Practical discussion about alignment of my receiver, SX-28

[N6YW]

Greetings
I will be blunt, I suck at aligning receivers. My last two attempts were done on 2 of the hardest that I know of, the RME 69 and SX-28. My sage guru Wayne Spring K6IRD had to give the SX-28 back to me because of work commitments, so I am stuck with having to do it myself which of course was the cause of the problem to begin with.  
So I reasoned, perhaps we could have a quasi step by step discussion that would allow not only myself to plod through this, but others could benefit from the posts by our resident gurus.
I have the rudiment test equipment, scope, basic RF generator, etc.
I have already recapped the rig and it works on 530 thru 1500 kcs, but all of the other bands are way off. I should add that I attempted the alignment by using the original manual and it just didn't translate to me. It doesn't sound very good either, so I don't think it's heterodyning correctly, such as when I do hear an AM station on 75 meters, it's very distorted.
So, in my perfect English and Southern Gentleman manners, allow me to ask for HELP!!!!
73 de Billy N6YW

[WBear2GCR]

I have not done an SX-28 myself. The one I got over the summer is next up after the BC-779 that is on the bench is finished (almost done now).

The first line of defense is to run a signal in the front end on the band that is selected. Usually I start with either the band that seems to work or the lowest freq band.

In the case of the SX-28 I've read enough to know that the entire set of coil assemblies must be physically removed from the chassis and recapped completely. Has this been done with yours? Resistors also should be checked carefully and in some cases replaced. It won't work properly with leaky and otherwise out of spec caps.

Anyhow, what I do in general terms is to follow the signal from the front end to the last IF with a scope. Alternately, the S meter is an indicator. One would look first at the plates of each stage to make certain that there is signal of adequate amplitude present. A working band gives something of a useful signal level to make comparisons.

Assuming you see signal all the way through to the end of the IF, then alignment makes sense.

The other way to work it is from the detector back up the IF to the RF. Here you would capacitively couple the sig generator at the proper frequency into the grid of each stage and watch to see that it is properly amplified and mixed. Then move back to the prior IF section until you reach the ANT input. In this way you can get at a problem area from "both sides".

After that, actual alignment usually consists of peaking/adjusting the IF for the proper center freq and shape, and getting the best settings for ends of band for the RF and VFO sections.

That's all I can offer until I do an SX-28 up myself. There are some pretty good write ups online about doing these receivers.

                                         _-_-bear

[Jim, W5JO]

The thing about aligning the SX-28 can be the dual AGC section.  That is a difficult read for anyone not real familiar with the work.  I suggest you ignore that part until the very last.  The RF and IF are pretty straight forward.  Just be sure to disable the AGC while doing those sections.

[KA0HCP]

Billy,
You have my sympathies.  I have gotten lost in an alignment before, and was able to eventually work my way out by relying on the Oscope, and methodically repeating the steps.  There is often interaction between the controls so when you make small progress it is worth repeating to home in.

Manuals:  Do you have both the civilian and the military tech manual for the SX-28A?  While there are small differences, the military manual is much more detailed.  Both available at the BAMA Mirror.

Resistors:  Check your resistors if you haven't already.  Some people take shortcuts and skip a comprehensive test of all the resistors.

If you are lost don't endlessly twiddle the coils, things seldom get better.  Take a break, even a few days until you are rested and in a calm state of mind.  Re-read the manual and start again with step One of the alignment.   Use your instruments and don't 'work by ear'.   

If the manual doesn't make sense:
-Ensure you have working copies of the manuals and schematic you can markup and highlight.
-Read through the procedures, making notes, highlighting, until you know all the test and adjustment points and what is being done.  Work one stage at a time

Hang in there, Bill

[WQ9E]

Make sure that the dials are properly aligned to the main and band spread capacitors before you start or you will end up going through the entire process again. 

On the upper bands it is very easy to end up with the HFO tracking on the wrong side which will make proper calibration and tracking impossible.  For this reason I like to use an external receiver to set the HFO adjustments to avoid aligning on the image.  If calibration and/or tracking on a band is really far off because of prior misalignment I go to the center of the band and with alignment capacitors at around 50% rotation I adjust the inductors for proper calibration/gain.  This gives you a reasonable starting point to then follow the regular alignment protocol in the manual.

