Re: {Collins} Receiver test data, Sherwood



Hi Gary,

The DR3 measurement is very useful, but it requires a fair bit of understanding lest you fall into pitfalls and create invalid data. The primary pitfalls involve 3rd order intermods produced by poor isolation between the two signal generators, and generator noise that falls into the passband of the receiver under test. I have purchased a couple of crystals that I use as noise clean-up filters on each signal generator to ensure signal generator noise isn't a problem. Once you are certain you aren't getting fooled by either of these two issues, you then must be certain the noise from the mixer LO isn't creating issues due to the "reciprocal mixing" phenomenon.

On top of all this, be aware that the DR3 number is dependent on the IF bandwidth of the receiver under test. If you look at Rob's data on S-Line units (75S-3, 75S-3B and 75S-3C), there are questions to be asked. I asked them all, but Rob couldn't recall the state of these receivers. One question is what IF filter was used when DR3 was measured. This is important since DR3 is measured with respect to the receiver noise floor. The sensitivities on Rob's data sheet are -145dBm, -146dBm, and -141dBm respectively for the 3 receivers. Since the noise figure of a properly aligned S-Line receivers is about 4-5dB, these sensitivities indicate they were measured with crystal filters, although the -141dBm for the 75S-3C doesn't really make sense for any standard S-Line receiver bandwidth. Perhaps this receiver wasn't properly aligned.

So, all the above are things that affect DR3 measurements. On top of that, DR3 of S-Line receivers without SB2 installed is dependent on the type of IF filter used (FA mechanical, Inrad mechanical or crystal). Looking at Rob's data (and the DR3 data in my article) you can discern that Rob's 75S-3B and 75S-3C likely have the SB1 and SB2 installed. The 75S-3 (and any 75S-3B without any service bulletins installed) will have considerably lower DR3, as indicated in Rob's data.

Bottom line is that DR3 measurement isn't a trivial undertaking, and I've encountered about every pitfall imaginable over 40+ years of designing receivers. You really want to know what you are doing. On the other hand, if you are curious, go ahead and make some measurements. As you try to confirm the validity of your data, you will definitely learn a great deal about receiver theory.

Don, W5QN

-----Original Message----- From: Gary J FollettDukes HiFi
Sent: Thursday, September 29, 2016 7:27 PM
To: Dave Bottom
Cc: w2iy@xxxxxxxxxxx ; Don Jackson ; Collins CCA Reflector
Subject: Re: {Collins} Receiver test data, Sherwood

There is no doubt that he knows more about receivers than just about anyone else. That is why I often look to his information when I am trying to learn.

However, I do make measurements all of the time, of a more critical nature, for my work, and thus I always have the concept Gauge R&R in the back of my head when I see comparisons between something producing a 104 dB reading (for any test parameter) and 105 dB. I would submit that those two receivers are the SAME and not try to rank them in order. It’s a tie…


Gary




On Sep 29, 2016, at 12:37 PM, Dave Bottom <ars.kd6az@xxxxxxxxx> wrote:

Back in 2009 when restoring my 75S-3B I was really pleased with the results except that there was virtually no perceptible difference between Fast and Slow AGC compared with my 75S-1 (from the 1970's) and I ran across some interesting material on the Internet about specifically my issue.

I sent an email off to get some further insights, not recognizing Rob's new call sign. It was only after a couple of exchanges and talking with him on the telephone I realized who I was talking with.

Rob does indeed own Collins gear, at least an S-3Line when I talked with him and I could tell he enjoys using it. The 75S-3B tested was undoubtedly his own well tuned receiver. The 75S-3C might have been someone else's and he may have not been inclined to mess with that person's radio, but since he is so approachable, just ask him!

By the way his input on tweaking my 75S-3B's AGC so it was Slow for SSB and Fast for CW or AM was a huge success. It was just as I was expecting it to be.

