42 AM, Owen B. Mehegan wrote:

> Hi Don -
>
> OK, I have gone through the transmit signal tracing steps and found a
> couple of anomalies. I'm referencing the specific section name and
> step number in the troubleshooting here.
>
> ALC:
> 3. Pin 2 of U8 (on the control board) shows 4.97VDC instead of the
> expected .7 - 2.5VDC. The manual tells me to check the component
> values in the RF detector, which I have. They are all correct.
>
> Transmit Mixer, Buffer, etc
> 2. Pin 4 of U10 shows .228 Vrms instead of the expected .016.
> 3. Pin 6 of U9 shows 1.148 Vrms instead of .2.
> 4. W6 shows 0.
> 5. D1 shows 0.
>
> So those seem _totally_ out of whack. These checks were done on 40m,
> which is where I'm having problems. Now what?
>

omehegan

Re: K2: Debugging low transmit power

Ah, so you did, Don. My mistake for not taking heed, I would have
pressed on with the signal tracing. However, while sitting here at work
and pondering the problem, I just re-read your message below for the 3rd
or 4th time, and something clicked: AHA! I _did_ put C6 in the wrong
holes!! When I realized you were specifically referring to a strip of 3
holes it jogged my memory. There's even an errata about this, which I
dutifully added to the manual, but it still confused me. The errata
says:

Add to the end of the step, "Use the outer of the three holes, not the
two holes closest to the left edge of the pc board."

Somehow I just interpreted that as "use the two holes closest to the
right edge." Since I'm at work at the K2 is at home, I thought I would
have to wait to confirm my mistake, but I remembered that I sent some
friends a picture of the board when I finished it. I just pulled it up
in my email, and I can clearly see that C6 is in the wrong holes!

I'll correct this tonight and see what that gets me. Thanks Don!

On Wed, 06 Oct 2010 13:29:17 -0400, Don Wilhelm <[hidden email]>
wrote:

> Owen,
>
> I tried to caution you not to get excited about RF voltage values
> that were higher than the listed expected values.
> That is normal for a K2 that is not detecting any output - the K2
> increases the drive in an effort to obtain more output. Since you
> have no output, the RF voltages on the working stages will be at the
> maximum value.
> The listed "expected" values are within 10% of what is to be expected
> for a properly working K2 at 5 watts.
>
> The transmit buffer has output and there is none at W6. Look at the
> schematic so you can identify the components between those two points.
> Basically it is the D6/D7 switch and the bandpass filter.
>
> Is there any chance you put C6 in the wrong holes? In the set of 3
> holes at the J15 location, the center hole should be empty. C6 goes
> in the holes for pins 1 and 3.
>
> 73,
> Don W3FPR

--
Owen B. Mehegan ([hidden email])
"He is a dangerous mixture of sophistication and recklessness which
makes one anxious about his influence on other boys."
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

Don Wilhelm-4

Re: K2: Debugging low transmit power

Owen,

I will have to take some of the blame about the wording in the manual
errata sheet. I had though specifying the outer holes of the set of
three was sufficient - but I was apparently thinking only of a "set" in
mathematical language terms (I am certain any mathematician would select
the right holes from that statement), so Ron got approval and made the
errata item that reads like yours.
After hearing about several builders who still were having trouble, we
decided to change the errata again to state explicitly that the middle
hole of the 3 should be empty (or some wording to that effect). We hope
this wording clears up the confusion once and for all.

73,
Don W3FPR

On 10/6/2010 5:31 PM, Owen B. Mehegan wrote:

> Ah, so you did, Don. My mistake for not taking heed, I would have
> pressed on with the signal tracing. However, while sitting here at work
> and pondering the problem, I just re-read your message below for the 3rd
> or 4th time, and something clicked: AHA! I _did_ put C6 in the wrong
> holes!! When I realized you were specifically referring to a strip of 3
> holes it jogged my memory. There's even an errata about this, which I
> dutifully added to the manual, but it still confused me. The errata
> says:
>
> Add to the end of the step, "Use the outer of the three holes, not the
> two holes closest to the left edge of the pc board."
>
> Somehow I just interpreted that as "use the two holes closest to the
> right edge." Since I'm at work at the K2 is at home, I thought I would
> have to wait to confirm my mistake, but I remembered that I sent some
> friends a picture of the board when I finished it. I just pulled it up
> in my email, and I can clearly see that C6 is in the wrong holes!
>
> I'll correct this tonight and see what that gets me. Thanks Don!
>

omehegan

Re: K2: Debugging low transmit power

Well that did it! I verified that I put C6 in the wrong two holes,
corrected it, and now I'm getting good power out on 40m! Thanks Don!
I'm still seeing only .3-.4 watts on 12m and 10m, but I realize I'm
getting ahead of myself. Should I just proceed through the alignment
until I get to 12m and 10m, or should I continue looking for another
fault?

