Elecraft

K2/100 - Power investigation

Classic

List

Threaded

12 messages Options

dave.wilburn

K2/100 - Power investigation

After trying unsuccessfully to get email to take data in tabs, I ended
up just putting the information on my website. Sorry for the trouble.

I was playing with an assortment of power meters, and noticed my K2
was reading lower than I expected. The closes things I had to a
gauge, was an IC-703+ that had just been to the shop, and repaired,
thus I felt it was likely to be putting out 10w.

When I compared a Bird 43 with an element I picked up on line, and one
that came with it, it read a bit low. But when I used an MFJ949e, it
was right on. Gotta run with what I have.

So I checked out the K2/100 at what it said was 10w/53w/111w (Full
Power). And I got the results, that are displayed here;

http://www.k4dgw.com/k2.html

The 10w settings seem a little bit more dispersed than I would have
thought. The 10m, 12m & 20m 53w settings seem a bit low. Then at
what the rig says is 111w/Full Power, the rig seems even lower.

Do I need to go back through and run the power calibration, or should
I start troubleshooting?

I have a KAT100, that was powered down, and the power was taken out of
the back of the K2/100.
--

Dave Wilburn
K4DGW
K2/100 - S/N 5982
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

Don Wilhelm-4

Re: K2/100 - Power investigation

Dave,

I cannot really give you a quick positive answer - the real answer is
"it all depends". To give a positive answer would require that you
operate into a dummy load which is 50 ohms non-reactive at all
frequencies in question. One can measure the dummy load with an antenna
analyzer - expect to see 50 + j0 at all frequencies of interest for
reliable and repeatable results.

I do not generally trust any wattmeter (including an uncalibrated Bird)
to provide any greater than 20% of full scale accuracy. If the full
scale reading is 100 watts, then the error can be as great as 20 watts
at any point on its scale. A properly calibrated KPA100 wattmeter can
do better than that.

I find that with a good 50 ohm dummy load and an 100 MHz oscilloscope
with a 10X probe (probe rated for 100 MHz), I can calibrate the KPA100
wattmeter to less than 5% accuracy on bands below 20 meters - at higher
frequencies, the 'scope and probe rolloff will indicate lower than
expected results. But I can state that once calibrated, the KPA100
power calibration can be trusted to greater accuracy than the
run-of-the-mill wattmeters.

Using a good 50 ohm non-reactive dummy load and measuring the RF voltage
across it:
At 4 watts, the Peak to Peak RF voltage will be 40 volts.
At 10 watts the p-p RF voltage will be 63.25 volts
At 40 watts, the p-p RF voltage will be 126.5 volts
At 100 watts, the p-p RF voltage will be 200 volts.

My dummy load is a heat-sinked Caddock MP9100-50.0-1% resistor rated at
100 watts with zero length leads that has been measured flat at 50 ohms
up to 60 MHz.

Up to the frequency limits of my oscilloscope, I trust my dummy load and
the peak to peak voltage measurement to provide more accurate power
measurement than any wattmeter available. I find that, after
calibrating with these tools, the typical KPA100 power indications are
within 5% of the actual power output.

I can also say that the TelePost LP-200 (and I assume the LP100) will
also produce power readings which are within 5% of the actual power
after proper calibration. I cannot make such statements about the
accuracy of the more common wattmeters which spec their accuracy as a
percentage of the full scale reading. A Bird with a 100 watt slug that
has been recently calibrated will have an accuracy of 5% of full scale,
which is a 5 watt error anywhere on the scale - so a recently calibrated
Bird with a 100 watt slug will indicate anywhere between 5 watts and 15
watts for an actual power of 10 watts. One can do better with a
precision 50 ohm load and a 100 MHz oscilloscope.

OK, that is my wattmeter rant for this month!!!

73,
Don W3FPR

David Wilburn wrote:

> After trying unsuccessfully to get email to take data in tabs, I ended
> up just putting the information on my website. Sorry for the trouble.
>
> I was playing with an assortment of power meters, and noticed my K2
> was reading lower than I expected. The closes things I had to a
> gauge, was an IC-703+ that had just been to the shop, and repaired,
> thus I felt it was likely to be putting out 10w.
>
> When I compared a Bird 43 with an element I picked up on line, and one
> that came with it, it read a bit low. But when I used an MFJ949e, it
> was right on. Gotta run with what I have.
>
> So I checked out the K2/100 at what it said was 10w/53w/111w (Full
> Power). And I got the results, that are displayed here;
>
> http://www.k4dgw.com/k2.html
>
> The 10w settings seem a little bit more dispersed than I would have
> thought. The 10m, 12m & 20m 53w settings seem a bit low. Then at
> what the rig says is 111w/Full Power, the rig seems even lower.
>
> Do I need to go back through and run the power calibration, or should
> I start troubleshooting?
>
> I have a KAT100, that was powered down, and the power was taken out of
> the back of the K2/100.

