RME 4350A Receiver Test Results

[W1VD]

Receiver: RME 4350A

Band	MDS	Blocking DR	Two-tone DR
		(20 kHz)	(20 kHz)
80 meters	-143 dBm	104 dB	81 dB
40 meters	-142 dBm	104 dB	80 dB
20 meters	-142 dBm	103 dB	78 dB

AM audio S/N: 49 dB


AM Audio Frequency Response: in dB, referenced to 1 kHz measurement   

100 Hz	200 Hz	400 Hz	600 Hz	800 Hz	1 kHz	2 kHz	3 kHz	4 kHz	5 kHz	6 kHz
0	+1	+1	+1	0	0	-3	-14	-23	-27	-32

AM Audio Distortion:

Mod	100 Hz	200 Hz	400 Hz	600 Hz	800 Hz	1 kHz	2 kHz
30%	14%	8.9%	7.1%	7.1%	7.1%	6.3%	7.9%
50%	20%	11%	7.1%	5.6%	5.0%	5.0%	7.1%
70%	28%	14%	7.9%	6.3%	4.5%	4.0%	6.3%
90%	32%	18%	11%	8.9%	7.9%	5.0%	8.9%
100%	35%	20%	14%	10%	7.1%	5.6%	8.9%

Notes: The AM distortion numbers show the somewhat typical increase at lower audio frequencies - most often caused by the low audio frequencies riding on the AGC line. Also there is the usual increase in audio distortion as the modulation percentage increases. This is due to the detector's inability to cleanly demodulate the higher amplitude levels.

Info on the measurement setups and test results on other receivers can be found at:

http://www.w1vd.com/BAreceivertest.html

[WA1GFZ]

Interesting that many of these old receivers had a great MDS and crappy everything else. Imagine when the bands were full of AM all the P&Ming

[N2DTS]

Plus, testing RX's at 20 KHz won't show how poor they really are.
At 5 or 10 KHz, (the typical bandwidth used on AM) they will look really bad....

Brett

[KM1H]

Are most of the new rigs any better?  Imagine the phase noise bandwidth from a healthy AM signal from most of the rice box crap on the market. Add that to the rest of the mix on the bands.

Those old radios would look a lot better if a real crystal or mechanical filter was added to the IF which happens to be what Ive done to several here. It doesnt require much work to improve the front end signal handling either.

Carl
KM1H

[WA1GFZ]

I would think mixers may need some help in older receivers along with better gain distribution.

[KM1H]

Yep, and some are really easy. Plug in a 6BY6 for a 6BE6. A 6SB7Y can sub for a 6SA7 or 6K8 in some sets. Those are the easy ones that yield a noticable improvement with no or minimal work.

The serious folk will go with the 6X8, 7360, 6ES8, and 12AT7 mixers mentioned here many times. Then there are the various RF amp tubes that can overcome mixer noise at 10M and at the same time run at only the gain needed to accomplish that.

Ive been preaching gain distribution for so long with usually blank stares that Id about given up hope

Carl

[WD5JKO]

Those old radios would look a lot better if a real crystal or mechanical filter was added to the IF which happens to be what Ive done to several here. It doesnt require much work to improve the front end signal handling either.

   Carl, I generally agree with you on issues, but what you said above makes me ask a question. Many of the old tube receivers have 1 or 2 RF stages followed by a mixer. The RF stages are AVC controlled, and the mixer if the 6SA7/6BE6 type may not have a high dynamic range. So in the case of say a RCA AR88 with dual RF amps (6SG7), and a mixer (6SA7) to 455 Khz we have a problem when a strong signal is say 20 khz away from the tuned signal. The RF amps will operate at maximum gain, and overload the mixer. This is because the selectivity of the IF is good enough to not see that strong signal, and therefore NO AVC is generated to cut back on the RF gain, hence a major overload.

  So I wonder why you said that adding better selectivity to the IF would improve these old radio's? Sure the added selectivity would help sort out a cluster of signals all close together, but wouldn't the added selectivity make the off channel signal handling issue worse?  By worse I mean keeping the RF stages at max gain when a big signal is just a few KC away.

  I have a RCA AR88, and live 100 yards from an OM SSB ham that runs an amplifier on 3918 Khz. When he talks, my AR-88 mutes - complete overload. if I switch to my Icom R75, I have no issues other than hearing the SSB guy's PTT relays click.
You also say that "It doesnt require much work to improve the front end signal handling either". I am all ears on this one, so what can I do to boost the dynamic range of the first 3 stages of an AR-88 when the strong off frequency signal is within the front end RF passband, but outside the IF passband such that NO AVC voltage is made to lower the front end gain?

Jim
WD5JKO

[w3jn]

Not to speak for Carl who I'm sure will jump in with his own good ideas, but there could be any number of modes by which the AR-88 is overloading. 

If you have standby connections, speaker wires, etc., the signal could certainly be coupling into the radio thru them as well as thru the AC line cord.   Although they're bypassed pretty well on an AR-88, the AR-88 isn't the sealed box that the R75 is.  Try disconnecting everything you can and see if that helps, then bypass and/or filter.