--
[hidden email] (Owen B. Mehegan)
'This above all: to thine own self be true,
And it must follow, as the night the day,
Thou canst not then be false to any man.'
--William Shakespeare

On Oct 6, 2010, at 5:05 PM, Don Wilhelm wrote:

> Owen,
>
> I will have to take some of the blame about the wording in the
> manual errata sheet. I had though specifying the outer holes of the
> set of three was sufficient - but I was apparently thinking only of
> a "set" in mathematical language terms (I am certain any
> mathematician would select the right holes from that statement), so
> Ron got approval and made the errata item that reads like yours.
> After hearing about several builders who still were having trouble,
> we decided to change the errata again to state explicitly that the
> middle hole of the 3 should be empty (or some wording to that
> effect). We hope this wording clears up the confusion once and for
> all.
>
> 73,
> Don W3FPR
>
> On 10/6/2010 5:31 PM, Owen B. Mehegan wrote:
>> Ah, so you did, Don. My mistake for not taking heed, I would have
>> pressed on with the signal tracing. However, while sitting here at
>> work
>> and pondering the problem, I just re-read your message below for
>> the 3rd
>> or 4th time, and something clicked: AHA! I _did_ put C6 in the wrong
>> holes!! When I realized you were specifically referring to a strip
>> of 3
>> holes it jogged my memory. There's even an errata about this, which I
>> dutifully added to the manual, but it still confused me. The errata
>> says:
>>
>> Add to the end of the step, "Use the outer of the three holes, not
>> the
>> two holes closest to the left edge of the pc board."
>>
>> Somehow I just interpreted that as "use the two holes closest to the
>> right edge." Since I'm at work at the K2 is at home, I thought I
>> would
>> have to wait to confirm my mistake, but I remembered that I sent some
>> friends a picture of the board when I finished it. I just pulled it
>> up
>> in my email, and I can clearly see that C6 is in the wrong holes!
>>
>> I'll correct this tonight and see what that gets me. Thanks Don!
>>
>

AC7AC

Re: K2: Debugging low transmit power

In reply to this post by Don Wilhelm-4

CONTENTS DELETED

The author has deleted this message.

Don Wilhelm-4

Re: K2: Debugging low transmit power

In reply to this post by omehegan

Owen,

Go through the transmit alignment for the bandpass filters before
declaring another problem.
Most K2s will have some power reduction on 12 and 10 meters, but if your
10 meter power output is greater than 10 watts it is OK.

73,
Don W3FPR

On 10/6/2010 10:16 PM, Owen B. Mehegan wrote:
> Well that did it! I verified that I put C6 in the wrong two holes,
> corrected it, and now I'm getting good power out on 40m! Thanks Don!
> I'm still seeing only .3-.4 watts on 12m and 10m, but I realize I'm
> getting ahead of myself. Should I just proceed through the alignment
> until I get to 12m and 10m, or should I continue looking for another
> fault?
>
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

omehegan

Re: K2: Debugging low transmit power

Last night I aligned all the bands, first on receive and then on
transmit. On 10m, when doing the receiver pre-alignment, I couldn't seem
to detect any audible change in signal level as I adjusted L12 and L13,
which worried me. But when I did the transmitter alignments I had no
trouble getting 2 watts on 10. I just had to tune L12 almost all the way
into its can. I don't think I checked to see if I was getting 10 watts
out on all the bands when I set the K2 to that level, so I'll do that
tonight, but it looks like I'm all set! Thanks for all your help Don,
and I'm happy to hear that the errata will be updated to clarify this. I
thought, "A diagram would really solve this problem," so I'm glad to
hear that's exactly what will be added. My friend said, "This is what
makes the K2 and Elecraft so great. They are constantly updating the kit
and the manual. You are building on the shoulders of a thousand
frustrated hams!"

I do have one curiosity question and one comment about this procedure.
The question: First I did the receiver pre-alignment, and then I went on
to the transmitter alignment. Both sections include these sentences:
"Switch to 20 meters (14100 kHz). Set C21 and C23 to their mid-points."
Then they have you adjust L8 and L9 on 20, then on 30, and then they
have you go back to 20 and adjust C21 and C23. My question is, should
the _transmitter_ alignment steps say to set C21 and C23 to their
mid-points? This would seem to undo the adjustments to them that you
make in the receiver pre-alignment section. Is this a mistake in the
manual or am I misunderstanding the procedure?

The comment: during the transmitter alignment procedure, I was confused
by the fact that I was setting the K2's power to 2 watts, and then
adjusting things in tune mode that would have the rig showing as much as
10-12 watts on its wattmeter. I didn't understand if I was supposed to
be just adjusting to see exactly 2 watts, or to see the maximum power.
And when I adjusted for maximum power, exited tune, then entered tune
again, I would see closer to 2 watts. A quick Google search turned up
this recent email thread:
http://www.mail-archive.com/elecraft@.../msg102763.html Of
course it had a helpful response from Don Wilhelm, so I felt better and
went on my way :) I would suggest that this information somehow be
codified in the manual for future builders, because to me it wasn't
intuitive that I was setting the K2 for 2 watts, and then seeing it hit
10+ as I went through the alignment steps.

