[TMC] GPR-90 RXD Tuning

[John Vendely]

Hi John,

The 4 Mc highpass would be used on all bands which utilize double 
conversion, since it's needed to provide rejection of RF input signals 
around 3.955 Mc which will work their way directly into the 1st IF.  So 
this filter would be used on the top three bands, any frequency above 
5.4 Mc.

The different simulation component values might stem from differences in 
the termination impedances you chose for the simulation vs. what 
actually exists in the receiver.  This could result in significantly 
different values.  I've never seen measured input impedance curves for 
this receiver.  The IF rejection filter response could vary a bit 
without affecting receiver performance too adversely.  They may  have 
tolerated some passband response variation, provided the filter provided 
good IF rejection and didn't hurt sensitivity too much.  Quite a bit of 
RF phase response variation over wide frequency ranges might also have 
been tolerated, provided the phase variation over small bandwidths 
(equal to the widest IF BW) wasn't too large.   BTW, it looks like some 
versions of the GPR-90, presumably the earlier ones, had only the BCB 
reject filter.  The 1st IF rejection filter may have been added later 
after problems were noted.  I would expect it was a necessity, and it's 
surprising it wasn't used in all versions.

73,

John K9WT

On 3/16/2014 3:42 PM, John Poulton wrote:
> I was a bit more curious about the two high-pass filters (it IS odd that
> TMC got the sense of the filter wrong in the writeup.. :), so I visited
>
> http://www-users.cs.york.ac.uk/~fisher/lcfilter/
>
> a site that allows one to design various kinds of filters.  I set the thing
> up for Butterworth, 5-pole, shunt input, with 2Mc cutoff, 300-ohm
> input/output impedance, and synthesized a filter very similar to the
> BC-band filter that's in the receiver.  The values calculated by the
> program are a bit different, but close.
>
> The other (4Mc) filter is a different story.  This is a series-element
> input/output filter of 7th order.  The on-line program produces values that
> are fairly different from those used by TMC, the main difference being that
> the center inductor, which is quite a bit smaller than the outer pair of
> inductors.
>
> Given the relatively rough phase response of the simulation I ran for the
> existing filter in the radio, I wonder if it might be possible to improve
> on this filter, and whether it could actually improve the receiver's
> performance.
>
> If I'm reading the schematic correctly, the 4Mc filter is only used on the
> top band, which most of us probably use rarely, if ever..
>
> 73, John K4OZY
>
>
> On Sat, Mar 15, 2014 at 1:08 AM, Richard Knoppow <1oldlens1 at ix.netcom.com>wrote:
>
>> ----- Original Message ----- From: "John Poulton" <jp at cs.unc.edu>
>> To: <jvendely at cfl.rr.com>
>> Cc: <wb9tow at egr.msu.edu>; "tmc" <tmc at mailman.qth.net>
>> Sent: Thursday, March 13, 2014 12:06 PM
>> Subject: Re: [TMC] GPR-90 RXD Tuning
>>
>>
>>   Just for the hell of it, I modeled the LC filters in the front end of
>>> the GPR-90 using HSPICE.  The filter comprising L28,L29,L30 and
>>> C115,C117 appears to be there to attenuate the B/C band. It's -6dB
>>> point is at almost exactly 2Mc and it has reasonably flat response
>>> in both magnitude and phase above 3Mc.
>>>
>>> The other filter, L32,L33,L34 and C138,C139,C140,C141 appears to be a
>>> high-pass filter with a cutoff at about 4Mc, but with significantly
>>> rougher mag and phase response.
>>>
>>> On higher bands, the two filters are cascaded, and the higher frequency
>>> filter (of course) determines the input response.
>>>
>>> For both, I loaded the output of the filter with 250 ohms, roughly 1/Gm
>>> for a 6AB4.
>>>
>>> John, does that square with what you expected from this filter?
>>>
>>> What does the circuit comprising L35, C143, and C142 do..??
>>>
>>> 73, John K4OZY
>>>
>>>       I was looking at the schematic and parts list in the GPR-90 hand
>> book from your site. This is the standard resolution one with the 1962
>> errata sheet in it.  Curiously, all the caps and coils in the RF are
>> described as being part of the "low-pass filter".  Of course with series
>> caps and parallel coils it can't be. I also have an earlier version of the
>> IB with the exact same error in it.  Very odd lapse of technical writing.
>>
>>
>>
>> --
>> Richard Knoppow
>> Los Angeles
>> WB6KBL
>> dickburk at ix.netcom.com
>>
> ______________________________________________________________
> TMC mailing list
> Home: http://mailman.qth.net/mailman/listinfo/tmc
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:TMC at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>
>