ad: w5yi

hallicrafters SR-2000 time delay relay?

Discussion in '"Boat Anchor" & Classic Equipment' started by WA6MHZ, May 20, 2013.

Thread Status:
Not open for further replies.
ad: L-HROutlet
ad: l-rl
ad: Subscribe
ad: Left-3
ad: K5AB-Elect-1
ad: Left-2
ad: MessiPaoloni-1
ad: L-MFJ
  1. K9AXN

    K9AXN Premium Subscriber QRZ Page

    Product detector/Mixer

    Good evening Jim:
    It’s all Perception, semantics, and loose definitions. And you are correct. A Product detector and Mixer are one and the same but the Product detector is a mixer designed uniquely for a particular purpose --- doesn’t have to be limited to audio but is assumed to be. Generally in the Ham world of legacy radios, it’s used to render audio from an SSB signal. It’s a mixer with one input being a fixed frequency and amplitude domain and second input containing variable frequency and variable amplitude domain content. Don’t believe I’ve heard anyone refer to Product detector in any other way than for SSB audio detection even though the very same design is used extensively throughout the SR-2000.

    There are two mixer designs in the SR-2000/SR400 and a balanced modulator for SSB generation. The first and only single element mixer is the 2nd xmit mixer and the rest are low noise triodes using the cathode and control grid.

    My definition distinguishing the difference between a Product detector and Mixer is this.
    A mixer has two inputs generally one fixed the heterodyne oscillator or BFO both frequency and amplitude being fixed and the second input, varying frequency and amplitude. This mixer type, the 2nd xmit mixer in the SR-2000 has two connections to a single control grid. The Heterodyne oscillator and 6.5mc generated SSB signal. In this configuration, if the 6.5mc signal level exceeds the heterodyne oscillator, severe distortion will occur. Same concept as vintage shortwave receivers where you had to vary the RF gain when listening to SSB to prevent distortion. This mixer in and of itself is a mixer not product detector. The Hallicrafters designers added circuitry around this mixer to prevent the 6.5mc SSB content from exceeding the heterodyne oscillator --- Fast and slow AALC. Looking at the mixer with AALC addition we have what could be construed a product detector only because the signal level was limited by external circuitry.

    The definition of a product detector to me is simply a mixer that uses two inputs where neither amplitude has a limiting effect on the other.

    A good example of the difference is the Hallicrafters SX-100 Mark1 and Mark2. The Mark1 behaves like a Mixer and the Mark2 like a Product detector when in SSB mode. What does that mean? The mark1 requires you to vary the RF gain dependent on the strength of the signal to prevent distortion and the Mark2, leave the RF gain at 10 and don’t worry about distortion. Sounds like a Product detector. What is the difference? In the Mark2 the BFO drive is increased so the signal will not exceed it, the AVC threshold is reduced, and delay time extended to further prevent the signal from exceeding the BFO signal. A dramatic difference. The Hallicrafters guys design limited and stabilized the signal. Is it now a Product detector? Does it behave like one? Depends on semantics, perception, and opinions.

    Hope this explains my use of the word Product detector --- maybe not for everyone. Thanks again for the question.

    Note: The SR-2000/SR400 series are designed using low noise Triode Mixer/Product detectors throughout with the exception of the 2nd xmit mixer.

    Have a super day Jim!

    Kindest regards Jim K9AXN
     
  2. WA6MHZ

    WA6MHZ Subscriber QRZ Page

    Struggling with the Balanced modulator today.

    Early this morning, I examined the SR-400, which has a nearly identical Bal Mod ckt, and took plenty of measurements.

    But I don't see anywhere NEAR those same values on the SR2K.

    At first I thought it was the Pot, so I tore that out and measured it. Nope, 250 ohms and good ohmage when rotating CCW to CW. No problems there.

    I see 150 Volts from the Regulator tube on both the anodes and cathodes of the 1N87 diodes. However, measuring the diodes, they are NOT shorted! C188, a .001, is also not shorted. On the anode of CR18, I see it switch from about 15V to 0V on RX to TX. Schematic says 16V to 1.5V.

    Scopewise, I see audio from the Audio stage on both sides of the diodes. Also see very low level RF from the carrier osc via the link inside T-4. It COULD be that the T4 Link is open, as I see 1.6K between pins 3 & 4 of T4. Think I will tear T4 out of the unit tommorrow, as that is a clue. The link should be very low ohmage, not 1.6K!
     
  3. AF6LJ

    AF6LJ Ham Member QRZ Page

    Sounds like you are on the right track.
     
  4. WA6MHZ

    WA6MHZ Subscriber QRZ Page

    I Yanked out the T4 and took it apart.

    OH HO!!!!!!!!!!!!

