From: "Clemens Paul" <cpaul@gmx.net>
To: <rsgb_lf_group@blacksheep.org>
References: <CAHAQVWMte_9GEF90SMu09atqVUWzKHh-Wt+cxjkHH+AcR4ggYQ@mail.gmail.com> <CAHAQVWNyPNOf-eY1GtAcQ+b0HKu+mv=DdaZStS8vWM9LZVfPRw@mail.gmail.com> <CAGZ4oo942D_pYYrBXbcLGhVtbBMWQaYTrTcLZQ+ak3BU5LAxtQ@mail.gmail.com> <CANi3vvs+9NwNp6sAKbvkRFYsUS_q5jjaBb6xGG0Z9pih4LALAg@mail.gmail.com> <CAA8k23R7JurcyPGnRk0Q0VP+24XmiFwJLk-HMCDZXn6EFaVeYQ@mail.gmail.com> <50f24ef2-2d1d-b537-7d66-e6d12ecf8cc3@n1bug.com> <8e720108-88eb-11a7-4f41-ef67c7ae74c9@n1bug.com> <1UQWeI91kE.7cbTRbTeCVJ@optiplex980-pc> <d3ff127b-1242-1e3c-5259-6bdcaf1fe692@n1bug.com> <CAA8k23RF3LtP-Yn+DhYmu=nSGY93_9Dujrmrssu8_QyHcRP0bQ@mail.gmail.com> <1UQWn0FzVr.1HxCa9v5tM0@optiplex980-pc> <CAA8k23ROq3zeza_ut7U-+mbYz+HhFAq6rp8ap28s+VH2nm=AVg@mail.gmail.com>
In-Reply-To: <CAA8k23ROq3zeza_ut7U-+mbYz+HhFAq6rp8ap28s+VH2nm=AVg@mail.gmail.com>
Date: Sun, 15 Jul 2018 22:43:00 +0200
Message-ID: <FFBB5CAFA52C4FB3BA7EB9A43FE43F32@Clemens0811>
MIME-Version: 1.0
Thread-Index: AdQcCFtwDe6sc84jSM2lcRTv7NBrxQAcB83g
Subject: RE: LF: RE: Simple transverter for 472kHz
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

Hello Andy,

>Yes, I agree with your figures and they are terrible.  -125dBm 
>MDS means -128dBm of noise in a CW (call it 300Hz) bandwidth, 

-125MDS means that noise is at -125dBm,not at -128dBm.
You feed a signal of a sig gen into the receiver and adjust its level until a RMS AF voltmeter at the receiver's
AF output shows an increase of 3dB compared to no signal at input.
This means that signal and noise have the same level, hence the increase of noise+signal by 3dB.
The dBm readout of the sig gen then is MDS.
ARRL use 500Hz as 'CW-BW'.
So noise figure is 21dB.
According to CCIR curves man made noise in a "quiet receiving site" is 60dB.
Probably today even higher.
So there is plenty of room to shift the transverter's/receiver's dynamic range up by additional input attenuation
if it should be necessary to avoid 3rd order IM.
The front end topology can be found in the manual and is:
500kHz 7-pole LPF => 475kHz roofing filter with ceramic elements => 5-pole 500kHz LPF => mixer 

>Could there have been something wrong with the test unit 
>perhaps?   

...or with the test setup at such a low frequency?

