[Fwd: Re: [R-390] A little help on sensitivity - where to look]

[James Cottle]

-------- Original Message --------
Subject: 	Re: [R-390] A little help on sensitivity - where to look
Date: 	Sun, 02 Oct 2005 16:04:06 -0700
From: 	James Cottle <jim_cott at earthlink.net>
To: 	Flowertime01 at wmconnect.com
References: 	<7e.72a5e6e6.30717b6f at wmconnect.com>



Roger,
  Thanks so much for your lengthy and helpful email. I did a little 
inspecting today and found two serious problems (one which is my 
original and probably one new one). The first was a single slug in one 
of the tuning racks that had broken free of its screw, meaning it was 
always stuck in the bottom of the coil form. The only place this was 
supposed to be the case was on the 1 MHz  megacycle setting. I think 
this was my major problem, so a little glue got it back to where it will 
track with the other two. Secondly, today I could not get very much at 
all to work...then I noticed that the first 6BA6/5749 (V501) looked 
really cloudy. On pulling this tube, it had cracked diagonally across 
the base of the pins. I would bet that even before its catastrophic 
failure, it wasn't all that good. So NOW I am all set to proceed with a 
proper IF and RF alignment. I am sure that I will have a healthy beast 
when done. Your swapping tips are greatly appreciated, for these are not 
common knowledge, even in the technical manuals. Thank you for your 
tips. It is truly a wonderful beast to have in tip top shape.
James
AC4EA
San Francisco

Flowertime01 at wmconnect.com wrote:

