| Home |
| Search |
| Today's Posts |
|
|
|
#1
October 11th 05, 01:20 AM
|
|||
|
|||
|
|
#2
October 11th 05, 12:56 PM
|
|||
|
|||
|
For VHF 3.2 would likely work well enough except if there is a
sporatic E opening and people tend to pile up. Then it's really wide! For filters I use microprocessor crystals in the ladder configuration. With the correct shunt C and 4-8 crystals you can make a very fine crystal filter with bandwisths from 2-400hz to as wide as you may want. The design process is documented elsewhere (see EMRFD and the handbook as well as internet). I build for 6 and 2m and have built very nice ladder filters this way. Also since microprocessor crystals in the range of 4-20mhz are dirt cheap it's also a help. Also the higher the frequency the less likely dual conversion is required to avoid images and allows the selectivity to be closer to the antenna (better overload performance). My latest 6m rig uses 12mhz crystals, 8 of them for a 2.3khz bandwidth at 6db and 3.9khz at 60db with symetrical skirts. Allison KB!GMX On Mon, 10 Oct 2005 20:20:05 -0400, Ken Scharf wrote: wrote: On Mon, 10 Oct 2005 20:48:51 +0100, John Wilkinson wrote: Hi, I have bought a 45MHz crystal filter with a bandwidth of +/-10KHz, according to the spec. Now does this mean a real bandwidth of 20KHz? Likely the case. Some spec total bandwidth other specify 6db edge as distance from the filter center frequency. If so when a receiver states a bandwidth of 7KHz, is that +/-7KHz, ie 14KHz? If they didn't put the +/- there then its 7khz total. If I am designing a good communications receiver from 6KHz-30MHz to receive AM, SSB and CW and want 2 good filters, what are the bandwidths I should consider? AM wide 12-16khz Am narrow 6-10khz SSB anywhere from 2-3khz with many around 2.4khz wide Any wider than 3khz will be poor in crowded bands. I happen to prefer 2.1 to 2.3khz. CW I've seen 1.4khz all the way down to 200hz most consider 400-600hz adaquate. In all cases the skirt selectivity usually bandwitdth measured from the 6 to 60db points are important indicators of filter quality and any value of 2 or less is good enough and 1.4 would be excellent. The idea is you'd ike to be able to put the offending signal outside the bandpass and well attenuated. What some builders do for CW is use the CW filter and use an peaked audio filter to narrow the audio band pass. Not quite as effective but often cheaper. Allison KB1GMX That 45mhz would make a good 'roofing' filter to use ahead of a 455khz if filter. Normally using a first if of 45mhz and a second if of 455khz would result in 'second order' images leaking in, but with a good roofing filter ahead of the second mixer the problem is solved. Some years ago, I bought a bunch of 9mhz if filters at the Dayton Hamvention. They are 3.2khz BW (at the 6db points) filters made by CF Networks for the Gonset Sidewinder transciever. I think this was a vhf rig, which would account for the wider bandwidth. However, these are 8 pole filters, with a shape factor of 1.08 @ 15db down (rises to 1.23 @ 45db down). With this shape factor, these filters probably have similar rejection of off frequency qrm as would a 2.4khz filter with a 2.0 shape factor. Since I have a bunch of them, I could use 2 or even 3 of them in the IF stage (one after the mixer, one before the detector, and one between stages). This would decrease the apparent shape factor even more. The nominal carier frequencies for these filters were 8998.0 khz and 9001.7 khz. With the 3.2khz bandwidth this put the cariers 250hz outside the stated bandwidth (-15db point for these filters). I wonder just how these filters would actually work out in a rig today. That 3.2khz bandwidth DOES seem a bit wide, but the crazy shape factor seems to make it worthwhile to try. |
|
#3
October 14th 05, 10:16 PM
|
|||
|
|||
|
|
|
#4
October 14th 05, 11:14 PM
|
|||
|
|||
|
On Fri, 14 Oct 2005 17:16:21 -0400, Ken Scharf
wrote: wrote: For filters I use microprocessor crystals in the ladder configuration. With the correct shunt C and 4-8 crystals you can make a very fine Allison KB!GMX I managed to buy over 400 pcs of 8.3886mhz crystals on ebay, for just pennies each. I am planning on trying to build ladder filters with them. These crystals are in the larger HC6/u size holders which are supposed to work better than the miniature size used in the micro- processor crystals. The HC6 parts work fine as do the HC18, 49 and so on. The real trick is doing the work to measure and check the crystals for use and then calculate the capacitors and termination impedence based on that. For a little work you get fine filters dirt cheap. First step would be to build the DDS vfo for the radio since I can program the DDS to function as a sweep generator for aligning the filter. With the DDS sweeping the output frequency while providing a sawtooth ramp to drive the scope sweep in step with the frequency sweep I could see the actual bandwith plotted on the scope. While I have a DDS to do that with I found that using the first "high" crystal in a VXO that gets calibrated worked as well with a lot less fuss. Then I can use the same osc to sweep the filter later to test it by adding a varicap doide. Allison |