It is very common to find the crystal filter element is dirty in receivers of this age and cleaning with alcohol will restore its proper performance.  National printed the procedure in their early manuals but basically you disassemble the crystal holder and clean the crystal and pressure contacts with isopropyl (rubbing) alcohol then dry with a clean paper towel, don't touch the crystal or holder with your fingers after cleaning.  You can do a before and after test with your signal generator and a scope or RF voltmeter, a healthy crystal will show a sharp and clean peak when the generator is swept through its resonant frequency.

[N2DTS]

I never had an sx28, I did have an sx17 which was not as complex I suspect, and it never needed alignment.
But I can say that an SDR receiver that picks up the IF is very helpful, short of an old warbulator it is hard to get the IF tuned right without being able to see the results.
You can also see the LO frequency and amplitude, drift, etc.

They really are spectrum analyzers that the ham can now afford.
The sdr-iq is very good for that sort of work.

[N6YW]

I appreciate the comments. This made me rethink my objectives and how to approach this.
The idea of cleaning the crystal makes perfect sense. I will start there, and then get that correct before I do anything. I also own a Flex 5000 A. It's a great instrument and am curious as to how I can utilize it's abilities, like as in hooking up a scope? That would make sense.
Like I said, this is a practical discussion. Thanks!

[KA0HCP]

One other tip:

If you have 'lost' the signal, use your Oscope and signal generator.  VERY SLOWLY, sweep the generator until you find a peak or increase in signal.  If it is too far from the 'center' it may be quite broad and not very 'high'.

  Then 'walk' the peak back to the IF frequency by small tweaks of the IF controls, then adjust the generator in the same direction.  Slowly repeat this in small steps.  The signal should build in height and narrow as it approaches the IF frequency.

*Be slowww and deliberate with your generator adjustments.  These are analog circuits and the coils can be quite slow in building resonances.  Some signal generators may have rather coarse adjustment knobs that compound the problem ...mine does.  Likewise Oscopes can be slow to capture the waveform.

*Consider using a frequency counter if your signal generator does not have a precision frequency readout.  Oscopes with frequency measuring can be slow to update at low signal levels and may be susceptible to aliasing, giving a false reading.

*Use the lowest signal level that provides a good response.  Too high a level can swamp the IF and give overly broad adjustment.

*Ensure you are using well constructed, shielded coax cables for signal insertion.  No place here for clip leads and scrap wire.   If the manual requires special loading devices....build them!

*Follow the manual, but in general radio [Frequency Coverage] are aligned first at the bottom end of frequency by adjusting the [RF] coils (L-Inductance) "L=Low end", then the C-Capacitance, "Cap=High end".  

Final thought:  If truly lost, start by aligning the IF's.  Once they are accurately centered the RF sections naturally follow and then the rest of the circuits.  

bill.     [clarified RF alignment]

[WBear2GCR]

If the Flex can receive at 455kc (I think that is the IF freq...) or whatever the IF freq, you just hook up coax from the point of interest (cap couple!) to the Flex receiver (disable xmit!!) and what you see on the screen is a picture of the passband!!

So now you can perfectly align the IF cans, one by one!!

There is no need to clean the xtals at this point, IF you have the selectivity in a position where the xtals are bypassed.

[KA0HCP]

http://antiqueradios.com/forums/viewtopic.php?f=5&t=263448

A recent discussion of a problematic SX-28 rebuild and align.

[KB2WIG]

B,

Try this site,

http://www.radioblvd.com/SX28Notes.html

There is plentyy 'o stuff about the SX 28.  There is a military manual floating around. If you don't have it, click on this link....

www.cryptomuseum.com/df/sx28/files/tm11-874.pdf

I had lots of fun with my '28.

