|
Re: Re. problem with crystal oscillators
Arv Evans
开云体育Chris, & others...Please take a closer look at the BITX IF string without any crystal filter.? There would be no control of the passband and thus no way to tune a specific frequency, other than as a very broadband DC RX with a converter ahead of it.? There would also be little rejection of image frequencies.? The crystal filter is the only frequency determining part of the IF amplifier string, so at least some minimal bandwidth control is probably necessary.? I have to admit that my first working receive-only mock-up of the BITX20 did use a pair of 10.7 IF transformers that were peaked at 10.0 MHZ in place of the crystal filter.? It sort of worked (very, very broad tuning and prone to oscillation) so I purchased the necessary 10 MHz crystals and rebuilt it per Farhan's instructions. Use of a single-crystal IF filter is interesting for QRP activists who may not need SSB.? The IF filter loss would be significantly less than the 10 to 12 db of the ladder filter, and with a variable capacitor (maybe a varicap) it should be possible to adjust the bandwidth to some extent.? Farhan & myself have discussed using the BITX for other modes, including CW by using a computer sound card or by making the microphone amplifier into a keyed oscillator (probably easier than unbalancing the modulator to get a keyed carrier). The most intriguing part of the BITX design philosophy is that it is relatively easy and inexpensive? to duplicate Farhan's design, and then to customize it for whatever special features you might want to try. Farhan uses his BITX on CW and PSK-31 as well as SSB, so we know it is capable of those modes.? Listed below are a few of the more obvious modifications for the basic design: (1)? Consider the possibility of adding VOX by using a small relay for the T/R changeover and driving that relay with an amplified and rectified portion of the microphone audio.? Anti-vox could be incorporated by integrating a small portion of the RX audio (in proper phase) with the VOX audio amplification.? For CW the same VOX circuitry could automate RX - TX changeover. ? (2)? If you use computer driven modes of operation, you should probably include a connector for your computer on the back of your BITX-xx.? This connector might include the coupling attenuators that are necessary to bring the sound card output down to the correct level to match your microphone amplifier input, as well as tapping the RX audio ahead of the volume control so that adjusting speaker volume would not disturb your computer RX signal level. ? (3)? Adding a simple FWD & REF power meter or SWR meter seems obvious and rather easy to do.? The same meter could also be used as an audio-derived S-Meter with a one-transistor amplifier and a couple of diodes. ? (4)? Depending on the type and impedance of your antenna, it might be possible to include the antenna tuner inside the BITX enclosure for a nearly self-contained station. ? (5)? There are many options for the VFO, even building two VFOs for offset frequency operation.? For my BITX40 I built a PTO instead of using the two-capacitor design of the BITX20.? If one has built a DDS (Direct Digital Synthesis) unit, then that could become the VFO for your BITX. ? (6)? At least one BITX20 builder has shown pictures of his unit that incorporates an LCD frequency display, and another has shown his BITX with either 7-segment LED and /or binary frequency displays.? ? (7)? While Farhan's original design does incorporate a linear amplifier section, it is possible to substitute any of several good linear amplifier circuits ( even a tube-type linear if you have one lying around ! ).? I expect to soon hear of BITX designs that run much more power using combinations of parallel and push-pull IRF510 finals. This amazing potential for altering and/or augmenting the basic unit is part of the magic associated with Farhan's original design.? Maybe we should start a "BITX MODIFICATIONS" sub-section as part of the forum...?? Hans, is it possible ( or even advisable ) to build a MODIFICATIONS sub-directory in the FILES section of BITX20@... that would support multiple users adding their circuits to that area? 73's Arv - K7HKL _._ On Sat, 2004-07-17 at 05:03, vdberghak wrote: Hi,why the intention to add one x-tal as a filter if you want to |
||||
|
Re. problem with crystal oscillators
Arv Evans