73
Clemens
DL4RAJ 


>-----Original Message-----
>From: owner-rsgb_lf_group@blacksheep.org 
>[mailto:owner-rsgb_lf_group@blacksheep.org] On Behalf Of Andy Talbot
>Sent: Sunday, July 15, 2018 8:50 AM
>To: LineOne
>Subject: Re: LF: RE: Simple transverter for 472kHz
>
>Yes, I agree with your figures and they are terrible.  -125dBm 
>MDS means -128dBm of noise in a CW (call it 300Hz) bandwidth,  
>that means a noise figure of around 21dB so there is no front 
>end gain, and almost certainly some loss before mixer. 
>
>So a -36dBm IP3 is appalling.   The increases to -14dBm at 
>75kHz just means there is some filtering in place.
>
>That low an IP3 is the sort of figure you'd expect, perhaps, 
>from a single bipolar mixer;  the sort of thing used in the 
>1960s and 1970s (as I did in my first homebrew HF receiver !)
>
>Could there have been something wrong with the test unit 
>perhaps?   LO power drastically reduced into a mixer?  It 
>would be difficult to make a mixer with that low an IP3  othewise
>
>
>Andy
>www.g4jnt.com
>
>
>
>On 15 July 2018 at 00:54, jrusgrove@comcast.net 
><jrusgrove@comcast.net> wrote:
>
>
>	Andy
>	 
>	Maybe I can shed further light on my comments ... 
>	 
>	The ARRL receiver measurement system was developed by 
>Wes Hayward of Tektronics back in the mid 70's and has been 
>used with minor changes since. Instead of calculating 3IP the 
>system quotes third order IMD dynamic range relative to the 
>noise floor (MDS - minimum descernable signal). The MDS is 
>deemed as the signal level at which a 3 dB increase over 
>'background' noise is noted - as measured by an audio 
>voltmeter.  The measurement is normally made with the receiver 
>in a cw bandwidth and is quoted in dBm. QST quotes the MDS for 
>the this transverter as -125 dBm. For the two-tone IMD dynamic 
>range measurement, two tones (at selected spacings) are 
>injected into the receiver and the level is increased until 
>the third order products are 3 dB above the background noise - 
>at the MDS level. The two-tone dynamic range is the difference 
>between MDS and the signal generator level. 
>	 
>	For this transverter with a -125 dBm MDS and 60 dB 
>dynamic range at 2 kHz tone spacing, the indicated signal 
>level to cause third order IMD products would be -65 dBm. At 
>75 kHz spacing the signal level would be -51 dBm. Calculating 
>IP3 for 2 kHz spacing using -125 dBm and -65 dBm produces an 
>IP3 of -35 dBm. Calculating IP3 for 75 kHz spacing using -125 
>dBm and -51 dBm produces an IP3 of -14 dBm. 
>	 
>	I normally think in dynamic range numbers, not IP3, 
>since I've been using the 'Hayward system' since the 
>beginning. Since you think in IP3 numbers how do those IP3's 
>look to you?
>	 
>	Jay W1VD  WD2XNS  WE2XGR/2      
>	 
>	 
>
>		----- Original Message -----
>		From: Andy Talbot <andy.g4jnt@gmail.com>
>		Reply-To: <rsgb_lf_group@blacksheep.org>
>		To: LineOne <rsgb_lf_group@blacksheep.org>
>		Sent: 7/14/2018 7:38:16 AM
>		Subject: Re: LF: RE: Simple transverter for 472kHz
>________________________________
>
>		Those IMD figures don't in themselves say 
>enough to say whether it is good or bad. 
>
>		Firstly, that quoted paragraph doesn't give the 
>input levels at which IMPs were 60 [75dB] down.   It is more 
>usual to give a third order intercept point  (TOIP)when 
>specifiying linearity; a figure quoted in dBm
>
>		Secondly, specifying IMPs at different 
>bandwidths is meaningless too, as there is no indication of 
>the receiver filtering.
>
>		So all in all, rejecting that receive converter 
>based on that quoted paragraph is going too far.
>
>		If a level-7 (+7dBm Local Oscillator) diode 
>ring mixer were to be used at the front end, with no preceding 
>preamp and minimal loss filtering, you might reasonably expect 
>a TOIP in the +15 to  +20dBm region.   That means IM Products 
>will be, (in dB below theinput) twice the amount the RF input 
>is below the TOIP.  An example :
>
>		TOIP = +20dBm
>		Two tone input at a level of -20dBm
>		IMPS   =    +20dBm - 2 * (+20dBm - -20dBm = 
>-60dBc on the input signals.  Or equivalent to -80dBm
>
>		So two -20dBm signal into a typical 7dBm LO 
>diode ring mixer give -60dBc third order products.
>		A diode ring is what any self respecting 
>designer might use on a basic LF receiver converter
>
>		A level 13 mixer (+13dBm) wpouild give a 
>proprtionately higher TOIP, perhaps +25 to +30dBm
>		For higher linearity / better strong signal 
>handling still,  a bus-switch mixer offers a TOIP perhaps +35 
>to +45dBm.   Like the Softrock receivers
>		
>
>
>		Andy 
>		www.g4jnt.com <http://www.g4jnt.com/> 
>		
>
>
>		On 14 July 2018 at 12:13, N1BUG <paul@n1bug.com> wrote:
>		
>
>			Ouch! Having operated on 160m and the 
>HF bands for several years
>			using a receiver that comes in at about 
>-63 dB on 2 kHz spacing
>			third-order IMD I would never again buy 
>something with that kind of
>			receiver "performance".
>			
>			Paul N1BUG
>			
>			
>			On 07/06/2018 09:45 PM, 
>jrusgrove@comcast.net wrote:
>			> Saw the writeup in QST and 
>immediately noticed the receive
>			> two-tone, third-order IMD at 2 kHz 
>spacing is 60 dB and at 75 kHz
>			> spacing is only 74 dB! Good luck with 
>that if you've got strong
>			> signals at your location.
>			> 
>			> Jay W1VD  WD2XNS  WE2XGR/2
>			
>			
>
>
>
>
> 
><http://www.avg.com/email-signature?utm_medium=email&utm_source
=link&utm_campaign=sig-email&utm_content=emailclient> 	 Virenfrei. >www.avg.com 
><http://www.avg.com/email-signature?utm_medium=email&utm_source
=link&utm_campaign=sig-email&utm_content=emailclient>  	
>
>