justbits
More fiber than ATT can handle
Premium
join:2003-01-08
Chicago, IL
 ·AT&T Midwest
| Excelcus L-222RJDP RF Interference Filter
Has anybody tried this product out?
Had anyone had a real RF problem fixed by using this product?
»www.excelsus-tech.com/index.cfm?···=1749757
»www.excelsus-tech.com/Downloads/···RJDP.pdf
Thanks,
--justbits |
|
RadioDoc
58ef2c0
Premium,ExMod 2000-03
join:2000-05-11 | I haven't used that one in particular, but it looks like a common-mode noise filter. If you have enough RF in the immediate vicinity to cause problems with your phone devices it might help.
What are you trying to fix? |
|
justbits
More fiber than ATT can handle
Premium
join:2003-01-08
Chicago, IL
·AT&T Midwest
1 edit | I was wondering if this product might be able to reduce the RF interference I seem to be getting.
For more of my background on what I'm trying to troubleshoot, check out this thread:
»Any 5100b technician page DMT/SNR data experts?
I've been having a repeating problem with noisy interference on my DSL line that recurrs every weekday. It also occurs at different times of day on weekends and holidays.
In the beginning the noisy interference was strong enough to cause my SpeedStream 5660, SBC SpeedStream 5360 and SBC SpeedStream 5100b to lose sync for the entire duration of the noisy interference signal. (Typically lasts almost exactly 45 minutes.) On June 7th, my pairs of wire were replaced... I went from 5.4/800 max attainable to 3.2/600 max attainable with this replacement pair of wires. And they also failed the same way for the next 3 days. As of June 11th, the noisy interference is not causing me to lose sync anymore, but it is causing the modem to reallocate DMT bit bins all over the place while the problem occurs. I have learned how to graph the 5100b SNRM & DMT bit bin data. From data I collect every 5 minutes (or more frequent), I produce graphs (and movies) that allow me to tell when the noisy interference is occuring and how long it lasts. The noisy interference acts the same way it has since May 1st. There's a ramp-up time where the SNRM on the lower download frequencies degrades. Then, the DMT bit bins start reallocating all over the place. Then, when the problem is over, the SNRM goes up and creates "spikes" at the frequencies where there were "DMT bit deficiencies" had formed during the noisy interference. (See the other thread for a better description.) If I power cycle or unplug & replug the DSL line, I get a perfect looking stairstep DMT bit bin chart and a nice flat plateau for the SNRM chart.
By the way, I have gone through extensive grounding precautions. Details in the other thread.
Just yesterday, the problem started manifesting differently. I had two random loses of sync during the afternoon and two periods of thousands of CRC errors for two 3-5 minute periods in the middle of this morning. The high CRC error count was enough both times to cause my PPPoE connection to drop... but not enough to cause a loss of DSL sync.
The current thought is that some cell tower upgrades in the area may be somehow affecting me. beach boy & crew and I have a pretty long thread going in SBC Direct.
Anybody in the Northwest Chicago area (Cumberland Ave & Montrose Ave, just south of Cumberland & I-90) have a spare spectrographic analyzer?  It needs to be able to hook up to an antenna as well as a telephone line. It needs to work in the 40 KHz to 1108 KHz range.
--justbits |
|
RadioDoc
58ef2c0
Premium,ExMod 2000-03
join:2000-05-11
·AT&T Midwest
| If you've got a portable AM radio you may be able to figure this out without fancy test equipment. Any signal strong enough to kill your DSL will certainly show up in the lower end of the AM broadcast band.
Take the radio and set it near the modem, turn it to a 'quiet' spot near the low end of the dial and remember what the noise sounds like when the modem is working. Then, when it's not, listen again (without moving the radio or changing the tuning) and see of you can detect any difference.
The problem with that filter is that is it works with common mode noise only. Your modem should already be able to deal with quite a bit of that without any modification at all. Those are useful if you live next door to a transmitter.
Cellular/PCS/etc frequencies are more than 1,000 times higher than what DSL uses. It's almost certainly not the cell site.
Don't rule out ComEd. They may load switch at certain times of the day, and given their proclivity for not doing maintenance until the bright green flash indicates where it's sorely needed, the real problem may be a bad insulator, etc., on a pole nearby. |
|
Jan Janowski
join:2000-06-18
Skokie, IL | reply to justbits
Low end or high end of AM band? |
|
justbits
More fiber than ATT can handle
Premium
join:2003-01-08
Chicago, IL
·AT&T Midwest
| I believe the biggest noise deficiency shows up between 70*4.1325 KHz and 90*4.1325 KHz.