开云体育MILAREPA@.............It would appear from your chart that the higher frequency crystals may not be oscillating, or that your frequency counter may not be capable of counting above 10 MHz. Lets explore the first possibility (that the higher frequency crystals may not be oscillating): ? (1)? Check the RF output of your crystal oscillator using a simple diode & capacitor RF detector and a high impedance DC volt meter.? If there is RF present it should be in the range of 4 to 12 volts. ? (2)? If the transistor you are using for the Crystal oscillator is not capable of operating at 10 MHz or above, it may be not oscillating with the higher frequency crystals.? The solution for this is to use a better transistor.? In the US we have 2N2222 and 2N3904 available for very reasonable prices so this is what I use for most of my HF work.? These devices are rated to 150 MHZ, and will usually oscillate at even higher frequencies. ? (3)? In your crystal oscillator circuit it is possible that one or more of the capacitors you are using is not the capacitance value that you think it is.? Check each capacitor on a capacitor tester to determine its actual measured capacitance.? Then verify on the schematic and parts list that you are using the correct value for the design.? If you do not have a capacitance meter, you can use the simple AC transformer and Digital VoltMeter (DVM) method outlined in one of my earlier emails to the forum.? A schematic and picture of this is also available in the PHOTOS section of the forum archives (look under my call sign, " K7HKL " ) ? (4)? The DC bias may not be correct in your crystal oscillator circuit.? Check the voltages on the oscillator transistor using a high impedance voltmeter.? The difference between base and emitter voltages should be approximately 0.7 volts for silicon transistors (it will be about 0.3 volt for germanium transistors, but I have not seen one of those antiques for several years).? The collector voltage should be about 1/2 the supply voltage.? If your oscillator design has an emitter resistor you can measure the voltage drop across that device and with a little math determine the collector current of your oscillator transistor. ? (5) It is possible that one or more of the resistors in your crystal oscillator circuit may not be close enough to the design value.? A quick ohm meter check on each resistor should detect this situation.? The resistor values are not particularly critical, but they should be within 10 to 20 percent of those indicated in Farhan's design.? Now, the "possibilities" for your counter being the problem: ? (1)? Many years ago I built a number of frequency counters using 7490 dividers and 7447 LED drivers.? One of the major problems experienced with those circuits was a lack of HF sensitivity in the input amplifier and count gating circuit.? Designing a frequency counter to have a very high input impedance and flat frequency response over a wide bandwidth is nearly impossible.? I finally resorted to engineering for a lower input impedance and using a very broadband input amplifier to drive my 7400 count gating circuit.? ? (2)? One mistake that is prevalent is to use a low impedance coaxial lead as the input to your frequency counter but configure the counter for high impedance input.? Obviously this is a contradiction in impedances.? If you are doing this you can try "terminating either end of your counter input coax with a low impedance (non-inductive) resistor.? Start with about 470 ohms and work up or down from there.? An alternative might be to use a very small capacitor (about 10 pf) in series between the counter input cable and the circuit under test.? This lets the counter see a higher impedance, and does not disturb the circuit-under-test as much by loading it with whatever input cable impedance you are using on your counter. ? (2)? If you have a signal generator that will go higher than 10 or 12 MHz you could verify that your counter is capable of reliably counting to those higher frequencies.? At the same time you might want to "calibrate" the sensitivity of your counter so that you will know how much RF voltage it takes at 5, 10, & 20 MHz to result in a reliable count of that frequency.? Do this by using a carbon potentiometer to attenuate the input to your counter and use your diode detector and high impedance voltmeter to measure the RF voltage level at which counting becomes unreliable.? Be sure to use the same test leads on your counter for both sensitivity calibration and for measurements on your BITX20 circuits. Regarding the use of 3.579 MHz crystals as IF Filters: Using 3.579 MHz crystals as IF Filters is an interesting experiment.? The older and larger sized crystals that I have available exhibit much different characteristics than those of smaller 10 MHz crystals.? I was unable to obtain consistent motional parameters from my test setup when using 3.579 MHz units, but the 10 MHz crystals worked first time.? For an experiment I selected a pair of 3.579 crystals that were 1.4 KHz different in frequency and used these to build up a half-lattice filter.? The half-lattice design