Elecraft

Incidental AM using FSK-D

Classic

List

Threaded

12 messages Options

Bill Lewis

Incidental AM using FSK-D

I am looking for information on how to eliminate incidental AM when using
FSK-D.

My K3 increases power by 10% when in Space as opposed to Mark.

Thanks much, Bill, W8NN

______________________________________________________________
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
Message delivered to [hidden email]

Joe Subich, W4TV-4

Re: Incidental AM using FSK-D

> My K3 increases power by 10% when in Space as opposed to Mark.

Two things here ... 10% is abut 0.4 dB. How did you measure the power
difference and did you do so with an instrument with enough resolution
to accurately measure that difference at the power level you were using?
For example a Bird Wattmeter can't possibly measure with that level of
accuracy nor is one likely to be able to read an analog meter that
closely.

I have measured my K3 on RTTY with a wattmeter traceable to NIST and
specified for 3% accuracy and have done so at four specific power
levels (25, 55, 80 and 105 Watts) to eliminate any ALC non-linearity
or PA compression from consideration. The specific Mark/Space power
levels were measured by placing the rig into transmit in FSK_D mode
with the FSK input closed (Mark), measuring the power output for a
given power setting, then opening the FSK input and noting the PO
(and change).

In all cases, the difference between MARK and SPACE was less than 2%
(less than the meter's specified accuracy) and at two power levels (55
and 80 W) the measured difference was less than 0.2W (<0.2 dB at 55W).

These measurements were made with standard "High tones" (2125/2295 Hz)
and the K3 was equipped with the optional 2.8 KHz, 8 pole IF filter.

73,

... Joe, W4TV

On 2014-11-10 8:42 PM, Bill Lewis wrote:

> I am looking for information on how to eliminate incidental AM when using
> FSK-D.
>
> My K3 increases power by 10% when in Space as opposed to Mark.
>
> Thanks much, Bill, W8NN
>
>
>
>
> ______________________________________________________________
> 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
> Message delivered to [hidden email]
>

______________________________________________________________
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
Message delivered to [hidden email]

Hank Garretson

Re: Incidental AM using FSK-D

Easy. Put a decent scope on the RF envelope.

73,

Hank, W6SX

On Monday, November 10, 2014, Joe Subich, W4TV <[hidden email]> wrote:

>
> > My K3 increases power by 10% when in Space as opposed to Mark.
>
> Two things here ... 10% is abut 0.4 dB. How did you measure the power
> difference and did you do so with an instrument with enough resolution
> to accurately measure that difference at the power level you were using?
> For example a Bird Wattmeter can't possibly measure with that level of
> accuracy nor is one likely to be able to read an analog meter that
> closely.
>
> I have measured my K3 on RTTY with a wattmeter traceable to NIST and
> specified for 3% accuracy and have done so at four specific power
> levels (25, 55, 80 and 105 Watts) to eliminate any ALC non-linearity
> or PA compression from consideration. The specific Mark/Space power
> levels were measured by placing the rig into transmit in FSK_D mode
> with the FSK input closed (Mark), measuring the power output for a
> given power setting, then opening the FSK input and noting the PO
> (and change).
>
> In all cases, the difference between MARK and SPACE was less than 2%
> (less than the meter's specified accuracy) and at two power levels (55
> and 80 W) the measured difference was less than 0.2W (<0.2 dB at 55W).
>
> These measurements were made with standard "High tones" (2125/2295 Hz)
> and the K3 was equipped with the optional 2.8 KHz, 8 pole IF filter.
>
> 73,
>
> ... Joe, W4TV
>
>
> On 2014-11-10 8:42 PM, Bill Lewis wrote:
>
>> I am looking for information on how to eliminate incidental AM when using
>> FSK-D.
>>
>> My K3 increases power by 10% when in Space as opposed to Mark.
>>
>> Thanks much, Bill, W8NN
>>
>>
>>
>>
>> ______________________________________________________________
>> 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
>> Message delivered to [hidden email]
>>
>> ______________________________________________________________
> 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
> Message delivered to [hidden email]
>

Don Wilhelm-4

Re: Incidental AM using FSK-D

Not so easy even with a decent scope.
50 watts will produce 141.42 volts p-p, and 55 watts will produce 148.32
vp-p.
That is a difference on 7.1 vp-p for a 10% difference in power.
That small difference is difficult to discern on a scope at that
vertical amplitude.
Yes, a small difference in amplitude can be discerned, but to quantify
it to less than a 5% difference is difficult to measure with a 'scope.