    I found that one wire in the T4 Secondary not only wasn't soldered, but had NEVER BEEN SOLDERED, from when it left the Hallicrafters Factory! The wire had been wrapped around the terminal, but NOT STRIPPED or SOLDERED!!!

    I was just about dancing around the table!!!

    I repaired it and now the T4 secondary was making good contact.

    HOWEVER, when I soldered it all back into the circuit, it STILL DIDN'T WORK!!!!

    No RF out of the Balanced modulator!!!!! WTF????!!!!!

    HAIR TEARING TIME!!!

    I will work again on it tomorrow!

    That should have fixed it for SURE!!!!!
     
  5. AF6LJ

    AF6LJ Ham Member QRZ Page

    Hmmm..
    That was interesting.
    Good luck, sleep on it, the answer may come to you.
     
  6. K9AXN

    K9AXN Premium Subscriber QRZ Page

    No carrier oscillator

    Hi Pat:
    Your getting close. You are correct. Both windings are under an Ohm. BTW that transformer is the same PN as T6 --- compare them.
    Measure the output of the carrier oscillator at test point D, the cathode of V9A. It should be approximately 5 to 7 Volts PEP. Even if the link is open you will see output at test point D --- try it in both USB and LSB mode. Note that if the slug has been turned it simply may not be oscillating. The adjustment is as in the manual --- turn the slug until the output peaks and begins to shut down. Then turn it the opposite way to 80% of its peak reading. When you hit the peak and turn it further, one direction will almost immediately stop oscillating and the other way will gradually reduce the output. Set it on the side that gradually diminishes, down to 80%. If all else fails switch to SSB LSB mode and vary the slug back and forth to see if it will oscillate.

    The 15/0 volts on the anode of CR18 would mean that it is shorted. You should see 6 volts receive and 1.8 volts transmit. Something is wrong with the measurements. Note the 16/1.5 volt levels are +16 RX and +1.5 volts TX

    The 16/1.5 volt line is the cathode of the 6AQ5A. It is used to switch from RX to TX. The 6AQ5A is run at over 400 degrees and has a tendency to weaken over time resulting in a 15volt RX and 1.5volt TX level rather than 16/1.5 volts. Even at 15 volts it will work ok but the 0 volt level that you found on that line is troublesome. If it is correct, there is a strangeness with C146 and or R101 which are attached to the cathode of V15 or V15 is bad.

    After you get the radio working would suggest adding a heat dissipating shield to the 6AQ5A. Will bring the envelope temp down to around 275 degrees --- will outlast you after that.

    A good odometer for the radio is how black and burnt up the T/R relay case is. Its next to the 6AQ5A.

    Good luck --- Kindest regards Jim K9AXN


     
  7. WA6MHZ

    WA6MHZ Subscriber QRZ Page

    Thanks, Jim!
    I DID verify that the voltage was 6V/1.5V on the CR18 anode. Was trying to use an old VTVM and this time measured with my lab DMM that I can safely use on LOW voltages. This is my company meter and If I blow it up, the boss will have KITTENS and ban all junk radios from my desk. So I use the VTVM for any voltages that are going to be over 100V.

    I messed and messed around with the Balanced modulator, and couldn't still get anything. But, looking down the road abit, on the Crystal filter, I DID get a proper SSB signal! Maybe my scope probe was loading down the Balanced modulator! However, it worked FINE on the SR-400 balanced modulator which has the nearly same circuit. SO not sure what is happening there.

    SO I traced the SSB signal through the Filter up to the grid of V2, where I see a 1.65Mhz signal that looks like a proper SSB signal. I could even null the carrier somewhat with the scope on that point, and do see a big signal on the TUNE position.
    TUNE: 1.24V P-P
    SSB: Up to about 3V P-P SSB signal looking correct.

    On pin 7 V2 Cathode), where the VFO signal is injected, I see a low .7V P-P signal at 4.5Mhz

    On pin 6 (V2-Plate) I see a fine signal on RX (about 2V P-P at 6.5Mhz), and a 4V P-P signal on SSB TX. However, the SSB signal isn't correct anymore. It is mostly carrier and just a little modulation peaks.

    Something is fishy with T1. Someone has moused with it. The top slug is almost all the way out of the shaft, and is a very loose fit. it seems to have lost the grooves inside for good turning. The lower slug seems to be about where they should be. Both peak the RX signal OK, and Receives seems normal though I haven't made any sensitivity measurements.

    Since T1 is used on both RX and TX, if it was really bad, then RX would be messed up too.

    I traded V2 with another 7059 to see if that was squirrelly, but it acts the same. Doesn't seem to be a tube socket problem, but I can tell the socket has been reworked some. All resistor values seem the same.