>Jim,
>
>I put this mail up last week to a similar question.
>
>Do you have a copy of the Y2K manual? It is on line and a must have read.
>
>Run the cal tones and tell us what bands have tones. You either have a tone 
>some where in a 1MHz band or you do not. If tones are missing on one or more 
>bands, then a band problem is to be resolved. If you are missing tones on whole 
>octaves
>(.500 -.999 ) (1-1.999) (2-3.999)(4 - 7.999) (8 -15.999) (16-31.999) Then you 
>are looking at a set of octave transformers, slugs and caps. If you are 
>missing two or more octaves, then you should look at the band switch.
>
>Likely you have enough AM stations to get through some weak tubes. The rest 
>of the receiver range just does not have enough antenna and signal to drive it 
>to the output. From day one the most likely problem was just plain old tubes 
>that have reached the end of their useful life.
>
>You may need to check your RF band switch if the receiver only works on the
>.5 - 1 and 1-2 octaves. Pull the RF deck and do the band switch by eye ball. 
>All the switch sections are not exact and you want to get the best possible 
>adjustment on all the octaves. The switch contacts carry current. If you get one 
>section just barely making contact, over time the contact will burn. You want 
>a good alignment when the octaves are changed going both up and down in range.
>
>Check your cam adjustments at 7 +000. The RF and IF slugs should all be about 
>mid range. If any are at the upper or lower limits, some real good alignment 
>is in order. You may need some tubes in the RF deck if the cam alignment is 
>good and the slugs are at the ends or there range.
>
>Now you get to checking tubes and RF alignment.
>
>Once you get to the point where you have a calibration tone every 100 KC you 
>have
>a working R390/A. Between a working R390/A and a wonderful receiving R390/A 
>is a range of work. A working R390/A and a wonderful receiving R390/A should 
>not be confused with a good looking R390/A. Each of these are different.
>
>Good looking R390/A are selling for over $1000.00 on Epay and may not work at 
>all.
>
>A good working tube and a real good tube is a range that a tube tester will 
>not evaluate for you. Old used tubes can work better than some new tubes. I do 
>not want to send you out shopping for bunches of new tubes, because what you 
>buy may not in fact be any really better than what you have. All tubes are not 
>equal. Swapping the 5749s around will change your signal to noise. Changing 
>the 6C4s will also make a difference. Finding a good 6DC6 can be a treasure 
>hunt. Swapping the 5814s around will make differences. The 6AK5s or 5654s also 
>make differences. Start looking for tubes for your receiver. Accept what you can 
>find when you find them. If you really need some tubes because some are just 
>bad and you have nothing else to use, then buy some new ones. Hopefully not 
>from Radio Shack. RCA and Sylvania are good. Other good brands are also around. 
>Old JAN tubes are likely OK.
>
>Then comes alignment. Do the mechanical alignment of the dial over run and 
>set the zero adjust to center before doing the RF cams. Once through the 
>mechanical alignment of the RF cams will get you OK. 
>
>Once the PTO is set you are OK. You can set the PTO against WWV and zero the 
>cal osc to WWV. The mechanical coil alignment of the RF deck will improve with 
>as many as 4 passes. If you change any RF tubes, 6C4, 6DC6, a realignment is 
>in order. The Y2K manual will get you through that OK.
>
>A signal generator and a volt meter will get you through any alignment you 
>need.
>
>Truth is that the cal tones and a volt meter will get you into good alignment.
>A good frequency counter that lets you set the PTO and BFO is a blessing. Not 
>required, but use it if you have it. 
>
>Real good reception comes from just swapping tubes into the same socket and 
>evaluating the results.
>
>As my ear is not calibrated, I use a signal generator and a volt meter. I 
>like my signal generator as I can turn the modulation on and off. This compares 
>CW to Modulated, AM. Some (Military) like to call this signal to signal plus 
>noise test.
>
>I hang a 580 ohm (600 Ohm) 1 watt resistor across the local output on the 
>terminal board with an AC volt meter. My AC voltmeter has a DB scale. A good 
>receiver will put out 1/2 Watt so you need a 1 watt resistor or spares to burn.
>
>Using any signal frequency you can and start swapping tubes for comparison. 
>Run all the tubes through a tube tester some where just to get the shorted ones 
>discarded. Watch them all to get out the ones that glow blue. In the receiver 
>circuit thump them all a time or two to get out the microphonic ones.
>
>Get all your 5749's or 6BA6's and sub them one at a time into the first IF 
>socket. Using the same level of signal generator input. inset a tube and compare 
>the audio output level with the signal generator modulation turned on and 
>turned off. 
>
>A good (acceptable) receiver will have 10 DB difference. A nice (up to 
>military spec) receiver will have 20 DB difference. 25 is very doable in today's 
>R390/A even with their age. 30 has been seen on many receivers and can still be 
>achieved today. (You may spend more on tubes and caps than you paid for the 
>receiver to get there.)
>
>So sticking several tubes into a socket and comparing them to each other you 
>can judge them for noise. Put the better performing ones into some of the 
>other tube sockets. Set the test up again and compare the tubes you pulled. Find 
>the best of what you have and use those tubes. Doing the 5814's needs two test 
>to get each side of the dual triodes. Run the 6AK6's in the last IF not the 
>audio deck. Swap the 6C4 into the second mixer above 8MHZ. If you are not 
>blessed with tubes, Just buy your self a new 6DC6. Compare it to the one or ones you 
>have and write that down some where (on the tube box side) so you can judge 
>it again at some later date.
>
>The Army ran these receiver 24 x 7 for six months or 4380 hours. Tubes would 
>go for a year or 8760 hours. We would check all the tubes every six months and 
>swap out the poor ones to get the receiver back up to minimum of 20 DB signal 
>to noise at 1/2 watt output. Also need 4uv sensitive to get the 1/2 watt. If 