73,
Don W3FPR

On 11/10/2014 9:23 PM, Hank Garretson wrote:

> Easy. Put a decent scope on the RF envelope.
>
> 73,
>
> Hank, W6SX
>
>
>
> On Monday, November 10, 2014, Joe Subich, W4TV <[hidden email]> wrote:
>
>>> My K3 increases power by 10% when in Space as opposed to Mark.
>> Two things here ... 10% is abut 0.4 dB. How did you measure the power
>> difference and did you do so with an instrument with enough resolution
>> to accurately measure that difference at the power level you were using?
>> For example a Bird Wattmeter can't possibly measure with that level of
>> accuracy nor is one likely to be able to read an analog meter that
>> closely.
>>
>> I have measured my K3 on RTTY with a wattmeter traceable to NIST and
>> specified for 3% accuracy and have done so at four specific power
>> levels (25, 55, 80 and 105 Watts) to eliminate any ALC non-linearity
>> or PA compression from consideration. The specific Mark/Space power
>> levels were measured by placing the rig into transmit in FSK_D mode
>> with the FSK input closed (Mark), measuring the power output for a
>> given power setting, then opening the FSK input and noting the PO
>> (and change).
>>
>> In all cases, the difference between MARK and SPACE was less than 2%
>> (less than the meter's specified accuracy) and at two power levels (55
>> and 80 W) the measured difference was less than 0.2W (<0.2 dB at 55W).
>>
>> These measurements were made with standard "High tones" (2125/2295 Hz)
>> and the K3 was equipped with the optional 2.8 KHz, 8 pole IF filter.
>>
>>

Myron WVØH

Re: Incidental AM using FSK-D

Don et al,

I would try the Elecraft DL1 and the power turned down on the rig to about 10 Watts. The DL1 is the most accurate RF power measuring device under $30 and competes with RF power meters costing much more. Couple it to an accurate DVM and you have a very decent power measuring setup. Of course it's absorptive and not in line like a Bird but hey, it's a start.

If you put the scope probe across the 25 Ω tap where the diode is connected and left the DL1 output open, you could get an accurate "sample" of the RF envelope across 25 Ω.

Myron WVØH
Printed on Recycled Data

> On Nov 10, 2014, at 7:38 PM, Don Wilhelm <[hidden email]> wrote:
>
> Not so easy even with a decent scope.
> 50 watts will produce 141.42 volts p-p, and 55 watts will produce 148.32 vp-p.
> That is a difference on 7.1 vp-p for a 10% difference in power.
> That small difference is difficult to discern on a scope at that vertical amplitude.
> Yes, a small difference in amplitude can be discerned, but to quantify it to less than a 5% difference is difficult to measure with a 'scope.
>
> 73,
> Don W3FPR
>
>> On 11/10/2014 9:23 PM, Hank Garretson wrote:
>> Easy. Put a decent scope on the RF envelope.
>>
>> 73,
>>
>> Hank, W6SX
>>
>>
>>
>> On Monday, November 10, 2014, Joe Subich, W4TV <[hidden email]> wrote:
>>
>>>> My K3 increases power by 10% when in Space as opposed to Mark.
>>> Two things here ... 10% is abut 0.4 dB. How did you measure the power
>>> difference and did you do so with an instrument with enough resolution
>>> to accurately measure that difference at the power level you were using?
>>> For example a Bird Wattmeter can't possibly measure with that level of
>>> accuracy nor is one likely to be able to read an analog meter that
>>> closely.
>>>
>>> I have measured my K3 on RTTY with a wattmeter traceable to NIST and
>>> specified for 3% accuracy and have done so at four specific power
>>> levels (25, 55, 80 and 105 Watts) to eliminate any ALC non-linearity
>>> or PA compression from consideration. The specific Mark/Space power
>>> levels were measured by placing the rig into transmit in FSK_D mode
>>> with the FSK input closed (Mark), measuring the power output for a
>>> given power setting, then opening the FSK input and noting the PO
>>> (and change).
>>>
>>> In all cases, the difference between MARK and SPACE was less than 2%
>>> (less than the meter's specified accuracy) and at two power levels (55
>>> and 80 W) the measured difference was less than 0.2W (<0.2 dB at 55W).
>>>
>>> These measurements were made with standard "High tones" (2125/2295 Hz)
>>> and the K3 was equipped with the optional 2.8 KHz, 8 pole IF filter.
>>>
>>>
>
> ______________________________________________________________
> 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
> Message delivered to [hidden email]