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

dave.wilburn

Re: K2/100 - Power investigation

Understand. I did pick up the DL1, and check the readings. They were
along the lines of what I experienced with the MFJ. But I could not
read it at a higher power, as it is only a 20w dummy load. But it
jibed with what I was seeing. I will get further readings with it and
a voltmeter.

Dave Wilburn
K4DGW
K2/100 - S/N 5982

Don Wilhelm wrote:

> Dave,
>
> I cannot really give you a quick positive answer - the real answer is
> "it all depends". To give a positive answer would require that you
> operate into a dummy load which is 50 ohms non-reactive at all
> frequencies in question. One can measure the dummy load with an antenna
> analyzer - expect to see 50 + j0 at all frequencies of interest for
> reliable and repeatable results.
>
> I do not generally trust any wattmeter (including an uncalibrated Bird)
> to provide any greater than 20% of full scale accuracy. If the full
> scale reading is 100 watts, then the error can be as great as 20 watts
> at any point on its scale. A properly calibrated KPA100 wattmeter can
> do better than that.
>
> I find that with a good 50 ohm dummy load and an 100 MHz oscilloscope
> with a 10X probe (probe rated for 100 MHz), I can calibrate the KPA100
> wattmeter to less than 5% accuracy on bands below 20 meters - at higher
> frequencies, the 'scope and probe rolloff will indicate lower than
> expected results. But I can state that once calibrated, the KPA100
> power calibration can be trusted to greater accuracy than the
> run-of-the-mill wattmeters.
>
> Using a good 50 ohm non-reactive dummy load and measuring the RF voltage
> across it:
> At 4 watts, the Peak to Peak RF voltage will be 40 volts.
> At 10 watts the p-p RF voltage will be 63.25 volts
> At 40 watts, the p-p RF voltage will be 126.5 volts
> At 100 watts, the p-p RF voltage will be 200 volts.
>
> My dummy load is a heat-sinked Caddock MP9100-50.0-1% resistor rated at
> 100 watts with zero length leads that has been measured flat at 50 ohms
> up to 60 MHz.
>
> Up to the frequency limits of my oscilloscope, I trust my dummy load and
> the peak to peak voltage measurement to provide more accurate power
> measurement than any wattmeter available. I find that, after
> calibrating with these tools, the typical KPA100 power indications are
> within 5% of the actual power output.
>
> I can also say that the TelePost LP-200 (and I assume the LP100) will
> also produce power readings which are within 5% of the actual power
> after proper calibration. I cannot make such statements about the
> accuracy of the more common wattmeters which spec their accuracy as a
> percentage of the full scale reading. A Bird with a 100 watt slug that
> has been recently calibrated will have an accuracy of 5% of full scale,
> which is a 5 watt error anywhere on the scale - so a recently calibrated
> Bird with a 100 watt slug will indicate anywhere between 5 watts and 15
> watts for an actual power of 10 watts. One can do better with a
> precision 50 ohm load and a 100 MHz oscilloscope.
>
> OK, that is my wattmeter rant for this month!!!
>
> 73,
> Don W3FPR
>
> David Wilburn wrote:
>> After trying unsuccessfully to get email to take data in tabs, I ended
>> up just putting the information on my website. Sorry for the trouble.
>>
>> I was playing with an assortment of power meters, and noticed my K2
>> was reading lower than I expected. The closes things I had to a
>> gauge, was an IC-703+ that had just been to the shop, and repaired,
>> thus I felt it was likely to be putting out 10w.
>>
>> When I compared a Bird 43 with an element I picked up on line, and one
>> that came with it, it read a bit low. But when I used an MFJ949e, it
>> was right on. Gotta run with what I have.
>>
>> So I checked out the K2/100 at what it said was 10w/53w/111w (Full
>> Power). And I got the results, that are displayed here;
>>
>> http://www.k4dgw.com/k2.html
>>
>> The 10w settings seem a little bit more dispersed than I would have
>> thought. The 10m, 12m & 20m 53w settings seem a bit low. Then at
>> what the rig says is 111w/Full Power, the rig seems even lower.
>>
>> Do I need to go back through and run the power calibration, or should
>> I start troubleshooting?
>>
>> I have a KAT100, that was powered down, and the power was taken out of
>> the back of the K2/100.
>

Vic K2VCO

Re: K2/100 - Power investigation

In reply to this post by Don Wilhelm-4

Don Wilhelm wrote:

> I find that with a good 50 ohm dummy load and an 100 MHz oscilloscope
> with a 10X probe (probe rated for 100 MHz), I can calibrate the KPA100
> wattmeter to less than 5% accuracy on bands below 20 meters - at higher
> frequencies, the 'scope and probe rolloff will indicate lower than
> expected results.