On Wed, 06 Oct 2010 23:41:09 -0400, Don Wilhelm <[hidden email]>
wrote:
> Owen,
>
> Go through the transmit alignment for the bandpass filters before
> declaring another problem.
> Most K2s will have some power reduction on 12 and 10 meters, but if
> your 10 meter power output is greater than 10 watts it is OK.
>
> 73,
> Don W3FPR

--
Owen B. Mehegan ([hidden email])
"He is a dangerous mixture of sophistication and recklessness which
makes one anxious about his influence on other boys."
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

Don Wilhelm-4

Re: K2: Debugging low transmit power

Owen,

That bit of "funny" procedure for setting L8 and L9 first on 20 meters
(while the capacitors are set to the mid-point) is an interim step that
only needs to be done once (unless drastic changes are made to the
30/20m bandpass filter). It assures that the L8 and L9 are set to
somewhere close to where they should tune on 10 meters. Having done
that once, it is not necessary to repeat it - thereafter align L8 and L9
only for 30 meters, and after that, align the capacitors for 20 meters.
It is easier to discern the peak using transmit - you are using a meter
instead of your ears. Doing the alignment in receive first assures that
you adjust the settings somewhere close and you should not have to
adjust very far to get the transmit alignment correct.

As a matter of fact - in all cases, you should adjust the inductors for
any band pair before adjusting the capacitors.
In other words, you can align any of the band pairs independently of the
other bands.
The pairs are 40 and 60 meters (align 40 first), 30 and 20 (align 30
first), 17 and 15 (align 15 first), and 12 and 10m (align 10 first). I
did not mention 80 and 160 yet because they must share a single setting
if the k160RX is installed. The way I do that is to first align 80
(with the frequency set to 3750), then go to 1850 kHz - I look at (and
remember) the slug position (which has been preset for 80m), then move
the slug to find the peak on 160, noticing which way I turn the slug -
it will not be far away, at most a turn - then I turn the slug back
halfway between those two points and call it done.

Yes, it is important to remember that you are trying to tune for a peak
(and not to 2 watts). In fact, the manual tells you to initially set
the power control to 2 watts, so you should already have about 2 watts
of output before you start tuning, and the power will go up as you come
closer to the peak. Do whatever you have to do to keep the power lower
than about 5 watts because you can miss the peak when the power goes up
- exit and re-enter TUNE or rotate the power knob a bit and the power
will come down.

Actually the initial 2 watts is not important either, I routinely use
anywhere from 1 to 2 watts when doing an alignment, and that gives me a
bit more 'headroom' when the power increases. My arbitrary limit of 5
watts is not critical either, but if it exceeds that, I back down to a
low power again and continue to search for the peak.

73,
Don W3FPR

On 10/7/2010 3:38 PM, Owen B. Mehegan wrote:

> Last night I aligned all the bands, first on receive and then on
> transmit. On 10m, when doing the receiver pre-alignment, I couldn't seem
> to detect any audible change in signal level as I adjusted L12 and L13,
> which worried me. But when I did the transmitter alignments I had no
> trouble getting 2 watts on 10. I just had to tune L12 almost all the way
> into its can. I don't think I checked to see if I was getting 10 watts
> out on all the bands when I set the K2 to that level, so I'll do that
> tonight, but it looks like I'm all set! Thanks for all your help Don,
> and I'm happy to hear that the errata will be updated to clarify this. I
> thought, "A diagram would really solve this problem," so I'm glad to
> hear that's exactly what will be added. My friend said, "This is what
> makes the K2 and Elecraft so great. They are constantly updating the kit
> and the manual. You are building on the shoulders of a thousand
> frustrated hams!"
>
> I do have one curiosity question and one comment about this procedure.
> The question: First I did the receiver pre-alignment, and then I went on
> to the transmitter alignment. Both sections include these sentences:
> "Switch to 20 meters (14100 kHz). Set C21 and C23 to their mid-points."
> Then they have you adjust L8 and L9 on 20, then on 30, and then they
> have you go back to 20 and adjust C21 and C23. My question is, should
> the _transmitter_ alignment steps say to set C21 and C23 to their
> mid-points? This would seem to undo the adjustments to them that you
> make in the receiver pre-alignment section. Is this a mistake in the
> manual or am I misunderstanding the procedure?
>
> The comment: during the transmitter alignment procedure, I was confused
> by the fact that I was setting the K2's power to 2 watts, and then
> adjusting things in tune mode that would have the rig showing as much as
> 10-12 watts on its wattmeter. I didn't understand if I was supposed to
> be just adjusting to see exactly 2 watts, or to see the maximum power.
> And when I adjusted for maximum power, exited tune, then entered tune
> again, I would see closer to 2 watts. A quick Google search turned up
> this recent email thread:
> http://www.mail-archive.com/elecraft@.../msg102763.html Of
> course it had a helpful response from Don Wilhelm, so I felt better and
> went on my way :) I would suggest that this information somehow be
> codified in the manual for future builders, because to me it wasn't
> intuitive that I was setting the K2 for 2 watts, and then seeing it hit
> 10+ as I went through the alignment steps.
>