Hank Garretson

Re: Incidental AM using FSK-D

In reply to this post by Don Wilhelm-4

Keep in mind that this is one case where instrument precisipn is more
important than instrument accuracy.

73,

Hank, W6SX

On Monday, November 10, 2014, Don Wilhelm <[hidden email]> wrote:

> Not so easy even with a decent scope.
> 50 watts will produce 141.42 volts p-p, and 55 watts will produce 148.32
> vp-p.
> That is a difference on 7.1 vp-p for a 10% difference in power.
> That small difference is difficult to discern on a scope at that vertical
> amplitude.
> Yes, a small difference in amplitude can be discerned, but to quantify it
> to less than a 5% difference is difficult to measure with a 'scope.
>
> 73,
> Don W3FPR
>
> On 11/10/2014 9:23 PM, Hank Garretson wrote:
>
>> Easy. Put a decent scope on the RF envelope.
>>
>> 73,
>>
>> Hank, W6SX
>>
>>
>>
>> On Monday, November 10, 2014, Joe Subich, W4TV <[hidden email]> wrote:
>>
>> My K3 increases power by 10% when in Space as opposed to Mark.
>>>>
>>> Two things here ... 10% is abut 0.4 dB. How did you measure the power
>>> difference and did you do so with an instrument with enough resolution
>>> to accurately measure that difference at the power level you were using?
>>> For example a Bird Wattmeter can't possibly measure with that level of
>>> accuracy nor is one likely to be able to read an analog meter that
>>> closely.
>>>
>>> I have measured my K3 on RTTY with a wattmeter traceable to NIST and
>>> specified for 3% accuracy and have done so at four specific power
>>> levels (25, 55, 80 and 105 Watts) to eliminate any ALC non-linearity
>>> or PA compression from consideration. The specific Mark/Space power
>>> levels were measured by placing the rig into transmit in FSK_D mode
>>> with the FSK input closed (Mark), measuring the power output for a
>>> given power setting, then opening the FSK input and noting the PO
>>> (and change).
>>>
>>> In all cases, the difference between MARK and SPACE was less than 2%
>>> (less than the meter's specified accuracy) and at two power levels (55
>>> and 80 W) the measured difference was less than 0.2W (<0.2 dB at 55W).
>>>
>>> These measurements were made with standard "High tones" (2125/2295 Hz)
>>> and the K3 was equipped with the optional 2.8 KHz, 8 pole IF filter.
>>>
>>>
>>>
> ______________________________________________________________
> 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
> Message delivered to [hidden email]
>

Myron WVØH

Re: Incidental AM using FSK-D

Precisely.

Myron WVØH
Printed on Recycled Data

>
> precisipn
______________________________________________________________
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
Message delivered to [hidden email]

Joe Subich, W4TV-4

Re: Incidental AM using FSK-D

In reply to this post by Hank Garretson

First, you are quibbling about less than 0.5 dB *even at 10%*
difference in claimed power output. Second, my measurements were
taken with a watt meter capable of .01 W (10 mW) precision and the
worst case difference was less than 2% (0.08 dB) - not the 10%
claimed.