Be careful if you use this method! I learned the hard way (destroyed
THREE probes) that the voltage rating for scope probes goes down as
frequency goes up. There is a little piece of paper that comes with the
probes; look at it, compute the voltage across your dummy load at the
power levels that you will be testing (100 watts is about 70.7 volts),
and make sure that the probe can handle it at the frequency of interest.
--
73,
Vic, K2VCO
Fresno CA
http://www.qsl.net/k2vco
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

Vic K2VCO

Re: K2/100 - Power investigation

Vic K2VCO wrote:
>
> Be careful if you use this method! I learned the hard way (destroyed
> THREE probes) that the voltage rating for scope probes goes down as
> frequency goes up. There is a little piece of paper that comes with the
> probes; look at it, compute the voltage across your dummy load at the
> power levels that you will be testing (100 watts is about 70.7 volts),
> and make sure that the probe can handle it at the frequency of interest.

I should have said 70.7 v RMS; the peak voltage will be 1.414 x 70.71 =
100v, which means that the peak-to-peak voltage will be 200v.
--
73,
Vic, K2VCO
Fresno CA
http://www.qsl.net/k2vco
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

Tom Hammond-2

Re: K2/100 - Power investigation

In reply to this post by dave.wilburn

Don Wilhelm wrote:

>I find that with a good 50 ohm dummy load and an 100 MHz
>oscilloscope with a 10X probe (probe rated for 100 MHz), I can
>calibrate the KPA100 wattmeter to less than 5% accuracy on bands
>below 20 meters - at higher frequencies, the 'scope and probe
>rolloff will indicate lower than expected results.

Surely you meant to type "BETTER than 5% accuracy..."... <G>

I catch myself doing the same damned thing here.

You going to FDIM? QRP ARCI rooms are now being reserved...!

Tom

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

dave.wilburn

Re: K2/100 - Power investigation

In reply to this post by dave.wilburn

With the DL1, when you do the calculation to get the power level, what
do you measure with? V-AC or V-DC?

Dave Wilburn
K4DGW
K2/100 - S/N 5982

David Wilburn wrote:

> After trying unsuccessfully to get email to take data in tabs, I ended
> up just putting the information on my website. Sorry for the trouble.
>
> I was playing with an assortment of power meters, and noticed my K2 was
> reading lower than I expected. The closes things I had to a gauge, was
> an IC-703+ that had just been to the shop, and repaired, thus I felt it
> was likely to be putting out 10w.
>
> When I compared a Bird 43 with an element I picked up on line, and one
> that came with it, it read a bit low. But when I used an MFJ949e, it
> was right on. Gotta run with what I have.
>
> So I checked out the K2/100 at what it said was 10w/53w/111w (Full
> Power). And I got the results, that are displayed here;
>
> http://www.k4dgw.com/k2.html
>
> The 10w settings seem a little bit more dispersed than I would have
> thought. The 10m, 12m & 20m 53w settings seem a bit low. Then at what
> the rig says is 111w/Full Power, the rig seems even lower.
>
> Do I need to go back through and run the power calibration, or should I
> start troubleshooting?
>
> I have a KAT100, that was powered down, and the power was taken out of
> the back of the K2/100.

Don Wilhelm-4

Re: K2/100 - Power investigation

In reply to this post by Vic K2VCO

The 10X probes I have are rated at either 300 volts or 600 volts, so no
problem.
Vic does bring up a very good point that you should know the ratings for
anything that you use in electrical measurements - not only to keep from
destroying it, but also to know whether the device will behave as you
expect it to. Sometimes what you see on the 'scope face is not really
what you are measuring. I know my 100 MHz rated probes create some
loading at 30 MHz, so I don't use those for measurements at that frequency.

I should have stated that 10X probes were definitely to be used, 1X
probes load too much.
At lower power levels, an RF Probe will provide very good results. The
DL1 is a good dummy load to use and it has a peak detecting RF Probe
attached across half of it.