That's from 289.28 KHz and 371.92 KHz.
When the noise is gone, a 8 dB to 12 dB gain on the most "bit impaired" (affected) carrier frequencies is seen. An 8-12 dB noise should be audible on an AM radio, shouldn't it?
It'd be nice if the DSL modem actually gave me noise floor data, but it doesn't... gotta wait for ADSL2/2+ to get that data from the modem. *sigh*
The dB ratings I'm giving here are based upon what I'm reading off of my SpeedStream 5100b modem.
--justbits |
|
RadioDoc
58ef2c0
Premium,ExMod 2000-03
join:2000-05-11
·AT&T Midwest
| The second harmonic of both of those should be audible at around 580 and 740 KHz, respectively, if there is an outside signal causing the problem. There aren't many RF things using those frequencies around here, though. More likely something riding along in the cable bundle you are in. |
|
d_l
Barsoom
Premium,MVM
join:2002-12-08
Reno, NV
1 edit | reply to justbits
With filter | 
With filter | 
With filter | 
Without filter | 
Without filter | 
Without filter |
 .
deblin 's Excelsus RFI filter was tested on my 6016/608 service with a ZyXEL P645-A1 modem which contains a Samsung-clone of an older Alcatel chipset. My line does not suffer an excessive RFI problem, but does show modest effects from three AM stations at 630, 780, 920 (sometimes), possibly from my laptop display at 830 and 880 KHz, and from an unknown source in the power line at about 1100 KHz.
The bin load results of three different sync tests with the filter and three without the filter are shown above. The only distinctive effect is that the RFI filter centers the peak bin loads in the upload frequency range whereas without the filter the peak bin loads are skewed to the low frequency end. The filter may have reduced the download noise margin by about 1.0 dB, but my noise margins fluctuate so much every day that the only way to conclusively prove this would be by exhaustive statistical analysis. The filter also may have actually increased the effect of the various radio stations and other RFI sources have on the bin loads, but once again these affected tones routinely fluctuate by one or two bits per tone bin between syncs so this is not definitive. Note that once the modem is synced, the bit loads do not change as bit swapping apparently is either not enabled on my linecard or my modem doesn't permit it.
One definite problem with the RFI filter on my line is that it causes impulse noise, CRC errors whenever my 220V appliances are used, i.e. oven or stove top. This is due to the coupling of the phone line with my house ground which is apparently a poor grounding point. The 220V impulse noise had long plagued my DSL line and I've only recently cured it by detaching a ground wire jumper connecting the metal sheath within my drop line from the ground nut on the station protector. I believe that metal sheath is also grounded at the F2 connection and that there was an electrical potential between the grounds. The grounding jumper on it was causing the sheath to act as an antenna to pick up impulse noise from the nearby power line drop and also amplify RFI from AM radio stations. Removal of the jumper eliminated the 220V noise problem and reduced the effect the radio stations have on my bit loads. Using the RFI filter on the line effectively restored the effects of that grounding jumper.
So the bottom line is that my poor test results with the RFI filter may not be typical due to my house ground. They are certainly different from the results deblin obtained with this filter. |
|
Seppy
join:2004-08-29
Aurora, IL
| reply to justbits
I think I might try one of these filters out. Ever since I have had my DSL(2years) it has not worked correctly. I also have a very long history in SBCDirect. They have put a NID Splitter outside to rule out my inside wiring and have replaced various parts in the RT and also did what they call a bunny something, where they try to get me as direct a line as possible? I am starting to wonder if it is a problem with the power in my house? How can I tell if my power is bad and causing problems? If I turn the breaker off and then back on for the whole house will that clear any power problems? I don't really know that much about the power in a house. I lose sync to the modem at least once a day sometimes almost every hour, but it is random because some days will be ok. |
|
justbits
More fiber than ATT can handle
Premium
join:2003-01-08
Chicago, IL
·AT&T Midwest
| A "bunny something"? What does it look like? Any idea?
A home run is the typical name for a short & direct run from the NID to your modem.
If you're having power line RFI problems, you could run out to RadioShack and get some ferrite cores and wrap your telephone line around them just before the line goes into the modem. RadioShack also has a telephone RFI filter that may help improve your line statistics.