Dynamic (scope) measurements may show a slightly greater dynamic
difference - if you can read them with the precision necessary to
detect a 3% ripple on a 140V RF envelope - due to wave shaping for
bandwidth control. Even then, such minor (0.08 dB) differences in
mark/space power are insignificant with the majority due to ripple
in the IF filter. Both the 2.7 and 2.8 KHz filters show > 1dB
ripple in the passband - see:
http://www.elecraft.com/K3/K3_filter_plots.htm

73,

... Joe, W4TV

On 2014-11-10 9:52 PM, Hank Garretson wrote:

> Keep in mind that this is one case where instrument precisipn is more
> important than instrument accuracy.
>
> 73,
>
> Hank, W6SX
>
>
>
> On Monday, November 10, 2014, Don Wilhelm <[hidden email]> wrote:
>
>> Not so easy even with a decent scope.
>> 50 watts will produce 141.42 volts p-p, and 55 watts will produce 148.32
>> vp-p.
>> That is a difference on 7.1 vp-p for a 10% difference in power.
>> That small difference is difficult to discern on a scope at that vertical
>> amplitude.
>> Yes, a small difference in amplitude can be discerned, but to quantify it
>> to less than a 5% difference is difficult to measure with a 'scope.
>>
>> 73,
>> Don W3FPR
>>
>> On 11/10/2014 9:23 PM, Hank Garretson wrote:
>>
>>> Easy. Put a decent scope on the RF envelope.
>>>
>>> 73,
>>>
>>> Hank, W6SX
>>>
>>>
>>>
>>> On Monday, November 10, 2014, Joe Subich, W4TV <[hidden email]> wrote:
>>>
>>> My K3 increases power by 10% when in Space as opposed to Mark.
>>>>>
>>>> Two things here ... 10% is abut 0.4 dB. How did you measure the power
>>>> difference and did you do so with an instrument with enough resolution
>>>> to accurately measure that difference at the power level you were using?
>>>> For example a Bird Wattmeter can't possibly measure with that level of
>>>> accuracy nor is one likely to be able to read an analog meter that
>>>> closely.
>>>>
>>>> I have measured my K3 on RTTY with a wattmeter traceable to NIST and
>>>> specified for 3% accuracy and have done so at four specific power
>>>> levels (25, 55, 80 and 105 Watts) to eliminate any ALC non-linearity
>>>> or PA compression from consideration. The specific Mark/Space power
>>>> levels were measured by placing the rig into transmit in FSK_D mode
>>>> with the FSK input closed (Mark), measuring the power output for a
>>>> given power setting, then opening the FSK input and noting the PO
>>>> (and change).
>>>>
>>>> In all cases, the difference between MARK and SPACE was less than 2%
>>>> (less than the meter's specified accuracy) and at two power levels (55
>>>> and 80 W) the measured difference was less than 0.2W (<0.2 dB at 55W).
>>>>
>>>> These measurements were made with standard "High tones" (2125/2295 Hz)
>>>> and the K3 was equipped with the optional 2.8 KHz, 8 pole IF filter.
>>>>
>>>>
>>>>
>> ______________________________________________________________
>> 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
>> Message delivered to [hidden email]
>>
> ______________________________________________________________
> 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
> Message delivered to [hidden email]
>

alsopb

Re: Incidental AM using FSK-D

In reply to this post by Joe Subich, W4TV-4

And what if this was done this with the standard 2.7 KHz filter? Filter
passband ripple has the the stock answer to what people are observing.

Also what about dynamic vs static differences. Is it possible in the
dynamic situation the differences would be greater?

It does sound like there is some confusion precision with accuracy. A
measuring device might offer high precision and resolution but lower
absolute accuracy. People often make accurate difference measurements
with with a device which less accurate than the delta one observes. If
the difference are small, they can be quite accurate.

73 de Brian/K3KO

On 11/11/2014 02:13, Joe Subich, W4TV wrote:

