AIDC-3733 2.4 Ghz to 144 Mhz IF after Xtal Change This is a low noise (they have been measured at 1.3 dB noise figure and 37 dB of gain), synthesized LO down converter that will work as is for 2401 MHZ Oscar AO-40 (previously known as Phase IIID). The LO is 2278 MHZ and the 2401 MHZ output is 123 MHZ. The converter uses tunable comb filters on the input and output and has two stages of low noise preamplification. This unit, by itself, beats the sensitivity of my Drake 2880 downconverter connected to a Down East Microwave Low Noise Preamplifier. The LO synthesizer is the standard divide by 256 that uses crystals in the 8-9 MHZ range. The crystal is plug in -- no soldering to remove it, in case you want to change the IF output to 144 MHZ. The unit is designed and specified for an input frequency of 2150 to 2686 MHZ operation and requires no retuning of the comb filters to work on 2401 MHZ. The IF amp is designed for output from 116 to 408 MHZ and measures flat over this frequency range and to 500 MHZ. It requires no modification to work on 123 or 144 MHZ or anywhere in the frequency range. The downconverter has a built in dipole and reflector for a dish feed, this combination eliminates another fraction of a dB loss over a separate feed plus cable. The downconverter unit is 1 1/4 inches square and 12 inches long. Voltage requirements are 15-24 volts through the F connector. 123 MHZ IF. The dish is a section of a parabola that is 38 inches high by 26 inches wide, it is painted white. The dish can be mounted horizontally or vertically. These Units can be bought from: Robert Seydler his Email is: k5gna@aol.com These are Emails from the amsat mailing list for the AIDC-3733 The crystal is divided by 256 for the synthesizer. Currently it has a 2278 LO, for 122 MHZ output for a 2400 MHZ input. If your radio tunes to 122, I wouldn't bother changing it. For a 2 meter IF, the LO is 2256 divided by 256 = 8.8125 MHZ for the crystal. The chip marked 510 is a Fujitsu MB510 prescaler. This chip can divide and incomming signal (up to 2.7 GHz) by 128,144,256, or 272 times. This divided signal in the 8 or 16 MHz range (with a 2.3 GHz signal in) is compared to the VCO and locked by the HC4046. What does this mean? It means that you have a choice of four crystals to achive the desired LO frequency. Here is a list: For a 2256 MHz Local Oscillator: 128x 17.625 MHz 144x 15.66667 MHz 256x 8.8125 MHz 272x 8.294118 MHz The desired division ratio is selected by putting either ground or Vcc on pins 3 and 6. More information on the MB510 can be found here: http://www.fujitsumicro.com/pdf/mb510.pdf?sec=prescalers Look around in the junkbox and see if any of the crystals that you have might work better with a different divide ratio. Some of the components in the 8 MHz oscillator might need to be changed when using the other divide ratios. The loop components might also need tweaking if going to the 16 MHz ones. Keep in mind that most crystals higher that 20 MHz or so might be overtone crystals. A crystal marked 28.4 MHz might oscillate close to 8.8 MHz. I have not tried the CB crystals sold at Radio Shark. Another interesting point is that the phase noise can be improved by several dB's by changing the multiplication factor. The calculated degradation of a PLL of this type is 20 x log(multiplication ratio). This value is theoretical and you actually get more degradation. This means that the degradation from the phase noise of the reference will be: x128 = 42.1 dB x144 = 43.2 dB x256 = 48.2 dB x272 = 48.7 dB >From these numbers it is possible to improve the phase noise by 6 dB by going to a 17.625 MHz crystal. Improvements will be less due to slight degradation of the Q of the crystal going from 8.8 to 17 MHz. Another very important note is that the bandwidth of the bandpass filters have passband at both the desired frequency (LO+IF) and the image (LO-IF). So when using a 2256 LO and receiving a signal at 2401, you are also receiving at 2546 MHz. This undesired signal or noise power passes through the bandpass and ends up as noise at 145 MHz. This will create a 3dB error in the noise measurements. LO leakage can also degrade these measurements. The chip marked R24 is the VCO transistor. The wavy trace on the lower left pin is the 2278 MHz resonator inductor . It is bypassed to ground and has a 220 ohm bias resistor. The 1PZ chip is the varactor . The varactor is fed by a 10K resistor marked 103. The round test pad to the right of the resistor is used to test the lock voltage. Idealy you want the voltage on this pad to be the same before and after the conversion of the LO frequency. I have not measured mine but it needs to be in the middle of the voltage rails so that the PLL has range to compensate for the temperature and other drifts. On a locked PLL the voltage can be adjusted by changing any of the resonator components (varicap, series cap, etched inductor). The signal out of the VCO goes to an Agilent (HP) AT-41511 transistor. The datasheet is here: http://literature.agilent.com/litweb/pdf/5965-8929E.pdf Most of the resistors around it are for biasing. The three resistors after