As far as turning off power, that's only useful if you've got an AM radio handy. Tune the AM radio into a frequency that doesn't have a radio station. Move the radio near the circuit breakers/fuses. If you hear loud cracking or loud noise that increases when you get near the fuse box, that's excessive RFI that you're picking up. Now, disconnect your main circuit breakers/fuses for the entire house. If the noise goes away, the RFI is caused by a device in your home. If the RFI is still present, it is likely the power lines that are radiating excessive RFI.
By turning on breakers/fuses one at a time, you can isolate which house circuit may be the RFI radiator.
In any case, if you have UPSes, make sure you disconnect everything from them and turn them off also. They can radiate RFI too.
Have you tried disconnecting all of your inside wiring except for a telephone line that runs to your DSL modem? Use the microfilter there to connect to one phone and the DSL modem. Have you eliminated your inside wiring from the problem equation?
In all cases, make sure you collect information about your DSL line before & after the problem. There's a program called ADSL Monitor that you could use. There is documentation on where to get statistics from other modems too. What kind of modem are you using? If you have a 5100b, I have some excellent visualization tools if you post relevant technician readout data.
There's lots of power line RFI information out there. Do some web searching. There's quite a bit of RFI information in the SBC FAQs and in other posts in this forum.
Just a note for those stalking me, I acquired a Excelcus L-222RJDP RF Interference Filter and a huge torrid ferrite core (3 inch diameter) yesterday. At my first glance at the data this morning, the L-222RJDP RFI filter does not solve my interference problem. I'll post more concrete data soon. I'll post some pictures of my insane ferrite core setup also.
--justbits |
|
resh
join:2005-08-27
Elk Grove Village, IL
| reply to justbits
Yes, I have the L-222RJDP RF filter. As a matter a fact I waited over two years to order DSL because I live ONE BLOCK AWAY from a 50kw AM station. I didn't bother ordering knowing I would get lousy results. Before DSL I could listen to the radio with the phone headset laying on the table! I had to use a voice RF filter at the NID and one at the phone.
When SBC came up with this $15.00 thing I thought it would be worth the research if I could fix this RFI. My search turned up the the excelsus site and as we know supply SBC their standard filters.
In addition to the $30.00 L-222RJDP I also bought a special "switching" DSL filter Z-275P2J $15.00, similar to the supplied standard filters but switches the voice line out when on hook for better DSL performance.
Note, I installed both filters at the time of installation so have no comparison without. Also I believe they may have switched me to an RT they recently installed (1500ft the tech said) at the same time of installation so again I have no before or after tests. SN is 21.5 db and I get 1300/327, I removed the two voice RF filters assuming they may filter the DSL spectrum.
|
|
justbits
More fiber than ATT can handle
Premium
join:2003-01-08
Chicago, IL
·AT&T Midwest
| reply to justbits
I bought a Radio Shack portable radio (TVh,TVl,Wx,AM,FM).
I connected my DSL modem.
I looped my telephone cord in a circle.
I tuned the radio in to a low AM frequency (between 500 & 600 KHz) that amplifies significantly whenever I get near _anything_ electronic that is not shielded. It's almost as if it is like one of those spy "bug-detector devices" that you can buy.
I placed the radio over an edge of the circle.
I could hear a radio station.
Then, I disconnected the DSL modem.
The audible radio station signal disappeared. That was kind of strange that the DSL modem was AMPLIFYING the radio station when the modems were syncing or in-sync, but that the audio is not present when they are out-of-sync.
I connected the Excelcus L-222RJDP RF Interference Filter into the telephone line _before_ the wire loop that runs into the DSL modem.
I connected the telephone line back up to the modem.
The audible radio station signal is no longer audible, but the amount of static is significantly higher.
With a 5100b, instead of getting a tall stair-step like DMT bit bin distribution on the download frequencies/carriers, the DMT bit bin distribution looks more like a mostly even plateau. Attenuation increased by 1.0 dB with the RFI filter. (From 58.5 to 59.5.)
With a 2wire 1800HG attached to the DSL line, the modem was reporting a significantly higher uncancelled echo.
--Steve |
|
wayjac
Premium,MVM
join:2001-12-22
Indy | Is the 2wire new to your network?
What do you think about the 2wire?
--
God bless our troops |
|
