From: "Clemens Paul" <cpaul@gmx.net>
To: <rsgb_lf_group@blacksheep.org>
References: <555BAAF1.4020800@posteo.de>
In-Reply-To: <555BAAF1.4020800@posteo.de>
Date: Wed, 20 May 2015 16:00:04 +0200
Message-ID: <C4417F1505A54A2B8FF6DFA28D2411C6@Clemens0811>
MIME-Version: 1.0
Thread-Index: AdCSeubTp77qYWd4T/6argssq0gO+wAiSRUA
Subject: LF: RE: FZ-02-B mechanical filter input matching
Content-Type: text/plain;
	charset="US-ASCII"
Content-Transfer-Encoding: 7bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Status: O

Hi Stefan,

>Now my question: Is it better to remove a BF981 amp stage 
>(about 16 dB to 18 dB gain!) in front of the mixer (SBL-3) to 
>get a lower RX noise or do i better reduce the gain on the 
>LT1028 stage on the AF side? The AF pass band is 11...18 kHz 
>(LO= 461 kHz)

A general design rule for cascaded amplification stages is that the 
noise figure of the first stage is the most crucial one for the overal noise figure. 
So I would leave the gain before the lossy mixer as it is and reduce gain after the mixer.

http://194.75.38.69/applications/mcl_nf_calc.html
http://www.custommmic.com/Calculators/Cascade-Analysis/

73
Clemens
DL4RAJ 

>-----Original Message-----
>From: owner-rsgb_lf_group@blacksheep.org 
>[mailto:owner-rsgb_lf_group@blacksheep.org] On Behalf Of DK7FC
>Sent: Tuesday, May 19, 2015 11:28 PM
>To: rsgb_lf_group@blacksheep.org
>Subject: LF: FZ-02-B mechanical filter input matching
>
>Hi MF, 
>
>Some of you are using the old Telefunken mechanical filter in 
>homemade receivers for 630m. I use them i all my MF receivers.
>
>There have been discussions about how to match the antenn to 
>that filter. 
>So far, i used this matching method: 
>https://dl.dropboxusercontent.com/u/19882028/MF/MF%20Filter%20s
chematic.png (recently using the 500 Ohm input rather than the 18k input) >i.e. an active impedance converter stage. 
>This has a few advantages:
>-The input of the filter can be matched perfectly to the jFET, 
>giving a minimun pass band ripple (less than 1 dB)
>-The matching is independent of the impedance of the voltage 
>source connected to the input of the jFET
>-You can use a 50 Ohm resitor on the input of the jFET, giving 
>a perfect matching of the cable for all frequencies
>-The input can even be used as an active antenna (very high Z)
>-Many receiver inputs can be switched in parallel! For example 
>a LF RX parallel to a MF RX plus one 50 Ohm resistor...
>-The input of the mechanical filter must not be exceeded to a 
>voltage above 2 V rms (inside the passband or on any 
>frequency?????) So, if the jFET is driven at 5V supply 
>voltage, the filter is protected against overvoltages!
>These were the arguments why i decided to use this technique 
>so far. Here is an image of the RX on my remote site, captured 
>today: 
>https://dl.dropboxusercontent.com/u/19882028/MF/MF%20mechanical
>%20filter%20active%20input.jpg
>
>But now, on my remote RX site i have a relatively large RX 
>antenna (T, 5m V, 20m H) with a 50 Ohm matching. Furthermore i 
>want to reduce the power consumption of the RX system. And i 
>want to try to improve the RX performance even more. So i've 
>made a modification, a simple transformer matching (nothing 
>new!!!!!!!) as recently discussed.
>Here the advanatges are:
>-No active parts in front of the mechanical filter, just wire, 
>Ls, Cs, cable and ferrite cores!
>-One active stage less, i.e. less power consumption, about 10 
>mA at 12V, i.e. 0.24 Ah per day!
>-A 'gain' (relative to the active stage which has a voltage 
>gain of - 6 dB due to the resistor matching) of 16 dB 
>(confirmed by measurements on SpecLab) because the transformer 
>makes 10 dB when transforming from 50 Ohm to the 500 Ohm input 
>of the filter
>-This gain makes one of the following amp stages obsolete, 
>i.e. likely i can remove one more active stage which will 
>result in even less power consumption, maybe 0.5 Ah per day, 
>which is much in winter!
>-Maybe less RX noise?
>-Maybe less unwanted IM products??
>
>Here is an image of the simple new input matching 
>https://dl.dropboxusercontent.com/u/19882028/MF/MF%20mechanical
>%20filter%20passive%20input.jpg
>One of the advantages of the active input matching is missing 
>now: the perfect matching. The pass band ripple is now about 2 
>dB but this is still acceptable for me if there are some 
>advantages on the other side.
>
>The second amp stage is not yet removed in my currently 
>running RX, so there is to much gain in the moment, i.e. a 
>smaller dynamic range. I just wanted to confirm the 16 dB to 
>decide the next steps.
>
>Now my question: Is it better to remove a BF981 amp stage 
>(about 16 dB to 18 dB gain!) in front of the mixer (SBL-3) to 
>get a lower RX noise or do i better reduce the gain on the 
>LT1028 stage on the AF side? The AF pass band is 11...18 kHz 
>(LO= 461 kHz)
>
>73, Stefan/DK7FC
>
>