73 Dave WI6R

Sent from my iPhone

On Sep 29, 2016, at 9:22 AM, Michael Tortorella <w2iy@xxxxxxxxxxx> wrote:

Granted that the test must be somewhat time-consuming (so it would be
difficult to expect Rob to undertake as many tests as needed to get a
reasonable statistical result), Gary's point about the small sample size and
unit-to-unit variability is indeed well taken.  The Holmdel ARC (K2DR) at
Bell Labs did something similar back in the late 1970s, making some MDS and
DR3 measurements on receivers owned by members of the club.  We had a
similar experience, lots of unexplained strange results, many probably due
to not aligning the units before the test.  It was fun, though, and
certainly early days not long after Wes Hayward's article in QST on the
matter.  Somewhere I have a table of the results we got, if anyone is
interested I can pass it along.
73
Mike W2IY

-----Original Message-----
From: Collins [mailto:collins-bounces+w2iy=verizon.net@xxxxxxxxxxxxx] On
Behalf Of Gary J FollettDukes HiFi
Sent: Thursday, September 29, 2016 12:05 AM
To: Don Jackson
Cc: Collins CCA Reflector
Subject: Re: {Collins} Receiver test data, Sherwood

Thank you for the response, a well spoken response I might add.

I'll read the referenced article as well.

Gary

On Sep 28, 2016, at 9:57 PM, Don Jackson <w5qn@xxxxxxxxxxx> wrote:

Hi Gary,

Ah yes, Rob Sherwood's receiver data is interesting, and I spoke with him
a couple of years ago before writing the article on dynamic range that
appeared in Q1 2016 Signal magazine. Yes, 2-tone dynamic range has become an increasingly important differentiator among receivers over the years because it is a good measurement of how well a receiver performs in a crowded signal
environment.

Rob did not test many Collins units (maybe only the ones listed), and more
unfortunately, did not record the status of service bulletins performed on
the units. In a nutshell, SB1 and SB2 affect the DR3 performance. SB1
improves DR3 by reducing the gain in front of the IF filter. SB2 improves
DR3 by adding selectivity prior to the mechanical filter, which,
unfortunately, contributes to 3rd order distortion. If you haven't read it, I recommend that you check out the article, and you will see comparisons of
various S-Line receivers with different combinations of service bulletins.
As well, the article describes how changing to an Inrad mechanical filter
will improve DR3.

If you have further questions, let me know and I will attempt to answer
them.

Don, W5QN
Technical Editor, The Signal

-----Original Message----- From: Gary J FollettDukes HiFi
Sent: Wednesday, September 28, 2016 8:46 PM
To: Collins CCA Reflector
Subject: {Collins} Receiver test data, Sherwood

I am always out to learn and the best way to do so it to ask hard
questions and learn from those who have the answers.

I was just wandering around the Sherwood site regarding receiver dynamic
range ratings (again) and I noticed a curious thing.

The Collins 75S-3B is rated 6 spots higher than the 75S-3C.

Now the only difference between the two (of which I am aware) is the
addition of the second crystal board and associated switches for added band
coverage.

I think maybe this tells you what the "Gauge R&R" (Readability and
Repeatability) is for this test.

Also, I was wondering how many samples of a given radio are tested, since
there MUST be some degree of unit to unit variation.

I know there are other things the drive receiver performance quality
measurement but I find it interesting that this one test gets such high
visibility when it can't even produce the same result from two different
versions of the same receiver.

Even Direct Digital SDR radios get this figure of merit rating even though
they have no mixers! Interesting.

Gary
*** You are subscribed to Collins as w5qn@xxxxxxxxxxx. If you wish to
unsubscribe, or modify your preferences please visit
http://mailman.listserve.com/listmanager/listinfo/collins ***

*** You are subscribed to Collins as w2iy@xxxxxxxxxxx. If you wish to
unsubscribe, or modify your preferences please visit
http://mailman.listserve.com/listmanager/listinfo/collins ***

*** You are subscribed to Collins as ars.kd6az@xxxxxxxxx. If you wish to unsubscribe, or modify your preferences please visit http://mailman.listserve.com/listmanager/listinfo/collins ***




This archive was generated by a fusion of Pipermail (Mailman edition) and MHonArc.