Elecraft › [K2]

Erratic transmit power -- ALC oscillating?

Classic

List

Threaded

6 messages Options

n7zf

Erratic transmit power -- ALC oscillating?

This is a new build of a rev A K2 (installed rev B mods), rework eliminators, no options installed, several other mods installed including the CW waveform keying mod (VALC circuit). Receiver is working well.

If I set the output power to 5 watts, the usually K2 reports the power as 4.2 watts or 6.3 watts -- sometimes 5.1 watts. I am seeing some unexpected things in the ALC circuit on the control board and am wondering if the ALC is causing the transmit power to fluctuate.

In both transmit (cw) and receive mode, on all bands, there is a 45 hertz 0.7 volt spike coming out of U10A (LMC660), causing Q8 to turn on for about 3.5 msecs every 22 msecs. The B-E junction of Q8 is limiting the spike voltage. See photo of signal on pin 1 of U10 -- same as base of Q8. Using a 10x probe, so vertical scale is 200 mV per division, horizontal scale is 5 msec per division.

At the junction of R22 (82K) and R21 (270K) I am seeing a 1volt peak-to-peak signal being fed back by C46 (.01). Is this expected? I see this 'ripple' continue as a 150 mV peak at Test Point 2, and also on the output of the Xmit Mixer U10-pin4 during transmit. I have removed the control board and verified all the correct components against the Rev. H schematic. I have traced all the connections in that circuit. I have rechecked against the keying mod instructions. The CW key waveform modifications appear to be working correctly but is the ALC circuit supposed to be oscillating?

The CW keying mod (rev A 2/3/2004) specifies C31 to be .01 uF. The Rev A-to-B upgrade (rev E 4/29/2003) specifies C31 to be .047 uF. I assumed that the CW keying mod was a more robust change and followed it (so C31 is .01 uF).

Same problem running off a power supply or a battery. I have installed new power amp finals and new power amp bias transistors. I have carefully inspected T1-T4 in the final, including removing T3 and rewinding it. I have tested the low pass filters with a network analyzer and found no problems. I have measured and inspected the RF Output Detector circuit (and have the 1% resistor at R68 which measures to be 225.0 ohms).

I am grasping at straws at this point and would appreciate troubleshooting suggestions.
Bob N7ZF

Don Wilhelm-4

Re: Erratic transmit power -- ALC oscillating?

Bob,

The spikes you are seeing in the ALC circuit are normal, and are not
associated with any problem you may be having.

Is the variation in power observed when you press the TUNE button? If
so, that is normal too. The level is not controlled as tightly in TUNE
as it is with normal keying as a help to those who would be using a
manual tuner with the K2.

You apparently have an oscilloscope - have you observed the keying
waveform by connecting the 'scope across a 50 ohm dummy load? Does the
waveform have a shape like that shown in the Keying Waveform Mod
instructions? If so, that is OK.

Does the power output oscillate with normal keying? The LED bargraph or
your 'scope across a dummy load should reveal the instantaneous RF
level, and it should not vary. If it does, there is an application note
that will tell you how to determine the optimum value of RF Board R98
for your particular K2.

Yes, you should have used the capacitor from the Keying Waveshape mod at
CB C31.

Are you having problems with HiCur messages and low output power? I am
confused by your statements about the PA and LPF - there was no problem
with that part explicitly stated - any problem there is not related to
the rest of your post.

73,
Don W3FPR

On 2/10/2011 4:00 PM, n7zf wrote:

> This is a new build of a rev A K2 (installed rev B mods), rework eliminators,
> no options installed, several other mods installed including the CW waveform
> keying mod (VALC circuit). Receiver is working well.
>
> If I set the output power to 5 watts, the usually K2 reports the power as
> 4.2 watts or 6.3 watts -- sometimes 5.1 watts. I am seeing some unexpected
> things in the ALC circuit on the control board and am wondering if the ALC
> is causing the transmit power to fluctuate.
>
> In both transmit (cw) and receive mode, on all bands, there is a 45 hertz
> 0.7 volt spike coming out of U10A (LMC660), causing Q8 to turn on for about
> 3.5 msecs every 22 msecs. The B-E junction of Q8 is limiting the spike
> voltage. See photo of signal on pin 1 of U10 -- same as base of Q8. Using
> a 10x probe, so vertical scale is 200 mV per division, horizontal scale is 5
> msec per division.
>
> At the junction of R22 (82K) and R21 (270K) I am seeing a 1volt peak-to-peak
> signal being fed back by C46 (.01). Is this expected? I see this 'ripple'
> continue as a 150 mV peak at Test Point 2, and also on the output of the
> Xmit Mixer U10-pin4 during transmit. I have removed the control board and
> verified all the correct components against the Rev. H schematic. I have
> traced all the connections in that circuit. I have rechecked against the
> keying mod instructions. The CW key waveform modifications appear to be
> working correctly but is the ALC circuit supposed to be oscillating?
>
> The CW keying mod (rev A 2/3/2004) specifies C31 to be .01 uF. The Rev
> A-to-B upgrade (rev E 4/29/2003) specifies C31 to be .047 uF. I assumed
> that the CW keying mod was a more robust change and followed it (so C31 is
> .01 uF).
>
> Same problem running off a power supply or a battery. I have installed new
> power amp finals and new power amp bias transistors. I have carefully
> inspected T1-T4 in the final, including removing T3 and rewinding it. I
> have tested the low pass filters with a network analyzer and found no
> problems. I have measured and inspected the RF Output Detector circuit (and
> have the 1% resistor at R68 which measures to be 225.0 ohms).
>
>

______________________________________________________________
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: Erratic transmit power -- ALC oscillating?