    By the way, I did find that someone had changed R98 in the carrier Oscillator circuit from a 6.8K down to a 1.8K! Maybe they thought they weren't getting enough Carrier signal. So the previous person who has worked on this and gave up has been down the same road I have. I changed the R98 back to a 6.8K as per the schematic.

    I do have an IECC Heat Shield on the 6AQ5A now as per your suggestion. The side of the Relay case was well burnt!

    So progress is stymied at this point. Not sure why I can't view the Balanced modulator sig, but if I do see it downstream, then it must be there and OK.

    73 for now. Time to put it back in the case for another week. Boss wants the old radios off the desk for work tomorrow!
     
  8. K9AXN

    K9AXN Premium Subscriber QRZ Page

    Hi again Pat:
    Good results. Would like to see 1v PP from the VFO but .7 will be just fine.
    The output from the balanced modulator will be about .7v PP. If you place your 10x scope lead on the output of the crystal filter you will see 1.5v PP in tune mode which is good.

    The 6800 ohm resistor was the last owners desperate effort to fix something that didn’t need fixing --- good choice to restore it back to 6800 ohms.

    T1 is your nemesis. T1 and T2 are super critical for the operation of that radio. As you know by now they are the 6.5Mc broadband system along with V3. Here is a quote from Bob Orwin, chief engineer for Hallicrafters regarding the adjustment of the 6.5Mc I.F. system. Do not touch T1 or T2: adjusting those transformers is for experts only.

    If the slugs are loose use tacky wax --- use it anyway, it does not harden.

    The factory adjustment process used sweep analysis which I also used for years. I’ll say it like it is @#$%^&. Takes a lot of schmoozing. Adjusting broadband amplifiers historically has been tricky when using sweep alignment. Not going to explain the details here but will explain them fully on my web site when time is available.
    I’ll probably get some flak for that statement but if you haven’t tried sweep analysis for broadband filters try it: even Collins uses swamping not sweep analysis.
    On my web site under service notes you will find the sweep analysis and swamping techniques both somewhat difficult. The swamping process is by far the easiest.
    Since I wrote those notes I have developed a super simple and field tested procedure. It is easy and results are as good as sweep analysis.


    1. Remove V12 the Heterodyne oscillator tube.
    2. Bands switch 14Mc.
    3. Frequency dial 14.050Mc.
    4. Scope probe 10X on V11AP2. Will be looking for 6.5Mc at 3V PP.
    5. Tune mode.
    6. Go to transmit. Don’t worry about Xmit RF settings.
    7. Adjust T1 and T2 top and bottom slugs for peak.
    8. Set frequency dial to 14.450Mc
    9. Repeat step 7 noting which slugs need adjustment.
    10. Set frequency dial to 14.050.
    11. Repeat step 7 at 14.450 and 14.050 until the adjustments do not need changing.
    12. If you have been successful you will measure approximately 3 volts PP and when tuning the VFO from end to end the signal will not exceed 3db of rippple.
    You may wonder what we just did. That broad band system is made to be 500Kc wide with a 3db skirt by adding a 2200 ohm resistor across each coil winding, and primarily using tight coupling. Tight coupling causes the output to look like a Bats wing --- the wave shape rises at the beginning droops in the center and rises again at the end.
    What you just accomplished is peak the wave shape at the normal 14.050 and 14.450 locations which are the normal high points of the peak droop peak shape.

    There are photos of the waveform when properly adjusted in the service notes section of the K9AXN web site.

    Good luck --- Kindest regards Jim K9AXN
     
  9. KB9LXP

    KB9LXP Ham Member QRZ Page

    You, sir are a Radio God!!! Best of all, I followed everything you said Jim!!!
    Best of luck, Pat....your in good hands.
    Mike



     
  10. WA6MHZ

    WA6MHZ Subscriber QRZ Page

    Last night, I scored TWO 8122s on Ebay for $76. Not sure if that is a good deal or not. The existing tubes in the SR-2000 are in completely UNKNOWN condition. While my problems are WAY WAY downstream from the finals, it will be comforting to know I have a 2nd chance if it turns out the current 8122s are bricked. Some sellers were getting $100 each for these tubes, so $76 for 2 isn't too bad (IF THEY ARE GOOD!!!). Of course, I COULD Have just bought 2 tubes in WORSE condition than the ones I already have, won't know until I get the rig working. The seller CLAIMS they are tested and good. Guess I will find out.... Someday.

    The last few weekends have seen priorites set the troubleshooting of the SR2K back. Hams brought rigs to the Radio Hospital for repair, which ate up most of the day. Then 2 new Yaesu FT-101Es came in and were processed. More Hams are coming for more Radio Surgery this weekend. So life is very BUSY here in the Radio Hospital. If ONLY I could work on these during the WEEK!
     
Thread Status:
Not open for further replies.

Share This Page