I don't understand why some folks seem to continuously convert from peak
to RMS and back again - when using a 'scope, the display will show the
peak voltage - I use the derived formula for power as Vp-p^2/8R when
working with a 'scope - that is square the peak to peak voltage and
divide by 400 if the load is 50 ohms - I find it easier than doing all
the conversion formulas.

73,
Don W3FPR

Vic K2VCO wrote:

> Vic K2VCO wrote:
>>
>> Be careful if you use this method! I learned the hard way (destroyed
>> THREE probes) that the voltage rating for scope probes goes down as
>> frequency goes up. There is a little piece of paper that comes with
>> the probes; look at it, compute the voltage across your dummy load at
>> the power levels that you will be testing (100 watts is about 70.7
>> volts), and make sure that the probe can handle it at the frequency
>> of interest.
>
> I should have said 70.7 v RMS; the peak voltage will be 1.414 x 70.71
> = 100v, which means that the peak-to-peak voltage will be 200v.

Don Wilhelm-4

Re: K2/100 - Power investigation

In reply to this post by dave.wilburn

David,

The diode in the DL1 recitfies the RF and you read the DMM in DC volts.
The DL1 uses a peak reading detector across half the load, so the DMM
will read 1/2 of the zero to peak RF voltage. At low power levels (less
than 1 or 2 watts), you should add the diode drop to the indicated
voltage before calculation of the power, but at higher power levels the
diode drop is not significant.

BTW: The power formula in the DL1 manual has a small error that shows up
when calculatiing low powers. The proper formula can be simplified to P
= 2x (Vdmm + 0.15)^2/50 - that is: add 0.15 to the DMM reading, square
it and divide by 25 - the 0.15 term is an assumption for the diode drop,
those who would like more precision can measure their particular diode
forward voltage drop.

73,
Don W3FPR

David Wilburn wrote:
> With the DL1, when you do the calculation to get the power level, what
> do you measure with? V-AC or V-DC?
>
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

dave.wilburn

Re: K2/100 - Power investigation

Ok, I got all kinds of numbers off of the DL1. The Bird seems a touch
low, showing 80 - 100w across all bands. Every time I measure it a
different way, I get different numbers. Very frustrating. The W1
quit after about 5 min. The MFJ meter gives one reading, the Bird
another, using the DL1 and doing a calculation, goes crazy on 10&12m
but more stable, and once again different on others. Friend is going
to loan me his Bird, think it is a 4309, and we will see what it
shows. I have put the Bird between the K2 & the KAT100, and
transmitted and then ID'ed, and got the 80 - 100w. Am I chasing my
tail?

Dave Wilburn
K4DGW
K2/100 - S/N 5982

Don Wilhelm wrote:

> David,
>
> The diode in the DL1 recitfies the RF and you read the DMM in DC volts.
> The DL1 uses a peak reading detector across half the load, so the DMM
> will read 1/2 of the zero to peak RF voltage. At low power levels (less
> than 1 or 2 watts), you should add the diode drop to the indicated
> voltage before calculation of the power, but at higher power levels the
> diode drop is not significant.
>
> BTW: The power formula in the DL1 manual has a small error that shows up
> when calculatiing low powers. The proper formula can be simplified to P
> = 2x (Vdmm + 0.15)^2/50 - that is: add 0.15 to the DMM reading, square
> it and divide by 25 - the 0.15 term is an assumption for the diode drop,
> those who would like more precision can measure their particular diode
> forward voltage drop.
>
> 73,
> Don W3FPR
>
> David Wilburn wrote:
>> With the DL1, when you do the calculation to get the power level, what
>> do you measure with? V-AC or V-DC?
>>
>

Don Wilhelm-4

Re: K2/100 - Power investigation

David,

You are chasing your tail since you are trying to use all those
wattmeters - they will all be different.
You say that you are using a wattmeter between the KPA100 and the KAT100
- that in itself can lead to a variety of readings. The only thing you
can really trust is a good dummy load. The DL1 is good.

Because of the scaling done in the KPA100 for the output of the
wattmeter to the K2, you can calibrate it at low power. Disconnect the
power cable from the KPA100 and power only the base K2. Now, you can
use your DL1 dummy load and do the calibration at 10 watts (the power
control is a lot easier to use with no PS connected to the KPA100). Be
certain your KPA100 wattmeter is balanced first (do that into a good 50
ohm dummy load or it will not be correct - use your DL1). Then set the
KPA100 forward power pot so the K2 display matches the power calculated
from the DL1/DMM reading. You may have to make several trials to
average things out, but once calibrated that way, your KPA100 wattmeter
will have greater accuracy than the run of the mill wattmeters - and it
will be accurate on both the low and high power scales.