In reply to this post by n7zf

Bob,

One thing that I failed to add, and might be a concern to you at first.
It is normal for the K2 to take 2 to 3 dit times to come up to the full
requested power after a change in the power knob setting, a change in
bands or a power on.

73,
Don W3FPR

On 2/10/2011 4:00 PM, n7zf wrote:

n7zf

Re: Erratic transmit power -- ALC oscillating?

In reply to this post by Don Wilhelm-4

Don,

Thank you for your suggestions. It sure is hard to debug circuits that have feedback loops.

I was using the TUNE button for the readings that I reported, and the radio was connected to a 50-ohm resistive dummy load. I did not know that the TUNE was an approximate power level.

The keying waveform matches the diagram in the mod instructions.

I have retested using my cw key. I think I am only seeing the 'erratic' power on 80m and 40m. Output on the other bands seesm to be consistently the same. At '5 watts' on 80m and 40m I would occasionally get around 4.2 watts and usually get around 6.4 watts.

The power will drift for about 3 to 5 seconds then stabilize. You mentioned that it should take about 3 dits for the power to stabilize, so it seems to take much too long on my K2. Having reinstalled the power amp finals this week, I do not believe that there is an issue with heatsinking or mounting of the power transistors causing the drifting.

I looked at the app note regarding R98 and made the change.
http://www.elecraft.com/Apps/Power_Control_Mod.html

With the potentiometer at 1000 ohms, I got 12.5 watts on 10m so I put in a 1K resistor. Then I retested at the 5 watt level. On 40m, for example, I now (usually) get 5.6 watts out instead of 6.x watts. But about 10% of the time I get about 4.3 watts out.

The reason that I changed the transistors and T3 in the power amp and tested the LPF is because this K2 draws more current than it should when transmitting. At 5 watts, the current draw is 2.04 to 2.18 amps on 80m through 15m. On 10 and 12m the current is 2.52 to 2.56 amps. The voltage drops from 13.6v to 13.1v on all bands (at 5watts). At 10 watts, the current is 2.50 to 2.75 amps.

Power amp transformer T4 was wound with the standard 3-turn winding.

Should I be concerned about the higher than normal current draw?

73,
Bob - N7ZF

Don Wilhelm-4

Re: Erratic transmit power -- ALC oscillating?

Bob,

I would not be concerned about the slightly higher current draw. Change
CAL CUR to 3.50 and you should have no problems. If you add the KPA100
you will need to increase the setting to 3.50 amps anyway, so do it
now. Despite the manual recommendations, many current K2s require more
than 2.50 amperes under some specific circumstances.

BTW - the CAL CUR setting can actually be used as a Low Battery Alarm.
The K2 tries to maintain the requested power level, and when the voltage
drops, that means more current must be drawn for any given power level,
so if you are operating on batteries, you can set the CAL CUR value to a
lower level to give you an indication that the battery voltage has
dropped to a level that requires increased current.

The RF gain will vary from K2 to K2, and since you can obtain greater
than 12 watts on 10 meters, that says your K2 has greater than average
RF Gain - and also says you did a meticulous job of construction -
congratulations.

73,
Don W3FPR

On 2/10/2011 10:51 PM, n7zf wrote:

> Don,
>
> Thank you for your suggestions. It sure is hard to debug circuits that have
> feedback loops.
>
> I was using the TUNE button for the readings that I reported, and the radio
> was connected to a 50-ohm resistive dummy load. I did not know that the
> TUNE was an approximate power level.
>
> The keying waveform matches the diagram in the mod instructions.
>
> I have retested using my cw key. I think I am only seeing the 'erratic'
> power on 80m and 40m. Output on the other bands seesm to be consistently
> the same. At '5 watts' on 80m and 40m I would occasionally get around 4.2
> watts and usually get around 6.4 watts.
>
> The power will drift for about 3 to 5 seconds then stabilize. You mentioned
> that it should take about 3 dits for the power to stabilize, so it seems to
> take much too long on my K2. Having reinstalled the power amp finals this
> week, I do not believe that there is an issue with heatsinking or mounting
> of the power transistors causing the drifting.
>
> I looked at the app note regarding R98 and made the change.
> http://www.elecraft.com/Apps/Power_Control_Mod.html
>
> With the potentiometer at 1000 ohms, I got 12.5 watts on 10m so I put in a
> 1K resistor. Then I retested at the 5 watt level. On 40m, for example, I
> now (usually) get 5.6 watts out instead of 6.x watts. But about 10% of the
> time I get about 4.3 watts out.
>
> The reason that I changed the transistors and T3 in the power amp and tested
> the LPF is because this K2 draws more current than it should when
> transmitting. At 5 watts, the current draw is 2.04 to 2.18 amps on 80m
> through 15m. On 10 and 12m the current is 2.52 to 2.56 amps. The voltage
> drops from 13.6v to 13.1v on all bands (at 5watts). At 10 watts, the
> current is 2.50 to 2.75 amps.
>
> Power amp transformer T4 was wound with the standard 3-turn winding.
>
> Should I be concerned about the higher than normal current draw?
>
> 73,
> Bob - N7ZF

n7zf

Re: Erratic transmit power -- ALC oscillating?

Okay, Don. I will continue then with installing the options.
73,
Bob - N7ZF