Amps
[Top] [All Lists]

Re: [Amps] A tale of two IMs What happens?

To: Roger <sub1@rogerhalstead.com>
Subject: Re: [Amps] A tale of two IMs What happens?
From: "Dr. David Kirkby" <david.kirkby@onetel.net>
Date: Thu, 22 Apr 2010 11:52:21 +0100
List-post: <amps@contesting.com">mailto:amps@contesting.com>
Roger wrote:
> 
> Gary Schafer wrote:
>> IM and cross modulation are the same thing. With the catv amps there are
>> many signals operating at nearly the same level that mix with each other to
>> produce new products, some of which fall on the frequency of existing
>> signals. The only way for IM products or cross modulation products to appear
>> is by multiplication.
>>
>> The levels of the IM or cross modulation products increase in strength, with
>> cascaded amplifiers because those product levels add together from one
>> amplifier to the next in a linear fashion. Each amplifier produces its own
>> IM products that fall on the same frequency as those produced by the
>> previous amplifier.

To be more precise it is the *voltages* which add. The voltages must be added 
as 
  vectors, not scalers, so it is not true to say the powers will add. They 
could 
cancel under some highly unlikely set of circumstances.

> I believe it's rare for any amp to have the output either in phase of 180 
> degrees out of phase with the input. BUT IIRC it's not all that complicated 
> to set the phase to what every you desire in the design. 

For a spot frequency, all one needs to do is have the right length of cable on 
the output - you can get any phase shift you want.

However, that will change the phase of the IM products by the same amount, so 
it 
does not achieve you anything useful.

> If it's true that each amp generates identical IM products in amplitude 
> and frequency  then it'd seem to be a relatively simple matter to set 
> the output phase 180 degrees from the input and end up with a net zero 
> IM for each even number of amps while each odd number would have only 
> those IM products from that amplifier, but I don't think it's quite that 
> straight forward or simple. Then again I was wrong one other time.

You are not wrong about it not being as simple as that.

I think to do this analytically one would have to consider the non-ideal 
amplifier as being a mixer and an amplifier. Perhaps I'll look at this at some 
point. Using Sage

http://www.sagemath.org/

should take out some of the hard work of the maths! (Though to be honest I know 
Mathematica better, but Sage is free, whereas Mathematica is several thousand 
pounds or dollars a copy).

Dave
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps

<Prev in Thread] Current Thread [Next in Thread>