requires use of a toroid transformer, and here I did use a ferrite core for that purpose.? If you don't have a suitable ferrite core available you can try the Tap Washer cores that Farhan suggests, but at this time I have no data on what might be the proper number of turns.?? I found the half-lattice filter using 3.579 MHz crystals to be adequate for SSB work.? However, the use of a half-lattice filter does not readily adapt to the bi-directional design used in the BITX20, so this may not be a good solution. It is also possible to use a single 3.579 MHz crystal in series with the signal path as the IF filter in simple superhetrodyne receivers.? Again this becomes problematic if one tries to use this arrangement in a bi-directional design like the BITX20, because the terminating resistor is on the wrong end of the crystal in transmit mode.? I suppose one could do some sort of diode switching of either the whole crystal filter or just the terminating resistor, but that is rapidly getting beyond the simple nature of the original BITX design objective. Here is a thought that I have been playing with for several days...Would it be possible to use a single 10 MHz (or other frequency) crystal as the IF filter in the BITX-xx design and simply center the carrier in the IF passband.? This would result in DSB (Double Sideband) instead of SSB (Single Sideband) transmission, but it might work to get a system on the air.? One possible problem with this is that single-crystal IF filters tend to be very narrow banded, in the order of 200 HZ, and have a steep roll off on one sideband only.? I am not sure what this characteristic would do to the audio quality of a DSB signal.? I guess that someone has to try it for us to know for sure...? OK, I have probably covered much more than you wanted to know. I hope this helps. Arv - K7HKL _._? On Fri, 2004-07-16 at 04:30, MILAREPA wrote: Dear forum, |
||||
|
Re: Inexpensive plastic screwdriver
Arv Evans
开云体育PaoloIn the US, plastic knitting needles are available for an attractive price.? These range from 1/8 inch to 1/2 inch diameter.? The smaller ones are easily ground to a screwdriver point using sandpaper taped to a flat surface. If one is careful you can even shape them to hex or octagon shapes for adjusting slug tuned coils. TIP: For trial testing of coils to see if they need increased or decreased inductance you can build a simple tool to temporarily shift the inductance up or down.? Use a plastic drinking straw and insert a small brass rod (about 1/4 inch long) into one end.? Insert a small rod of ferrite material (a chunk of broken ferrite ring works fine) in the other end.? Now you can temporarily insert either end into your coil-under-test and observe the results.? The brass will cause an increase in frequency (decreased inductance) and the ferrite will cause a decrease in frequency (increased inductance). No, I have not tried this with the "Tap Washer Toroids" used in the BITX20, but I will later today because I am curious to see what the results might be. Arv _._ On Tue, 2004-07-13 at 01:52, Paolo Cravero as2594 wrote: Hi, |
||||
|
Re: BITX20 & BITX40
Arv Evans
开云体育BharatThe discrete component AF Amplifier file is in the FILES and PHOTOS sections of the BITX20 forum on Yahoo.com? You should be able to view and copy them from there.? If you have problems please send me an email and I will reply directly to your email with an attached file of the audio amp that I used.? Please be aware that I used a 741 Op-Amp in place of the LM-386 in my BITX20, and built the discrete component audio amplifier outboard in a speaker box.? With just the Op-Amp (configured for a gain of 100) there is plenty of audio to run headphones (32 ohm Walkman style headphones with the earphone elements in series).? The discrete audio amp is necessary only if you want to drive a speaker.? If you want to eliminate need for an Op-Amp, you could probably just use a single transistor gain stage (much like Farhan's microphone amp circuit) and a simple emitter follower for your receive audio.? I will try to work up a circuit diagram and test that sometime in the next few days. ----------------------------------------------------------------------------------------- CONSTRUCTION TIPS: (1) Build both the RX audio section and the TX Microphone amplifier.? By powering up both simultaneously you can drive the RX amp with your TX amp and microphone.? This lets you test both audio sections before you build the rest of the circuit. (2) Build the IF Filter and the BFO.? You can couple the output of your BFO (use a 3 pf capacitor) to the filter input and tune the BFO to determine your filter characteristics.? You will need a method of accurately measuring the BFO frequency, and a simple RF detector & high