On Fri, 2009-04-17 at 00:36 -0400, faginbagin wrote: > > I don't think it's a board problem. I think it's an incoming signal > > problem. If your cable signals on some channels are so strong that they > > start overdriving the analog receiver front end (bumping into non-linear > > regions of the amplifiers), then you can get intermodulation products > > which will show up as noise on other channels. The 3rd order > > intermodulation products can be particularly strong at frequencies > > *higher* than the two or more strong channels that caused them. > > That's an interesting hypothesis. FWIW, about a year ago, the cable > company upgraded our connection from outside. We're 200 feet from their > box, and I think they replaced an RG-6 cable with RG-7, maybe RG-8 (not > sure if I remember what the guy told me). > That goes into the attic, > where it's split to the cable modem and to an inline amplifier. It then > goes into an 8 way splitter. From there it goes to various rooms, > including mine. I'm pretty sure it's RG-6 cable going to a wall mount. > Then a 50 foot RG-6 cable goes to a 4 way splitter to 3 foot RG-6 cables > to the TV, PVR-150 in old computer, and the HVR-1600 NTSC & ATSC/QAM > connectors. I first noticed the static when I had a extra split, 8 > way to 2 way to a cheap 4 way. And mythtv was having trouble finding all > my clear QAM channels. I was hoping eliminating the 2 way split and > replacing the cheap 4 way with a better quality one would allow me to > get all my QAM channels and possibly clear up the static.
Please double check the guidelines here: http://www.ivtvdriver.org/index.php/Howto:Improve_signal_quality An unterminated splitter output can cause reflections that show up as elevated noise floor. For the record there are two things that can happen to your signal: you can lose signal power or you can get additional noise or interference added. Although the net result may be the same (degraded signal to noise ratio), the mechanisms, and hence remedies, may be different. > I would have > thought there were lots of places where I was losing signal strength, > but maybe not enough to make the HVR-1600 happy. > > Since my PVR-150 doesn't have a problem with static coming off the same > splitter, I'm wondering if maybe the analog components on the HVR-1600 > are inferior to the PVR-150. First, I have had a 4 way spliteer where one (and only one) of the output ports was bad. It was a brand new expensive one too. :( Second, if my guess about intermodulation products being introduced by strong incoming signals in the front-end amp is correct, weak signals are not the problem, strong ones are. The strong signals get clipped when they overdrive the amp and spectral components end up somewhere else in the RF spectrum (i.e. on another channel) showing up as noise. Since BTSC audio is FM, and since you have watchable video so a strong enough FM audio signal, and since natural FM noise sources are rare, it would make sense that intermodulation products could create the audio static. I can't call the HVR-1600 analog tuner inferior in this case, because it could be that the tuner's 1st stage amplifier has a higher gain, if the problem is due to intermodulation products. Or it could be inferior in that the 3rd order intercept points of the components before the IF filter are lower than that of the PVR-150's tuner. Without knowing both tuners and the having the spec for both, it would be hard to tell which is better under what circumstances. > If so, do you have any opinion about what > would be a better dual tuner card? Too bad they're not allowed to sell > analog only cards anymore. > > > I think I have detected the symptoms you describe on broadcast channel > > 32 (579.25 MHz) on my setup with my HVR-1600. The static is slight, but > > it is noticable and consistent. > > I sure hope it's the same root cause. Well, my spreadsheet tells me that a 3rd order intermodulation product of channel 26 (543.25 MHz) and channel 20 (507.25 MHz) located at 2*f1-f2 = 579.25 MHz lands in channel 32. Since Channel 26 and 20 are strong channels, and will certainly pass right through the UHF preselector filter into the first amp, I'm not looking any further for possible intermods - those stations are the ones that matter. I've got a few tests to run: 1. I tried to use about 100 ft of RG-6 as an in-line attenuator to get rid of possible intermods. It didn't have a noticeable effect on the static on channel 32. I don't know how much attenuation I'm getting out of my length of cable, but maybe not enough..... 2. I'll build a bridged-T in line attenuator sometime later this weekend, where I can control the attenuation. 