26-02-2017, 09:13 PM
Hi Doug and Peter,
Thankyou both for your very helpful replies.
I will answer in order.
Doug,
I take your point about how an addition (or subtraction) of turns at the top of the helical filters could, indeed almost certainly will, change the characteristic impedance of the resonator, and the coupling in/out, and this in turn may no longer be optimum. Your idea of compressing the winding to increase both L and C, to lower the resonant frequency, is a good one. I think that would probably also change the characteristic impedance but it might have less effect on the coupling, because it does not change the ratio of turns above and below the tapping point. (Actually there are taps only at the input and output - the inter-stage coupling is done via slots between the cavity sections.) Alas, Doug, your memory is good. The winding is indeed on a grooved former!
By the way, I should probably mention here that I was mistaken in an earlier post when I said that the google-groups article (Andrew, VK4CRT)suggested adding 1/2 a turn to the helical filters. It suggested that for W1 band radios (470-500 MHz), but NOT for T Band (403-420 MHz) or U Band (450-470 MHz) radios. However our experience is that 1/2 turn is needed for U Band radios, but resulting sensitivity still leaves a lot to be desired.
I agree that receiver sensitivity is only one part of the consideration in repeater design. Receiver sensitivity, transmitter power and cleanliness, diplexer isolation, and rejection of nearby signals are all strongly interacting factors. However, receiver sensitivity relative to transmitter power is very important, to ensure good performance with hand-held transceivers, which will generally have higher sensitivity than the repeater, and much lower power. So, ideally I would like to start by saying what sensitivity and power I want, and then engineer the isolation, whether by separation of antennas, filtering (diplexer) or separation/shielding of the receiver from the transmitter or indeed some combination of these. I generally leave the antenna gain out of this consideration, because it works equally for receive and transmit (i.e. capture area and gain increase in proportion to each other). (There are considerations of the maximum desirable gain of course, in relation to structural issues, lightning, and site location.)
Having said all that, I would also say that I would not go for a ridiculously high receiver sensitivity, and I would not risk putting a low-noise preamp in front of a repeater receiver. I think that would be asking for trouble. But I have run W1 FM828s in commercial service for many years (around 475 -480 MHz) and I have found them to be very happy to work with full sensitivity and 15 watts output, all in the one radio, and without detectable de-sensing. (I did have a very good diplexer with band-pass filtering.) I think the standard FM828 sensitivity specification (0.5 dB for 20 dB of noise quieting or 0.3 dB for 12 dB SINAD) is a good compromise - not too much to cause problems, but enough to perform reasonably well when accessed by hand-held radios. However, I think your point about separate transmitters and receivers is very relevant for UHF CB repeaters, where the offset is only 750 kHz, rather than the 5 MHz which we use for amateur repeaters.
------------------------------------
Peter,
Thanks for the two links. I had forgotten about the great series of articles by Will McGhie in AR /Repeater Link, and I will also revisit those. They give almost "too much information" about audio routing, so I think now have that covered. I'm not sure whether I missed something, but my impression is that there is not anything to help me on tuning of UHF FM-828 receivers on the WARG web-site.
My next step will be do some measurements on un-modified FM-828s and I may eventually report on those in a new post under FM, Repeaters, D-STAR. So far I have acquired a few W-Band units and a T-Band unit. I still need to acquire an unmodified U-Band radio. I have been waiting to have enough time for this project for about 9 years. I still don't have time, but I have started thinking about it now, and I don't seem to know how to stop! But it may take a while...
Best wishes,
Lew.
Thankyou both for your very helpful replies.
I will answer in order.
Doug,
I take your point about how an addition (or subtraction) of turns at the top of the helical filters could, indeed almost certainly will, change the characteristic impedance of the resonator, and the coupling in/out, and this in turn may no longer be optimum. Your idea of compressing the winding to increase both L and C, to lower the resonant frequency, is a good one. I think that would probably also change the characteristic impedance but it might have less effect on the coupling, because it does not change the ratio of turns above and below the tapping point. (Actually there are taps only at the input and output - the inter-stage coupling is done via slots between the cavity sections.) Alas, Doug, your memory is good. The winding is indeed on a grooved former!
By the way, I should probably mention here that I was mistaken in an earlier post when I said that the google-groups article (Andrew, VK4CRT)suggested adding 1/2 a turn to the helical filters. It suggested that for W1 band radios (470-500 MHz), but NOT for T Band (403-420 MHz) or U Band (450-470 MHz) radios. However our experience is that 1/2 turn is needed for U Band radios, but resulting sensitivity still leaves a lot to be desired.
I agree that receiver sensitivity is only one part of the consideration in repeater design. Receiver sensitivity, transmitter power and cleanliness, diplexer isolation, and rejection of nearby signals are all strongly interacting factors. However, receiver sensitivity relative to transmitter power is very important, to ensure good performance with hand-held transceivers, which will generally have higher sensitivity than the repeater, and much lower power. So, ideally I would like to start by saying what sensitivity and power I want, and then engineer the isolation, whether by separation of antennas, filtering (diplexer) or separation/shielding of the receiver from the transmitter or indeed some combination of these. I generally leave the antenna gain out of this consideration, because it works equally for receive and transmit (i.e. capture area and gain increase in proportion to each other). (There are considerations of the maximum desirable gain of course, in relation to structural issues, lightning, and site location.)
Having said all that, I would also say that I would not go for a ridiculously high receiver sensitivity, and I would not risk putting a low-noise preamp in front of a repeater receiver. I think that would be asking for trouble. But I have run W1 FM828s in commercial service for many years (around 475 -480 MHz) and I have found them to be very happy to work with full sensitivity and 15 watts output, all in the one radio, and without detectable de-sensing. (I did have a very good diplexer with band-pass filtering.) I think the standard FM828 sensitivity specification (0.5 dB for 20 dB of noise quieting or 0.3 dB for 12 dB SINAD) is a good compromise - not too much to cause problems, but enough to perform reasonably well when accessed by hand-held radios. However, I think your point about separate transmitters and receivers is very relevant for UHF CB repeaters, where the offset is only 750 kHz, rather than the 5 MHz which we use for amateur repeaters.
------------------------------------
Peter,
Thanks for the two links. I had forgotten about the great series of articles by Will McGhie in AR /Repeater Link, and I will also revisit those. They give almost "too much information" about audio routing, so I think now have that covered. I'm not sure whether I missed something, but my impression is that there is not anything to help me on tuning of UHF FM-828 receivers on the WARG web-site.
My next step will be do some measurements on un-modified FM-828s and I may eventually report on those in a new post under FM, Repeaters, D-STAR. So far I have acquired a few W-Band units and a T-Band unit. I still need to acquire an unmodified U-Band radio. I have been waiting to have enough time for this project for about 9 years. I still don't have time, but I have started thinking about it now, and I don't seem to know how to stop! But it may take a while...
Best wishes,
Lew.