impedance DC meter to measure filter output. ----------------------------------------------------------------------------------------- For the front-end Band Pass Filter of the BITX40 I used a pair of 10.7 MHZ IF Transformers from an AM/FM transistor radio.? These required 55 PF in parallel in order to bring them to resonance on 40 Meters.? I used the whole inductor and ignored the tap on these transformers.? The secondary winding of these was also ignored, but it might be interesting to experiment with coupling the input and/or output via those windings.? A copy of the schematic and picture of this bandpass filter is in the PHOTOS section of this forum.? Look in the directory showing my call sign (K7HKL). NOTE: There was no particular reason for using the 10.7 IF transformers other than that I had them available and wanted to experiment with using them in a 40 Meter receiver.? You can also use Farhan's design for the front-end BPF (Band Pass Filter) and re-calculate & rewind the coils for 7.0 to 7.3 MHZ coverage.? IDEA:? Maybe someone could experimentally determine the "inductance versus turns" for tap washers and publish it on this forum.? With that info we could easily scale Farhan's design to almost any band. My VFO is still not in a final stage of construction.? I have built a PTO (also shown in the PHOTOS section) but am still trying to obtain exactly 10 KHZ per turn with good temperature stability.? So far, that has eluded me!? I have spread and compressed the coils to adjust the frequency coverage, but the turns on one end of the coil interact with the other end of the coil.? I seem to be getting close to my goal, but it is still not at the point where I have an acceptably even spread of frequency from one end of the band to the other.? Your BITX40 VFO will need to tune from 3.0 to 2.7 to cover 7.0 to 7.3 MHZ RX & TX if you use the standard 10 MHZ IF & BFO.? Initial calculations indicate that this places all first-order spurs and all 2nd & 3rd harmonics outside the 40 Meter band.? I admittedly have not done a detailed analysis of complex mixing products to insure that there will be absolutely no unwanted birdies within the passband. NOTE: VU2ITI has published a good article ( Ceramic_Resonator_VFO_by_VU2ITI_SPARK.pdf ) on using a ceramic resonator for VFO/VXO applications.? If you can find a copy of that document you might be able to apply something like this for your BITX40, assuming you can find a ceramic resonator that is in the 3.0-2.7 MHZ range. At this point I also have not completed an acceptable linear amplifier section for 40 Meters.? Plans are for this to probably be a push-pull set of IRF-510 devices, but work on another transceiver (not the BITX40) has taken priority over that effort.? Obviously you can alter the inductors of Farhan's design to make it work on 40 Meters, but I want to go for a bit more power on 40, thus I will need to work up a different design for the linear. For my BITX40 I have built two different IF Filter modules.? At this point I am not sure which to use.? The ladder filter using 10 MHZ crystals did not tune up well for me (the one for my BITX20 worked perfectly), probably because I used a set of older & larger 10 MHZ crystals with apparently much different characteristics.? I also built a lattice filter for 10 MHZ.? This seems to have less loss and a steeper slope for exclusion of the upper sideband energy.? However, I am still tinkering with both filters and am not ready to publish anything about them yet.? If you are contemplating near-future completion of your BITX40 you should probably closely follow Farhan's design for the IF Filter and use the same brand (KDS) crystals that he recommends. In the FILES section of the forum you will find a spreadsheet program that calculates the VFO frequencies for BITX designs.? This will also let you explore a design that uses crystals other than 10 MHZ for the filter and BFO.? This might be handy if you are contemplating use of something like TV color-burst crystals (3.579 MHZ) for your filter and BFO (this is not recommended for a BITX40 though because your 3.579 MHZ BFO second harmonic would be at 7.158 in the 40 Meter band).? My longer-term plan is to use color-burst crystals for an LF (0.160 to 0.190 MHZ) version of the BITX, but that will have to wait until I finish the BITX40. I will continue to publish portions of my BITX40 here on this forum as each section is finished and? tested.? If you have any particular problems please feel free to email me (I may be having the same problem...) and maybe we can possibly resolve them together.? The other members of this forum are also a good source of ideas and solutions. 73's Arv - K7HKL _._ On Mon, 2004-07-12 at 23:46, Bharat D Balsavar wrote: Hi Arv, |
||||
