[Amps] Measuring Q with an SWR analyzer

Karl-Arne Markström sm0aom at telia.com
Thu Jan 5 05:48:32 EST 2006


This can be made with simple formula, and by considering a few approximations as valid.

First, the Q must be "reasonable", this means that Q will have to be at least 5 - 6,
second the frequency excursions must be "reasonably" small, otherwise the 
asymmetric frequency response of the pi-network will show up. 

Also, the component losses in the network are assumed to be negligible.

The following analysis regards the pi-network as one-pole resonant network, which is
valid near the resonant frequency and for small frequency offsets.

The general expression for the transfer properties of a single-pole circuit is

E/E0 = 1/(SQR(1+ (2Q* (|f-f0|/f0))^2))

where E = output voltage after the network, E0 is the driving voltage, Q is the network Q, f0 the resonance
(center) frequency and |f - f0| is the frequency excursion. Twice |f-f0| is the bandwidth BW.

Through algebraic manipulations, approximate Q can be solved for as:

Q = (f0/BW) * SQR((1/((E/E0)^2)-1))


The term E/E0 contains implicitly the SWR at the network input, as it can be transformed to a power transmission coefficent 
= (1-rho^2) = (1-(SWR-1/(SWR+1))^2) = (E/E0) ^2.

SWR = 2, rho = 0,33 or return loss = 10,5 dB equals a power transmission coefficient of 0,89 (11% of the power is reflected)
and the corresponding E/E0 would be 0,94. 

If the -3 dB power transmission coefficient (0,707 voltage) is inserted; the term under the radical sign becomes 1, 
and the formula reduces to the familar Q = f0/BW


73/

Karl-Arne
SM0AOM

----- Original Message ----- 
From: "Bill Turner" <dezrat1242 at ispwest.com>
To: <amps at contesting.com>
Sent: Thursday, January 05, 2006 3:08 AM
Subject: [Amps] Measuring Q with an SWR analyzer


> A question for the math experts:
> 
> I would like to measure the Q of a pi network plate circuit directly 
> using an SWR analyzer.
> 
> Let's say the tube is in circuit with power off. I have connected a 
> 2200 ohm resistor from anode to ground to simulate the tube's plate 
> load impedance. I connect an SWR analyzer (50 ohm) to the output, set 
> to 14 Mhz, and tune the tank circuit for a perfect 1:1 SWR at that 
> frequency. Now I vary the frequency of the analyzer by a certain 
> amount - say 1 Mhz for discussion purposes - and now the SWR reads 
> 2:1, so the 2:1 SWR bandwidth is 2 Mhz at 14 Mhz.
> 
>  From this is it possible to calculate the Q of the tank circuit? 
> Obviously, the higher the Q, the smaller the 2:1 SWR bandwidth, but 
> is it possible to calculate it exactly?
> 
> This would be a very useful thing to do in order to check one's 
> calculations against reality.
> 
> Thanks in advance,
> 
> 73, Bill W6WRT
> 
> _______________________________________________
> Amps mailing list
> Amps at contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
> 
> 
> -- 
> No virus found in this incoming message.
> Checked by AVG Free Edition.
> Version: 7.1.371 / Virus Database: 267.14.13/221 - Release Date: 2006-01-04
> 
>


-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.371 / Virus Database: 267.14.13/221 - Release Date: 2006-01-04



More information about the Amps mailing list