>
> > My K3 increases power by 10% when in Space as opposed to Mark.
>
> Two things here ... 10% is abut 0.4 dB. How did you measure the power
> difference and did you do so with an instrument with enough resolution
> to accurately measure that difference at the power level you were using?
> For example a Bird Wattmeter can't possibly measure with that level of
> accuracy nor is one likely to be able to read an analog meter that
> closely.
>
> I have measured my K3 on RTTY with a wattmeter traceable to NIST and
> specified for 3% accuracy and have done so at four specific power
> levels (25, 55, 80 and 105 Watts) to eliminate any ALC non-linearity
> or PA compression from consideration. The specific Mark/Space power
> levels were measured by placing the rig into transmit in FSK_D mode
> with the FSK input closed (Mark), measuring the power output for a
> given power setting, then opening the FSK input and noting the PO
> (and change).
>
> In all cases, the difference between MARK and SPACE was less than 2%
> (less than the meter's specified accuracy) and at two power levels (55
> and 80 W) the measured difference was less than 0.2W (<0.2 dB at 55W).
>
> These measurements were made with standard "High tones" (2125/2295 Hz)
> and the K3 was equipped with the optional 2.8 KHz, 8 pole IF filter.
>
> 73,
>
> ... Joe, W4TV
>
>
> On 2014-11-10 8:42 PM, Bill Lewis wrote:
>> I am looking for information on how to eliminate incidental AM when using
>> FSK-D.
>>
>> My K3 increases power by 10% when in Space as opposed to Mark.
>>
>> Thanks much, Bill, W8NN
>>
>>
>>
>>
>> ______________________________________________________________
>> 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
>> Message delivered to [hidden email]
>>
> ______________________________________________________________
> 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
> Message delivered to [hidden email]
>
>
> -----
> No virus found in this message.
> Checked by AVG - www.avg.com
> Version: 2012.0.2247 / Virus Database: 4189/8045 - Release Date: 11/10/14
>
>

Phil Wheeler-2

Re: Incidental AM using FSK-D

As a practical matter "My K3 increases power by
10% when in Space as opposed to Mark": is this
really something to be concerned about?

Phil W7OX

On 11/10/14 10:51 PM, Brian wrote:

> And what if this was done this with the standard
> 2.7 KHz filter? Filter passband ripple has the
> the stock answer to what people are observing.
>
> Also what about dynamic vs static differences.
> Is it possible in the dynamic situation the
> differences would be greater?
>
> It does sound like there is some confusion
> precision with accuracy. A measuring device
> might offer high precision and resolution but
> lower absolute accuracy. People often make
> accurate difference measurements with with a
> device which less accurate than the delta one
> observes. If the difference are small, they can
> be quite accurate.
>
> 73 de Brian/K3KO
>
>
> On 11/11/2014 02:13, Joe Subich, W4TV wrote:
>>
>> > My K3 increases power by 10% when in Space
>> as opposed to Mark.
>>
>> Two things here ... 10% is abut 0.4 dB. How
>> did you measure the power
>> difference and did you do so with an instrument
>> with enough resolution
>> to accurately measure that difference at the
>> power level you were using?
>> For example a Bird Wattmeter can't possibly
>> measure with that level of
>> accuracy nor is one likely to be able to read
>> an analog meter that
>> closely.
>>
>> I have measured my K3 on RTTY with a wattmeter
>> traceable to NIST and
>> specified for 3% accuracy and have done so at
>> four specific power
>> levels (25, 55, 80 and 105 Watts) to eliminate
>> any ALC non-linearity
>> or PA compression from consideration. The
>> specific Mark/Space power
>> levels were measured by placing the rig into
>> transmit in FSK_D mode
>> with the FSK input closed (Mark), measuring the
>> power output for a
>> given power setting, then opening the FSK input
>> and noting the PO
>> (and change).
>>
>> In all cases, the difference between MARK and
>> SPACE was less than 2%
>> (less than the meter's specified accuracy) and
>> at two power levels (55
>> and 80 W) the measured difference was less than
>> 0.2W (<0.2 dB at 55W).
>>
>> These measurements were made with standard
>> "High tones" (2125/2295 Hz)
>> and the K3 was equipped with the optional 2.8
>> KHz, 8 pole IF filter.
>>
>> 73,
>>
>> ... Joe, W4TV
>>
>>
>> On 2014-11-10 8:42 PM, Bill Lewis wrote:
>>> I am looking for information on how to
>>> eliminate incidental AM when using
>>> FSK-D.
>>>
>>> My K3 increases power by 10% when in Space as
>>> opposed to Mark.
>>>
>>> Thanks much, Bill, W8NN

alsopb

Re: Incidental AM using FSK-D

Phil,

That's the right question. There is no doubt that turning a pure FSK
modulated signal into an FSK + AM modulated (QAM) signal increases
bandwidth. The question is how much and how far down are the additional
side bands. I'm sure there are guys here who can either calculate or
measure the effect. It would be interesting to know the answer.

As a practical matter, mother nature does this to almost all FSK signals
via propagation effects. Often received mark and space amplitudes
different by a LOT more than the amounts we're talking about here. It
is interesting to watch this. Given this mother nature bandwidth
smearing, it is doubtful that the real world impact of K3 QAM vs pure
FSK is significant.

