[hpsdr] IMD in Ham transmitters

[Kjell Karlsen]

Hi All.

I think it is better to stay with the "ARRL" definition of IMD products  
relative to the PEP level. When I started to work with Spectrum Analysers  
back in 1968, the commercial definition followed this and I have not seen  
anything else later. The rules, at least then, was that all IMD products  
should be -32 dBc refereed to the claimed PEP for the TX.

When using 2 tones, the PEP is 6 dB higher than each of the 2 tones. If 3  
tones are used, the PEP is 9 dB higher and so on. It is always necessary  
to know the (max) PEP of the PA and the 1dB compression point is very near  
this level.

Keith: What you say about the 3 times increase in IM3 is correct at low  
levels of the output power. But when the power is near the saturation  
point, the IM3 increase more than 3 times the output. Using Hermes as TX  
and Mercury as RX it is easy to see this. The broadband OPAMP used in  
Hermes and Penny is extremely linear up to 400mW but above, the IM3  
increase much more than 3 times.

I think this shows that to know the maximum PEP and refer to this when  
measuring IMD is important. But anyway, if you specify the reference, no  
problem.

73, Kjell



På Sun, 20 Jan 2013 09:27:34 +0100, skrev Helmut, DC6NY <dc6ny at gmx.de>:

> ***** High Performance Software Defined Radio Discussion List *****
>
> Hi guys,
>
> first of all you have to know at least the -1 dB compression point of  
> your
> amp, better the transmission characteristic. Makes no sense to measure in
> the area of saturated power. This defines your max. PEP-power. With 2  
> tones
> this level is already achieved with -6 dB power per tone. This has  
> nothing
> to do with the distance to the IM products. They are defined as distance  
> in
> dBc relative to the two tone levels.
>
> 73, Helmut
>
> -----Ursprüngliche Nachricht-----
> Von: hpsdr-bounces at lists.openhpsdr.org
> [mailto:hpsdr-bounces at lists.openhpsdr.org] Im Auftrag von John
> Gesendet: Sonntag, 20. Januar 2013 01:50
> An: Keith Gooley
> Cc: hpsdr at lists.openhpsdr.org
> Betreff: Re: [hpsdr] IMD in Ham transmitters
>
> ***** High Performance Software Defined Radio Discussion List *****
>
> Years ago I used to measure IMD on Collins equipment.  Their specs at
> that time were always in dB below each of two equal tones.  If you think
> about it the peak of the envelope with only two equal tones is twice
> each tone peak value, so Collins spec was more conservative than other
> manufacturers of that time.
>
> In those days the spectrum analyzer and two tone oscillators took up a
> full single bay rack (on wheels) called a 478R-1.
>
> Sorry I incorrectly sent my answer to your post as an answer to Lyle's
> in my last Email.
>
> John  WoGN
>
> On 1/19/2013 4:38 PM, Keith Gooley wrote:
>> ***** High Performance Software Defined Radio Discussion List *****
>>
>> Hi All,
>>
>> If the commercial suppliers are claiming an extra 6 dB for PEP then
>> that doesn't seem right in my view. The IMD level increases at a rate
>> greater than the increase  in power, 3rd order IMD which is mostly
>> what we are talking about, increases at 3 times the rate of increase
>> of the 2 signals generating the IMD product. So if the output power is
>> doubled (3 dB increase) the IMD increases by 9 dB. Hence the IMD level
>> relative to the main signal is 6 dB worse.
>>
>> 73 de
>> Keith
>> VK5OQ
>>
>
> _______________________________________________
> HPSDR Discussion List
> To post msg: hpsdr at openhpsdr.org
> Subscription help:
> http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org
> HPSDR web page: http://openhpsdr.org
> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/
>
> _______________________________________________
> HPSDR Discussion List
> To post msg: hpsdr at openhpsdr.org
> Subscription help:  
> http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org
> HPSDR web page: http://openhpsdr.org
> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/
>
>


-- 
Sendt med Operas revolusjonerende e-postprogram: http://www.opera.com/mail/

 1358852378.0