it are a small attenuator pad going into the mixer. This pad can be removed to make a stronger signal source. The mixer diodes are inside the chip marked SH2. There are two diodes inside that package.The 2.4 GHz signal is feed on the upper left of the mixer and the IF comes out below the etched "U". This "U" looks like a LO notch or something. Receiver stages: The small coax at the input of the second filter is a notch filter. It can be adjusted to notch the image (around 2546) by cutting it. How much? You need test equipment. I would just remove the thing. The part marked 414 is an Agilent AT-414xx transistor for additional gain. The LNA board works on 12 volts and requires no negative supply. It can be used for a much better downconverter. Things I don't like or would do different: 1. I would remove the input filter. This would buy me better noise figure. Overload should not be a problem if aiming the antenna away from powerful transmitters.The second filter would be modified to filter the image frequency and the LO. For weak signal work it is very important to filter the LO at the receiving frequency (2401). The LO specially on these types of converters has a lot of undesirable noise at 2401. This degrades your noise performance. 2. The LO is a very poor performer. This design is just not suitable for SSB work! I would totally get rid of the LO and replace it by a 2256 design based on a 94 MHz multiplied crystal in an oven. The mixer should be ok. I plan on taking the converter to work this weekend to see if I can come up with reproducible modifications to improve this converter. I personally think that it can be modified to really perform well for mode S reception. I will keep everyone updated. Pieter N4IP From K5GNA: My experiment with he AIDC 3733 downconverter has been going for the last 11 hours and the downconverter has stabilized in frequency and of course, thermally. The frequency of the LO is within 100 HZ of where it was for the last 10 hours, after it had been on for about an hour. The drift rate is still very slow, cycling and moving only one to three Hertz per count (1 Hertz gate time) of the counter. There are none of the rapid, and large, frequency excursions as observed when the covers are off and the crystal and VCO are subject to air currents. The easiest way to measure the frequency of the LO with the covers on is by injecting a stable signal into the RF input of the downconverter and measuring the frequency at the IF with a stable counter. With a 2401 MHZ signal input, the counted output at 123 MHZ will indicate the frequency of the LO. Preferably both the signal generator and counter are locked to the same frequency standard. This test would be very hard to do with a free running signal generator. Unless you want to move the LO frequency beyond the limits of the lock range, I can't see the LO circuit needing any major fixes. The answer to the rapid frequency jumping problem that has been observed is thermal stabilization of the circuits. Just leaving the converter powered up with the covers in place is one solution. Bob K5GNA k5gna@aol.com Art and all: REVISITING THE GAIN MEASUREMENT: I measured the gain of a few more downcoverters using various tests. The gain that you measured at 2400 MHZ that was about 15 dB low can be attributed to the 2400 MHZ stub and possibly the RF tap's insertion loss if you were using one. Cut the stub, and the gain will be closer to what is indicated on the downconverter. After checking a few downconverters, I found that the stubs can notch right on 2400 MHZ or far enough off to not affect a measurement at 2400 MHZ. Some showed proper gain, some were up to 10 dB low at 2400 MHZ MEASURING THE RF TAP POWER INSERTER'S INSERTION LOSS: I measured the through loss on various taps on an HP 8753A network analyzer -- the loss is between 3 and 4 dB AT 123 MHZ and 144 MHZ. These numbers were also duplicated with an HP 8660C signal generator and an HP 436A power meter. I checked the ones for Pacific Monolithics, the Emerson and ones made for these downconverters -- they were all within this range. With the stub's insertion loss at 2400 MHZ and the RF tap's insertion loss at 4 dB, your gain numbers could be 15 dB or so down. This loss would depend on the stub's actual insertion loss at your frequency of measurement (try the same test at about 50 -150 MHZ higher) and whether or not you were using an RF tap. Once again, the stub's loss occurs after 99% of the NF has been determined and does not really affect NF. It only reduces the amount of attenuation you will put in a VHF receiver to get the noise level to a reasonable level. However, as I previously posted, it is an easy task to cut the center conductor of the stub's connection at the preamp's output and the hole goes away. TNX & 73, Bob Seydler k5gna@aol.com Mark, I posted a message on the BB on 4/3, here is an excerpt: 2. Notch in response at 2400 -- there is a 5-8 dB notch in response at 2400 MHZ or so on some of the converters. The little piece of rigid coax that is on the output of the preamp section (before it goes to the second comb filter) appears