73 de Brian/K3KO
On 11/11/2014 06:56, Phil Wheeler wrote:

> As a practical matter "My K3 increases power by 10% when in Space as
> opposed to Mark": is this really something to be concerned about?
>
> Phil W7OX
>
> On 11/10/14 10:51 PM, Brian wrote:
>> And what if this was done this with the standard 2.7 KHz filter?
>> Filter passband ripple has the the stock answer to what people are
>> observing.
>>
>> Also what about dynamic vs static differences. Is it possible in the
>> dynamic situation the differences would be greater?
>>
>> It does sound like there is some confusion precision with accuracy. A
>> measuring device might offer high precision and resolution but lower
>> absolute accuracy. People often make accurate difference measurements
>> with with a device which less accurate than the delta one observes.
>> If the difference are small, they can be quite accurate.
>>
>> 73 de Brian/K3KO
>>
>>
>> On 11/11/2014 02:13, Joe Subich, W4TV wrote:
>>>
>>> > My K3 increases power by 10% when in Space as opposed to Mark.
>>>
>>> Two things here ... 10% is abut 0.4 dB. How did you measure the power
>>> difference and did you do so with an instrument with enough resolution
>>> to accurately measure that difference at the power level you were using?
>>> For example a Bird Wattmeter can't possibly measure with that level of
>>> accuracy nor is one likely to be able to read an analog meter that
>>> closely.
>>>
>>> I have measured my K3 on RTTY with a wattmeter traceable to NIST and
>>> specified for 3% accuracy and have done so at four specific power
>>> levels (25, 55, 80 and 105 Watts) to eliminate any ALC non-linearity
>>> or PA compression from consideration. The specific Mark/Space power
>>> levels were measured by placing the rig into transmit in FSK_D mode
>>> with the FSK input closed (Mark), measuring the power output for a
>>> given power setting, then opening the FSK input and noting the PO
>>> (and change).
>>>
>>> In all cases, the difference between MARK and SPACE was less than 2%
>>> (less than the meter's specified accuracy) and at two power levels (55
>>> and 80 W) the measured difference was less than 0.2W (<0.2 dB at 55W).
>>>
>>> These measurements were made with standard "High tones" (2125/2295 Hz)
>>> and the K3 was equipped with the optional 2.8 KHz, 8 pole IF filter.
>>>
>>> 73,
>>>
>>> ... Joe, W4TV
>>>
>>>
>>> On 2014-11-10 8:42 PM, Bill Lewis wrote:
>>>> I am looking for information on how to eliminate incidental AM when
>>>> using
>>>> FSK-D.
>>>>
>>>> My K3 increases power by 10% when in Space as opposed to Mark.
>>>>
>>>> Thanks much, Bill, W8NN
>
> ______________________________________________________________
> 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
> Message delivered to [hidden email]
>
>
> -----
> No virus found in this message.
> Checked by AVG - www.avg.com
> Version: 2012.0.2247 / Virus Database: 4189/8045 - Release Date: 11/10/14
>
>

alsopb

Re: Incidental AM using FSK-D

I want to retract the conclusion. Don't do this kind of mental exercise
at 2AM.

There is a big difference in the "bit rate" between QSB and RTTY.

1)Those engaged in Frequency Measurement Tests typically observe carrier
frequency shifts <1 Hz over several minutes due to propagation effects.
Most of this is due to Doppler effect of moving reflecting layers.
Amplitude variations are due to multipath signal phase difference.
Anyhow the effective "bit rate" is low.

2)FSK with 45 baud amplitude modulation, one would expect a series of
sidebands. They would occur outside the usually occupied spectrum for a
series of RY's (essentially a square wave)-- maybe +/- 200 Hz and up.
It probably makes a significant difference in sideband amplitudes and
placements, whether the FSK transition is phase continuous or not. The
text content being sent affects amplitudes of the additional sidebands.

Maybe these conclusions made at 3:45 AM are more accurate.

Again it would be neat to compare the calculated bandwidths for the two
point QAM scheme (F1,amplitude 1), (F2, amplitude 2) to the ideal FSK
(F1, amplitude 1), F2 (amplitude 1) modulation schemes.