Gud luck,

KLC

[N6YW]

Okay
After some reorganizing my assumptions about this receiver, I have found that not all is as bad as I previously thought. Maybe half of the bands are off, so it gives me the idea that I have the IF pretty darn close even though I haven't measured anything yet.
The Flex Radio does indeed pick up SX-28 frequency. So, I am building a capacitor coupled probe so I can see how close the IF really is. I am on the road tomorrow helping Dennis W6DQ move his monstrous collection of heavy metal transmitters from Fullerton to Inyokern Ca.
Mans game... Collins 550G, KW-1, Western Electric, KW Desk, ugh...
So, I will have to wait until the weekend to get my test setup organized. My new manual will have arrived by then, the military version complete with fold out schematics on laser printed quality paper stock.I am going to go to Staples and have them scan and print a large version mark up working schematic so I can make notes, use the high lighter etc.
You guys really helped to steer me in the right direction and I appreciate it very much.
Getting things organized really helps to format my brain.
73 de Billy N6YW

[N2DTS]

The sdr's are sensitive so no need to have a direct connection, just get a probe close an you can pick up signals to check LO frequencies.
For the IF, yes, hook a cap probe to the last IF an tune the whole thing to get the best response.

[W7TFO]

I bought a Rigol DSA-815TG spectrum analyzer w/tracking generator.

It does IF strips like nothing else I've ever used.  (everything else, too).

$1400, was worth every penny.

73DG

[N2DTS]

And you can get an sdr-iq for about $300.00 used and it makes a great piece of test equipment and its a good receiver. Goes down to 100Hz. Yes, 100 Hz!

I used a spectrum analyzer when making my home brew receivers and it really helped see how things were working.
Freq counters are erratic, need more input, can lock on a harmonic, and you can not see amplitude or shape of IF filter response.

[Opcom]

IIRC the AGC amplifier has its own tuned circuit on the IF. If it is not aligned right the AGC won't work right and signals can be distorted.

[N6YW]

I got it working pretty darn well although the crystal phasing and selectivity needs additional tweaking, I am satisfied at it's sensitivity. I enjoyed listening to the 75m AMI group tonight.
What a wonderful rig even though it's a beast to lift, I'm okay pairing it up with the BC-610-E.
Worth every bit of effort.
Bill Feldman N6PY (SK) was a great man. I used to participate in many a Collins Collectors Net QSO's discussing radio projects he was involved in and while it was most always focused on Collins equipment, he was especially fond of the SX-28 and wrote a very good method of aligning the IF, which I did. I think I have become better acquainted on how to do it.
Still, I have a lot to learn but the satisfaction of getting it up and running is quite nice.
The bottle of beer was well deserved.

[N2DTS]

Glad you got it working well.
The sx17 I had was very nice for an AM receiver, narrow selectivity was good for situations where things were close, wide was very good fidelity without being excessive, and the audio sounded great.
 

[Opcom]

How do you like the audio quality and power output?

[N6YW]

How do you like the audio quality and power output?

It does now that I replaced the crappy old output transformer with one that has an 8 ohm tap and better frequency response. When I get the R-12 speaker it will really kick butt!

[N6YW]

So, let's face it... a 70 year old cabinet that looks pitiful should get the sparkling finish that this iconic radio deserves.
Something worth considering for those of you who are contemplating repainting the cabinet of the SX-28 with a wrinkle finish. It's laborious and messy and requires a great amount of skill to do it correctly, not to mention a proper environment in which to do it.
I have two words: Powder Coating.
Many powder coating facilities are capable of applying a wrinkle powder coat finish that rivals the look and texture of the traditional finish. My cabinet was done down in Orange County in the city of Orange. They did an excellent job and it looks brand new, like it just left the factory. I was skeptical at first, but my skepticism was short lived once I saw it finished.
Another reason for going this route is the obvious advantage of durability, as nothing that I know of can compete.
The quality of the wrinkle is excellent and uniform and while I am good at painting the wrinkle finish, I must say I am very spoiled by this alternative method. I would suggest you find a local powder coating facility and take them a small project for starters to see how well they do. Perhaps an FB-7 or SW-3, and definitely the HRO dog house supply case!
I am sure you will be happy you did.