to be the culprit. It connects near the rear comb filter and runs up above the circuit board to near the front comb filter. It appears to be a 2.4 GHZ stub. I cut one at the connection to the preamp output, and the gain is now flat across the range. THE CUT IS MADE ON THE CENTER CONDUCTOR OF THAT COAX. I have a picture of this part of the downconverter on an Ebay auction. To view it, go to ebay.com, search for auction number 1227956478. It is the upper right picture, click on it to make it larger. The left side of the stub (shorted end) can be seen sticking up at the bottom by the open front end filter (left part of the picture) and the connection that needs to be cut is connected near the little green thing on the output side (right part of the picture) of the preamp where it connects into the second filter. I hope the picture and description helps. Bob Seydler K5GNA k5gna@aol.com Joe, The stub is more of a measurement problem than an operational problem. When people have measured the gain or the response, they can see the hole in the response in some downconverters at 2400 MHZ. Some downconverters have the hole somewhere else where it doesn't affect the measurement at 2400. They think they that it is broke because the gain doesn't read what it says on the cover. As far as NF goes and downconverter performance, the stub has very little affect. As long as you have no great losses from an excessively long feed line, the gain actually needs to be reduced at the receiver. I use the 10 or 20 dB pad in my IC R-8500. These downconverters are actually made for a fairly long run of RG-59 cable. That is why people are seeing S meter readings of S-5 and above in their receiver with just the added converter noise. Remember TSI makes these things with 3 different IF gains that give 17, 30 and 37 dB of total gain. For weak signal work, the gain should be low, and your S meter reading should be low with the converter is turned on. TNX & 73, Bob Seydler K5GNA k5gna@aol.com I obtained 2 aidc3733 and tested one with the antenna element, since the N connector was not yet available from Bob Seydler, and here are my results: Sweeper and antenna: hp87610 into a 3 inch wire, about 4 inches from antenna. Adjusted drive at 2500 mhz for 10dbm output, this is the point where it just starts To compress. Step each 5 mhz and plot results. Power meter gigatronics both under pc control with labview, done at work after hours. IF 40-430 MHz. RF 2080-2710 band pass filter. 2395 mhz: -12.5db relative notch Instead of a notch at 2400 we need the notch at 2150 to eliminate the image noise from the preamp and whatever comes in the antenna, because the LO is in the passband. The easiest thing for me to try is put an 8 mhz xtal in the LO and see if it will lock at 2048 mhz. If it does this puts the image noise out of the bandpass filter. Makes 2400 come out at 322 mhz which is fine for me because the recievers are icom7100s. A 2256 LO with a 8.8125 mhz crystal would put image noise (2256 -2080 = 176 mhz still in the 2m passband but a 432 crystal 2400-432=1968 is way below where it may lock up. 7.6875 mhz. I have 8 meg xtals so thats what I am going to try. After I get the N connector I can put it on the big antenna, I may just switch between that one and my old one to retain the wide freq range of the old LNB (IF 950-1450 LO 3650). May have to put an rf relay out there and stuff the additional converter out there. I am testing the aidc3733 to see how far the LO will move and still stay locked up, because even after the 2400mhz notch open stub is removed(unsolder the wire), I dont want the image noise in the passband and want to get 2.2-2.3 ghz for space shuttle and other commercial satellites like tdrss ranging beacons. I connected a power meter to the converter output thru a dc block. Sweeping it gives the following: Note the dropout right where we want it to work 2401: Siggen needed -8.5 dbm to get 10dbm out at 4 inches away. freq dbmoutAnt1 dbmoutAnt2 1995000064 -10.113 -10.902 2020000000 -5.468 -5.815 2044999936 0.432 -0.173 2084999936 7.622 8.245 2120000000 8.595 8.421 2150000128 7.921 7.837 2195000064 7.329 6.518 2240000000 7.858 6.489 2255000064 -1.121 -2.65 2260000000 -7.914 -9.556 2264999936 -18.211 -19.265 2270000128 -19.886 -21.44 2300000000 -2.294 -3.293 2304999936 2.866 2.072 2310000128 6.064 5.27 2315000064 7.051 6.346 2335000064 6.534 5.781 2350000128 6.559 6.145 2375000064 4.845 4.639 2384999936 1.957 1.956 2395000064 -2.557 -2.569 2404999936 0.809 0.635 2415000064 4.863 4.923 2430000128 8.186 8.024 2440000000 9.383 9.143 2455000064 10.098 10.132 2484999936 10.066 10.219 2500000000 10.138 10.016 2560000000 5.981 8.242 2620000000 7.18 8.11 2695000064 5.505 4.839 2720000000 4.291 2.738 2750000128 -1.398 -3.891 2780000000 -7.779 -10.822 2840000000 -19.095 -20.589 ----- Original Message ----- From: "Mark L. Hammond" To: "amSAT-BB" > http://camel.campbell.edu/~hammond/ham/3373_stub_cut.jpg > > Hope this is right since I already cut two of my units here ;-) > > On the first unit, I can see nearly 3 S-units improvement on 3' dish > (outside) from my harmonics generator in the