73 de Brian/K3KO

On 11/11/2014 08:03, Brian wrote:

> Phil,
>
> That's the right question. There is no doubt that turning a pure FSK
> modulated signal into an FSK + AM modulated (QAM) signal increases
> bandwidth. The question is how much and how far down are the additional
> side bands. I'm sure there are guys here who can either calculate or
> measure the effect. It would be interesting to know the answer.
>
> As a practical matter, mother nature does this to almost all FSK signals
> via propagation effects. Often received mark and space amplitudes
> different by a LOT more than the amounts we're talking about here. It
> is interesting to watch this. Given this mother nature bandwidth
> smearing, it is doubtful that the real world impact of K3 QAM vs pure
> FSK is significant.
>
> 73 de Brian/K3KO
> On 11/11/2014 06:56, Phil Wheeler wrote:
>> As a practical matter "My K3 increases power by 10% when in Space as
>> opposed to Mark": is this really something to be concerned about?
>>
>> Phil W7OX
>>
>> On 11/10/14 10:51 PM, Brian wrote:
>>> And what if this was done this with the standard 2.7 KHz filter?
>>> Filter passband ripple has the the stock answer to what people are
>>> observing.
>>>
>>> Also what about dynamic vs static differences. Is it possible in the
>>> dynamic situation the differences would be greater?
>>>
>>> It does sound like there is some confusion precision with accuracy. A
>>> measuring device might offer high precision and resolution but lower
>>> absolute accuracy. People often make accurate difference measurements
>>> with with a device which less accurate than the delta one observes.
>>> If the difference are small, they can be quite accurate.
>>>
>>> 73 de Brian/K3KO
>>>
>>>
>>> On 11/11/2014 02:13, Joe Subich, W4TV wrote:
>>>>
>>>> > My K3 increases power by 10% when in Space as opposed to Mark.
>>>>
>>>> Two things here ... 10% is abut 0.4 dB. How did you measure the power
>>>> difference and did you do so with an instrument with enough resolution
>>>> to accurately measure that difference at the power level you were
>>>> using?
>>>> For example a Bird Wattmeter can't possibly measure with that level of
>>>> accuracy nor is one likely to be able to read an analog meter that
>>>> closely.
>>>>
>>>> I have measured my K3 on RTTY with a wattmeter traceable to NIST and
>>>> specified for 3% accuracy and have done so at four specific power
>>>> levels (25, 55, 80 and 105 Watts) to eliminate any ALC non-linearity
>>>> or PA compression from consideration. The specific Mark/Space power
>>>> levels were measured by placing the rig into transmit in FSK_D mode
>>>> with the FSK input closed (Mark), measuring the power output for a
>>>> given power setting, then opening the FSK input and noting the PO
>>>> (and change).
>>>>
>>>> In all cases, the difference between MARK and SPACE was less than 2%
>>>> (less than the meter's specified accuracy) and at two power levels (55
>>>> and 80 W) the measured difference was less than 0.2W (<0.2 dB at 55W).
>>>>
>>>> These measurements were made with standard "High tones" (2125/2295 Hz)
>>>> and the K3 was equipped with the optional 2.8 KHz, 8 pole IF filter.
>>>>
>>>> 73,
>>>>
>>>> ... Joe, W4TV
>>>>
>>>>
>>>> On 2014-11-10 8:42 PM, Bill Lewis wrote:
>>>>> I am looking for information on how to eliminate incidental AM when
>>>>> using
>>>>> FSK-D.
>>>>>
>>>>> My K3 increases power by 10% when in Space as opposed to Mark.
>>>>>
>>>>> Thanks much, Bill, W8NN
>>
>> ______________________________________________________________
>> 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
>> Message delivered to [hidden email]
>>
>>
>> -----
>> No virus found in this message.
>> Checked by AVG - www.avg.com
>> Version: 2012.0.2247 / Virus Database: 4189/8045 - Release Date: 11/10/14
>>
>>
>
> ______________________________________________________________
> 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
> Message delivered to [hidden email]
>
>
> -----
> No virus found in this message.
> Checked by AVG - www.avg.com
> Version: 2012.0.2247 / Virus Database: 4189/8045 - Release Date: 11/10/14
>
>