luns@spocom.UUCP (Luns Tee) (04/09/91)
In hopes of doing something about the familiar problem of getting a VCR to do timer recordings of more than one channel including one piped through a descrambler, I've been playing with Radio-Shack's block converter. Meant for other uses, this can also take the channel 3 output from a descrambler, and bump it up to cable channel 91 (with a frequency inverted version around channel 70). Feeding this into a cable combiner with the normal cable service lets the VCR record any non-scrambled channel normally, and record scrambled programs on channel 91 provided the descrambler is on. Now, my problem is that the channel 91 and 70 signals aren't all that come out of the block converter. There's some crosstalk, and some of the original channel 3 gets through, as well as many other signals all over the spectrum. Piping the signal through the UHF side of Radio-Shack's UHF/VHF VCR to TV combiner (which is a highpass T filter) gets rid of channel 3 completely, but there's still some pretty strong interference higher up in the VHF cable channels. Two approaches that I'm considering for getting rid of this interference are 1) build a highpass filter with a pretty sharp rolloff between 626 Mhz (channel 91) and 432 Mhz (just above the highest cable channel in use). Cutoff would preferably be higher in this range should the cable company decide to add some more channels in the future (and I prey they never reach 91!) 2) in hopes that the interference is a conversion of unwanted signals picked up in the signals fed *into* the converter, build a lowpass filter with a cutoff at (or a little above) 66 Mhz (the top of channel 3). Perhaps make cutoff 72 Mhz instead should I decide to modify my VCR to tap only a channel 4 output from the RF-modulator and feed that in as well. Now, I'm not experienced enough (high-school) to go about designing either filter from scratch, but I found in a book a highpass filter with a very sharp rolloff (and over 30 db attenuation below the rolloff), but at 40 Mhz. In theory, I could take all the values in the schematic and divide them by 15 to get a cutoff at 600 Mhz, but values get so small (around 7 pf for most of the capacitors, with one around 2, and inductors of about 7 nH) that stray capacitances and inductances take over. The second filter would be much easier to design I suppose, but I'm afraid that the interference may be generated within the converter - perhaps from harmonics of what's piped in. Otherwise, I'd think it's the preferable approach. Suggestions anyone?
ppa@col.hp.com (Paul Austgen) (04/11/91)
In my opinion, you should always try to buy a commercially available filter whenever possible. Filter design is tricky and time consuming. You might check with local electronics suppliers that carry CATV/MATV product lines. The output of a VCR is generally very "dirty", and a bandpass filter on the output would probably help. You should also be aware that if you are combining signals on the cable feed, you run the potential of interfering with aircraft and other commercial radio, and are probably in violation of local regulations of the cable company. Such tinkering needs to be done properly, or lots of trouble may result.