The R75 uses an entirely different design philosophy.  It has little front end selectivity, little gain before the mixer, (I also presume that the mixer is double balanced and presents a loss at this point) and up-converts to a low VHF range where it hits a roofing filter.  Then it's downcoverted one or more times and presumably run thru the ultimate selectivity filters.  Most of the gain is AFTER that roofing filter.

The AR88 relies upon pretty decent front end selectivity, mainly to reduce images - not strong adjacent signals.  The front end tubes all have gain as does the mixer.  The selectivty is progressive - ie the ultimate selectivity is attained by successive IF stages, not all at one with a multipole crystal filter.  So what you have is considerable off-channel signal blasting its way pretty far into the radio that's been amplified by successive stages, making things worse.  If things are bad enough you could end up with self-rectification at one of the tubes thus causing blocking as you describe.

At the end of the day we don't know exactly what's causing your AR-88 to block or exactly where the problem is.  A good xtal or mechanical filter right after the mixer is a good start, presuming the signal's not sneaking in thru the power line or other connections.  The mixer is the next obvious suspect, but not sure if you want to go dirilling and blasting on that radio! 

[KM1H]

Jim, old radios have RF gain controls that many dont realize are there Some control the RF, others RF and IF and a few IF only which is rather useless.

In addition the concept of a switchable front end attenuator has been around for over 50 years.

I was never impressed with the AR-88 I had many years ago and the only good RCA radio IMO is the RBB & RBC. Those do not overload. Of course they cost around 4 times more in similar years. Without studying the AR-88 design I can ony generalize. Changing to a 6SB7Y mixer or a more modern circuit on a plug in adapter. The variable mu 6ES8 dual triode mixer or 7360 would be the best IMO. Reduce the full on RF gain with circuit changes down to just enough to overcome mixer noise. If you dont care about the highest band that can be a substantial reduction especially in the 2nd stage. Use a 6EH7 on an adapter for the 1st stage which will further reduce the noise and gain can be lowered even more; that frame grid tube has a very high overload tolerance. Monitor the AGC curve, study tube curves, and make changes as required.

If you look at Jays HRO-60 test, note the IMD performance is among the best and that is with 2RF and 3 IF stages. I have to assume that was with the RF gain cranked wide open. The front end selectivity of any HRO is superior to its competition. Ive made a few changes to mine including an AM ceramic filter. It holds its own under 75M battle conditions, 10M AM E skip pileups, and fooling around in CW contests. That filter squares off the skirt selectivity right down to the noise floor and the radios 12 tuned circuits at 455 kc add to the brick wall. Someday I'll set up an IMD test and compare to Jays. National offered a mechanical filter adaptor and a few radios were sold with it to commercial customers. I might look at building a switchable version some day

Carl
KM1H

[w3jn]

Indeed.  Far and few between are the radios whose RF gain controls actually control RF gain.  In many radios the RF gain pot is connected to the cathode bus of the IF tubes and actually controls IF gain, not RF gain.

In the case of the AR-88, it's one of the few that throws an adjustable (thru the RF gain control) negative bias on the AVC line, thus providing gain control for ALL stages.  Better is the CR-88 diversity variant which has separate RF and IF gain controls...

[WD5JKO]

John, Carl, all;

   Thanks for the comments. My AR-88 is now owned by Geoff, W5OMR. It is still at my QTH waiting on his pick up after a repair. This radio was bought by Ozona Bob, W5PYT sight unseen for big bucks. What he got was an AR88 missing the wiring harness, two front panel rotary switches, and The IF can with the crystal phasing winding.

  Bob quite disgusted, screwed by another ham (again), offered the receiver to me so long as I promised him that I'd try to get it going again, and I did try, and to a large extend succeeded.

   My question to Carl was academic in that I have no intention at this point to drill or modify. Thanks for the responses though. Yes thanks for mentioning adding an RF attenuator, or reducing the RF gain control. 

   The funny thing about an AR-88 hooked to no antenna is that it is very quiet even with the audio control at maximum and the antenna trimmer peaked. Hook it to even a short piece of wire, and it explodes with audio in response to the RF picked up. With my HP-606 Signal generator, 30% moduilated AM and 0-1 uv scale, I can hear a tone all the way down the scale. It is weak, but there. No shortage of sensitivity. With large signals up to 1 v rms 30% modulated, the audio is clean. Everything changes though when that signal is outside the IF passband and yet within the RF passband. Overload city!

Ive been preaching gain distribution for so long with usually blank stares that Id about given up hope

   Amen brother! The AR-88 needs a little less RF gain, more mixer dynamic range, and more AF gain with a better tone control circuit. Sure you could add one of those Kiwa made 455 Khz IF filters as well like Brett did to his HB receiver. Beefing up the power supply is important as well since the multi section C-L-C-L-C filter is resonant somewhere around 80-100hz. Maybe someday Geoff will follow this thread. 

   My love affair to the big RCA has passed, as is my promise to Ozona Bob. Bob may not be around any longer, but I did fulfill my commitment to him.

Jim
WD5JKO