>you do not have a calibrated signal generator this means nothing. It also has 
>no impact on your ability to get your R390/A working very good. You can 
>compare tubes using the Cal tone and BFO on and off. It will let you compare the 
>same tube type in same socket and judge them from best to not best. You can them 
>insert the best to the front end and work down the line from there with what 
>you have.
>
>If you have a signal generator that puts out 150 uv at 455 you can get the IF 
>deck into shape real fast. You need 150 uv in to the IF deck by moving the IF 
>out jumper wire over to the IF input and feeding the BNC connector on the 
>back panel. Set the RF gain adjust on the IF deck to -7 volts on the diode load. 
>Set the RF gain to mid range and adjust the generator for -7 volts on the 
>Diode load. Set the band switch to .1 setting and rock the signal generator 
>frequency for maximum signal level through the 455Khz crystal. You can zero the BFO 
>against the signal generator. This will get you amazingly close. If you think 
>you generator is close, set it to 150 uv and then set the RF gain to -7 volts. 
>Set your generator modulation to 30%.
>
>Open the IF bandwidth back up to 2KC. Turn the BFO off. Start swapping 
>between modulation on and modulation off, you need a 30 DB difference. If you do not 
>have this 30 DB difference in the IF deck and Audio deck string, you will 
>never get a 20 DB difference for the full receiver. You will get 30 DB in the IF 
>deck and have the meter needle bumping around. This random noise will not get 
>it. You may get 28 and have a rock solid flat meter needle. This may be OK and 
>things are just not all that exactly calibrated. So accept this and get on 
>with life. Better tubes will come in the future. A bouncing needle may be a 
>leaky cap. More likely its a noisy tube. Over time and repeating these test, you 
>will get a feel for what is passable. Just get your receiver as good as you can 
>with what you have today. Enjoy it.
>
>The IF front end 5749's will make the most difference. The 6AK6's are next in 
>order. The 5814's will also make a difference. You can swap poor ones into 
>the line audio path. Some 5814's are in the limiter, 455 cathode follower, and 
>AGC circuit. You can get the better tubes into the critical path and put some 
>of the others elsewhere until you can find some better tubes.
>
>Once you are getting 30 DB in the IF deck and audio string, you know that end 
>of the receiver is good. You can move over to the RF deck. In the RF deck you 
>run 4 uv into the antenna input and look for a 20 DB difference between 
>modulated signal generator to unmodulated signal generator.
>
>You can set the IF band switch to the .1 position and rock the generator 
>frequency into the band pass. Once you tweak over peak, you can roll the KC knob 
>for maximum signal point. Doing the cap or core slug alignment in the RF deck 
>within 50KC of the specified number in the procedure will not cause you any 
>grief in the final alignment and signal to noise ratio. Get the generator close 
>and use the KC knob to get max signal. Then do adjustments.
>
>When you are done, go find the Chuck Riddle RF gain setting procedure and use 
>that to get the best receiver sensitive setting.
>
>Swapping tubes in the RF deck will make improvements. When comparing tubes, 
>do not try to align the RF deck for each tube. Just plug what you have of each 
>tube type into the same socket for comparison.
>
>You will get real hot great gain tubes. Some will have lots of noise. Some 
>lower gain tubes will give better noise ratios. Its just a plug and try process.
>
>Once you get the best of what you have sorted and the best moved to the front 
>of the line the receiver will improve in sensitive. Do the RF deck alignment 
>more than once over with a set of tubes. It will make a difference.
>
>If the R390/URR receiver TM there is a procedure to feed both sides of the 
>balanced antenna input from one signal generator wire through 2 each 68 ohm 
>resistors, one to each side of the balanced antenna input. This set up lets you 
>adjust the first antenna cap in the octaves of the RF deck. Any resistor pair 
>between 50 and 120 ohms has been shown to work for this test setup. Grab a pair 
>of resistors and do this alignment on your receiver at least once. While any 
>single ended input setup will show no difference in output or signal to noise 
>having these caps balanced will improve the minimum signal you can hear. If you 
>ever get to feed the receiver from a balanced antenna, you will want these 
>caps adjusted. The R390/A URR manual has never covered this alignment procedure.
>
>I do use an antenna match box and band pass filter with my R390. It has a 
>balanced output into the receiver. So I do this alignment.
>
>You have a working receiver. It works on some bands. What it needs now is 
>within your ability to provide. The US Military taught thousands of guys and some 
>gals to convert that receiver into an up to spec receiver with under 4 hours 
>of hard work using only a screwdriver, spline wrench, volt meter, 600 ohm 
>resistor, signal generator and one hand behind their back for safety. It took the 
>instructors 50 weeks to teach the required theory and mechanical skills to 
>each student. It took 40 hours to teach someone every thing the military mind 
>knew about an R390/A and what was needed to service any problem that Receiver may 
>ever have. Two Instructors taught me what I needed to know as part of a class 
>of 10 guys. I used what I learned to fix receivers for 8 years back in 68 to 
>75. I have still not found a R390 problem I could not isolate and fix. The 
>credit goes to the receivers. The engineers at Collins did an awesome job. For as 
>many parts as these receivers have they are still flat reliable and simple. 
>There is not any problem in that receiver you cannot fix. Ask hear on the R390 
>reflector and you will more help and humor than you will need to get er done.
>
>Roger KC6TRU
>
>_____________________________________________________________
>R-390 mailing list
>Home: http://mailman.qth.net/mailman/listinfo/r-390
>Help: http://mailman.qth.net/faq.htm
>Post: mailto:R-390 at mailman.qth.net
>Unsubscribe: http://mailman.qth.net/mailman/options/r-390
>
>  
>