3. I'll try and twiddle the audio processing in the cx18 driver to use the tuner AF out instead of the SIF. If the origin of the static is not somewhere in the tuner but in the CX23418 digitization and broadcast dematrix, then static shouldn't persist under this configuration. You, however, may want to just try bypassing you in line amplifier and seeing if the static goes away. > > There are relatively few strong over the air station where I am at. It > > shouldn't take me too long to build up a spreadsheet of all the possible > > intermodulation products that pairs of strong stations below channel 32 > > could create on channel 32's frequency band. > > > > If I'm right, then attenuationing the strong stations with an external > > attenuatior or filter should make the static go away. > > Would such a device be limited to a narrow frequency range? I guess I'm > wondering if it would interfere with my ability to tune clear QAM > channels. Although, since the HVR-1600 has separate inputs, I guess it > doesn't matter. A simple attenuator would just be resistive. Bypassing your inline amp temporarily would have a similar effect. > > Since I can reproduce this, I'll try and debug this without you going > > through too many hoops. Hopefully the symptoms won't go away. Over the > > air propagation can be weather and time of day dependent... > > Don't worry about me going through hoops. I see this as an opportunity > to contribute. I wish OTA was an option for me. According to antennaweb, > I would need a large directional antenna to get most of the stations I > want. I would need three to get all the stations I want. Three antennas combined would hurt signal to noise ratio. You'd get the signal from one antenna with the noise from 3 antennas. Manmade and galactic noise dominate at VHF and UHF. With 3 antennas pointed towards 3 cities you'll get more manmade noise and would be more likely to have a strong galatic noise source in the beam of an antenna (i.e. pointed at the galactic disk) at any given time of day. > I'm ENE of most > Detroit area stations, WNW of two Canadian stations we like, and S of a > Flint station we like. > > > This all looks OK, after a cursory glance. You may want to unload > > lirc_i2c while testing, just to eliminate any possible unknown due to > > lirc.. lirc_i2c is going to use the first I2C master of the CX23418 to > > talk to the Z8 IR microcontroller chip. The analog tuner commands are > > sent on the second I2C master of the CX23418 chip. > > Done. lirc_dev is still there, hope that's OK. Yeah. > > I was set to look for just NTSC-M. You were set to autodetect NTSC-M > > NTSC-M as implemented in Japan, or NTSC-M as implemented in Korea. That > > shouldn't have affected things, but you can try and set it back to just > > looking for NTSC-M with v4l2-ctl, to see if it make a difference. > > Must be a myth thing. It gives you a choice between NTSC and NTSC-JP, no > mention of Korea. I will try setting it to just NTSC-M. v4l2-ctl can do it. I wouldn't worry too much about it though. > >> Anyway, I'm now going to hook up the DVD player and see what happens. > > > > For inputs other than the analog RF tuner, I suspect you won't get the > > static. > > > > Also for an RF source with only one (e.g a VCR) or only a few weak (e.g. > > rabbit ears and a UHF loop or bow) RF channels, I suspect your > > HVR-1600's analog tuner won't produce the static. > > Hmm, I do have an old Gemini cable box. I think it's 20+ years old. > Would it be a useful data point if I hooked it up to the HVR-1600? Well. We know a cable box will only output on channel 3. There is no chance for intermodulation products and you will also only exercise the low VHF part of the analog tuner. Your problem on channel 29 manifests with the UHF part of the tuner with many channels available. If the static persists with the cable box attached, we'll know it's not a signal problem if one assumes the analog tuner is not defective. > > Meh, no big deal. Although if you can prevent the line wrap next > > time.... > > I'll try to remember. > > > The driver will refuse to stuff custom VBI packets in a DVD stream, > > since the driver doesn't know how to do this in software. That prevents > > your stream from getting corrupted. > > Mind if I go off-topic and ask about the status of VBI? The ivtv > driver adds a private data stream containing VBI data to a DVD stream. I > assumed that eventually, the cx18 driver would do the same. If I change > to MPEG-PS, will I get that VBI stream now? If you need a tester for > this feature, I'm very interested! It works, both sliced and raw. I've got to make improvements for non-NTSC standards though. The interlock on insertion of private packets only in the PS is strictly a cx18 driver software thing. I know the insertion will corrupt the MPEG TS and would likely not do well with the MPEG-1 streams. So I put in an interlock to only let it happen with the PS. If there are stream types where you know from experience ivtv wouldn't corrupt the stream, let me know and I'll add it back for cx18. Here's what I think about enabling private packet insertion without in depth analysis of anything other than the PS and TS: OK 0: MPEG-2 Program Stream No! 1: MPEG-2 Transport Stream No? 2: MPEG-1 System Stream ??? 3: MPEG-2 DVD-compatible Stream No? 4: MPEG-1 VCD-compatible Stream ??? 5: MPEG-2 SVCD-compatible Stream > I did test the S-Video/Line-In inputs. No static. And channel 29 still > has static, despite removing lirc_i2c. Good to know baseband audio isn't having problems. Regards, Andy > It's getting late and I'm feeling less pressure since it looks like > you've got a plan. I will do more testing with your parameters over the > weekend. > > TTYL, > Helen _______________________________________________ ivtv-users mailing list [email protected] http://ivtvdriver.org/mailman/listinfo/ivtv-users