|
#7
October 15th 05, 02:55 AM
|
|||
|
|||
|
On Sat, 15 Oct 2005 00:24:39 GMT, "Dale Parfitt"
wrote: Just sweeping a filter designed for SSB would be fine I suppose. I recently had an opportunity to hear the difference between a stock FT-1000 CW filter and one homebrewed with attention paid to group delay- the difference was very clear to hear- in favor of the homebrew filter. \ Dale My prefered filter has a gausian to 6db shape for less ringing and group delay. I work for that goal. However, try the KK7B Phasing rigs for sound. They are direct conversion SSB (image rejecting) so all the selectivity is in the audio bandpass. I use a miniR2 and T2 pair on 6m and filter artifacts like group delay aren't there. Transparentcy is a good word to describe it. Allison KB1GMX |
|
#8
October 15th 05, 03:52 PM
|
|||
|
|||
|
A common mistake in years past was to try to put all of the selectivity into
a single super-deluxe crystal, mechanical or digital filter. These filters quite often have "raspy" noise interference at the edges of the passband (especially in CW mode) due to the enhancements of spectral noise peaks at the very sharp band edges, caused by the conversion of the phase statistics of noise to amplitude statistics (each edge of the filter acts like a phase disciminator). This raspy noise interferes with weak signals. The filter can be equalized for group delay, as mentioned, with improvement in the problem. The band edges can be softened, with good results. A much better way is to use two or more intermediate-performance filters in cascade. This method softens the edges so that the effect is greatly reduced. It also improves overall shape factor. A cascade in this manner of identical bandpass or audio lowpass filters tends in the limit toward the Bessel or even the Gaussian response. Digital filters can also use a method called Transition Band Sampling (see Oppenheim and Schafer 1975 or Oppenheim and Willsky 1983). All of these results are related in principle to the Central Limit Theorem of statistics. The cascaded filter approach is also very beneficial in other respects, in particular the reduction of wideband noise in high-gain IF amplifiers. This noise degrades AGC performance and adds audio frequency noise to the product detector output. Bill W0IYH "Dale Parfitt" wrote in message news:b%X3f.301$W32.225@trnddc06... Just sweeping a filter designed for SSB would be fine I suppose. I recently had an opportunity to hear the difference between a stock FT-1000 CW filter and one homebrewed with attention paid to group delay- the difference was very clear to hear- in favor of the homebrew filter. \ Dale |
|
#9
October 15th 05, 02:50 AM
|
|||
|
|||
|
On Fri, 14 Oct 2005 19:38:37 -0400, Ken Scharf
wrote: That would work fine, but with the DDS, I can program the actual frequency range to be swept and probably be able to calibrate the scope face to read the actual frequency 'break' points on the filter. Using the vxo method will get you a working filter quickly no doubt, but I will still need the DDS vfo for the finished rig, so I just figured I'd do that first. Makes sense. I found the other way easy when DDS chips were 55$ each! The varicap sweep is calabrated to the scope so that was not an issue for sweeping the filter. I found that using wideband noise and a sound card was better. How many 'rocks' did you use in an SSB filter? I've seen some designs on the web with 6 crystals, would the shape factor be any better with 8 or more? (with over 400 crystals in the junk box I can go crazy, but I'd still have to find the capacitors :-). I'd say 4 is a useable minimum. With that I'll add the skirts at 40db down are not very good though. I've used 6-8 to get a good 6-60db shape (under 2:1). There is a problem if you go for too many. The filter can have enough group delay that while it's shape is good, the sound has a hollowness. The caps, once you figured the qalues you will likely end up using parallel values. IE: 232pf may be a 220+12pf or a 220 and a 4-20pf trimmer. Allison |
| Reply |
| Thread Tools | Search this Thread |
| Display Modes | |
|
|
Similar Threads
|
||||
| Thread | Forum | |||
| How to measure soil constants at HF | Antenna | |||
| FA - R. L. Drake SW8 'portable' World Band Shortwave Communications Receiver | Shortwave | |||
| Constant bandwidth TRF circuit | Shortwave | |||
| a page of motorola 2way 2 way portable and mobile radio history | Policy | |||
| CCIR Coefficients METHOD 6 REC533 // AUCKLAND --> SEATTLE | Shortwave | |||
Hybrid Mode