If you do it right, you can judge the accuracy of those external
wattmeters by comparing them to your KPA100.

You can expect some variation from band to band because the diodes are
not entirely linear with frequency. Pick one band and stick with it
while doing the calibration. Attempting to calibrate on several bands
is an exercise in frustration. I like to use 40 or 30 meters, but if
your operation is mostly on 20 meters, then that may be a better choice
for you.

If you are attempting to compare wattmeters with a live antenna,
remember that common mode current on the feedline can also have some
influence, and the amount of influence may be vastly different from one
meter to another - as I said, the only thing you can trust during these
calibrations is a known dummy load.

73,
Don W3FPR

David Wilburn wrote:

> Ok, I got all kinds of numbers off of the DL1. The Bird seems a touch
> low, showing 80 - 100w across all bands. Every time I measure it a
> different way, I get different numbers. Very frustrating. The W1
> quit after about 5 min. The MFJ meter gives one reading, the Bird
> another, using the DL1 and doing a calculation, goes crazy on 10&12m
> but more stable, and once again different on others. Friend is going
> to loan me his Bird, think it is a 4309, and we will see what it
> shows. I have put the Bird between the K2 & the KAT100, and
> transmitted and then ID'ed, and got the 80 - 100w. Am I chasing my tail?
>
>
> Dave Wilburn
> K4DGW
> K2/100 - S/N 5982
>
>
>

dave.wilburn

Re: K2/100 - Power investigation

Yea, not having a good day. Switching dummie loads and powers and
such. Smoked the DL1, small dummy load, large power. Hobbies are
supposed to be fun....right? ;)

Dave Wilburn
K4DGW
K2/100 - S/N 5982

Don Wilhelm wrote:

> David,
>
> You are chasing your tail since you are trying to use all those
> wattmeters - they will all be different.
> You say that you are using a wattmeter between the KPA100 and the KAT100
> - that in itself can lead to a variety of readings. The only thing you
> can really trust is a good dummy load. The DL1 is good.
>
> Because of the scaling done in the KPA100 for the output of the
> wattmeter to the K2, you can calibrate it at low power. Disconnect the
> power cable from the KPA100 and power only the base K2. Now, you can
> use your DL1 dummy load and do the calibration at 10 watts (the power
> control is a lot easier to use with no PS connected to the KPA100). Be
> certain your KPA100 wattmeter is balanced first (do that into a good 50
> ohm dummy load or it will not be correct - use your DL1). Then set the
> KPA100 forward power pot so the K2 display matches the power calculated
> from the DL1/DMM reading. You may have to make several trials to
> average things out, but once calibrated that way, your KPA100 wattmeter
> will have greater accuracy than the run of the mill wattmeters - and it
> will be accurate on both the low and high power scales.
>
> If you do it right, you can judge the accuracy of those external
> wattmeters by comparing them to your KPA100.
>
> You can expect some variation from band to band because the diodes are
> not entirely linear with frequency. Pick one band and stick with it
> while doing the calibration. Attempting to calibrate on several bands
> is an exercise in frustration. I like to use 40 or 30 meters, but if
> your operation is mostly on 20 meters, then that may be a better choice
> for you.
>
> If you are attempting to compare wattmeters with a live antenna,
> remember that common mode current on the feedline can also have some
> influence, and the amount of influence may be vastly different from one
> meter to another - as I said, the only thing you can trust during these
> calibrations is a known dummy load.
>
> 73,
> Don W3FPR
>
> David Wilburn wrote:
>> Ok, I got all kinds of numbers off of the DL1. The Bird seems a touch
>> low, showing 80 - 100w across all bands. Every time I measure it a
>> different way, I get different numbers. Very frustrating. The W1
>> quit after about 5 min. The MFJ meter gives one reading, the Bird
>> another, using the DL1 and doing a calculation, goes crazy on 10&12m
>> but more stable, and once again different on others. Friend is going
>> to loan me his Bird, think it is a 4309, and we will see what it
>> shows. I have put the Bird between the K2 & the KAT100, and
>> transmitted and then ID'ed, and got the 80 - 100w. Am I chasing my tail?
>>
>>
>> Dave Wilburn
>> K4DGW
>> K2/100 - S/N 5982
>>
>>
>>
>