shack! Wow...quite an > improvement. Not very "technical," but it's the only way I can test. You still have the image noise there too. The noise will go down if you can notch out the 2150 freq assuming you still have 2278 LO. NN0V, Your measurements are accurate, they match mine and others who have made these measurements and posted them. The two preamp filters are designed to pass about 2.1 to 2.7 GHZ and the IF amp works from around 50 to almost 500 MHZ. The notch can be at 2400 MHZ or nearby. Others have made quite a few comments on this BB on changing the frequency of the LO. My long range plans are to change the LO to 2.00 GHZ on a downconverter. With this, the 2.05 to 2.50 GHZ input signal would come out at 50 to 500 MHZ on my receiver. This takes care of the image problem and gives me the extended range on my RX. If you want the notch at 2150 MHZ, then the small rigid coaxial cable that makes the stub needs to be replaced with a slightly longer stub. You can then cut it until the notch is were you want it. If you want the notch higher in frequency, then the stub needs to be shortened. I have measured the stub, the cut end is not a short, it is an open circuit at the cut end. Obviously you need some good test equipment to change the stub's frequency. The two bandpass filters have tuned elements that are "fingers" that are various lengths, to provide the bandpass characteristics. The longer ones could be cut at the ends to shorten them a little and to bring the lower frequency bandpass limit up a bit. The elements could also be replaced with new ones, cut out of thin copper, with all the elements cut to around 2.3 -2.45 GHZ. Even a small single sided circuit board could be made to duplicate these elements and be designed for passing 2.3 to 2.45 GHZ. These changes would take a lot of time and effort and some good test equipment to implement. Most people will find it works OK as is. TNX & 73, Bob Seydler K5GNA k5gna@aol.com >From K5GNA, After a few inputs, I think that my original comments on unsoldering the stub can be ignored. Cut it. First look at Mark Hammond's posted picture of it and then cut it with a very small cutter. ============================================== I got quite a few suggestions about the diode/resistor multiplier for a 2.4Ghz signal generator. Thanks to all. This is somewhat lengthy, but hopefully it'll help others who might be struggling with similar lack of equipment. The diode multiplier is this one: http://members.aol.com/k5oe/signal_gen.gif That image can be found on K5OE's pages under "Drake 2880 Mods" http://members.aol.com/k5oe/ Hi all, The AIDC 3733 converter can NOT be powered with 12-13Vdc, or whith the "preamp" supply of a transceiver,without a previous modification: the voltage is internally regulated with a 7812 regulator. Such regulator have a voltage drop of about 1.5-2 volts, the resulting voltage will be inaceptable for the 3733 to work properly. I began to make experiments whith the diode Signal Generator on 50mhz,and the FT847 supplying directly the 3733 via the coax. I noticed on 123.1 mhz a strong noise ,but not at all the clear signal expected. Then i measured the voltage at the F connector of the converter, it was about 11,5V , because of the loss in the coax cable (only a few meters). At the imput of the 7812 regulator(inside the 3733),it was somewhere around 11.0 V,(loses in the coils) and at the output 9,6 V. Desoldering the coil at the input and suplying 13 V directly bring back to life the converter,with a clean signal, but the voltage on the output of the 7812 regulator was still around 10.6V, and consequently, not regulated. I connected it on my spectrum analyser, (it suplies the converter via the coax), and the signal was nice but with a fast and quick (small) oscillation. This oscillation did not *aparently* degrade the reception quality,the sound from the SG seemed clean. I then suplied the 3733 with 18V and the signal becomes about 40% stronger,NF improved a lot and it stopped oscillating. here are some solutions for this problem: -Supply it with 15-18V via a bias tee.Probably the best. -remove the 7812 regulator and short in an out pins (the supply will not be regulated any more, that may increase the drift,it depends of the quality of your power supply.) This way, you can try to use the "preamp" power suply of your transceiver,but be carefull with the (DC) loss in the cable.Check the voltage in the 3733 with the lenght of cable that you will use.(at the output of the 7812.Should be 11.9 V) -Make a hole in the cover or side of your converter,fit a feed thru capacitor,short the in/out of the 7812 regulator and supply it separately with 13V. I hope that will help.I had to remove my DEM transverter because the "weatherproof" box i made stopped its mains function (yes,stopping water),and instead to repair it i consider to try the AIDC converter.The specifications seems better anyway. I have fixed the 3733 directly to the feed,soldering the 31,5 mm probe directly to it. I hope today i will be able to get telemetry again. Hoping that will help, Best 73 and good luck. Fred, CU8AO Flores Island.