RFI filter suggestion

[doyall]

Anybody got and ideas for an in-line DIY RFI filter (or failing that, a suggestion for a store-bought one) to prevent my powered sub from picking up the signal from a tower a mile away when using the LFE input?

[moldyoldy]

You'll need to determine the "point of entry" of the offending RF. To eliminate the input side, merely unplug the audio cable, power it on and set to max gain. The presence or absense of the phenomena will determine what to do next.

[doyall]

It is the LFE cable itself that picks up the signal. When I connect the sub using speaker level output the RFI does not exist. Is there some type of filter I can run the LFE signal through to eliminate the interference?

[X_X]

It is the LFE cable itself that picks up the signal. When I connect the sub using speaker level output the RFI does not exist. Is there some type of filter I can run the LFE signal through to eliminate the interference?

Yes, but I would use that as a last resort. A more effective alternative might be to boost the shielding on that cable. You could either upgrade the cable, or go the cheap route and wrap it in Teflon plumber's tape. Don't laugh- it works.

[moldyoldy]

Sounds like the LFE cable you're using makes a decent antenna. A good-quality double-shielded coax cable should suffice, but I'm not suggesting anything particularly expensive either. Some considerations;

Minimize length--anything over 6-8 ft invites noise. Less is even better.
Don't coil up excess cable.
Don't run cable parallel to any power cords. Signal cables that must cross power cords should do so at right angles.
No Y-adapters.

Another possibilty if you have cable TV; investigate to ensure the TV cable is well-grounded (outdoors).

Before spending any $, check your sub manual or mfr website for suggestions.

If all else fails, a ferrite core installed on the LFE cable might help, for example, this one from Radio Shack;

http://www.radioshack.com/product/index.jsp?productId=2103979&cp=&pg=2&y=3&origkw=rfi&kw=rfi&kwCatId=2032058&x=18&numProdsPerPage=100&parentPage=search

[Chas]

If you have confirmed your cable is good, then try soldering a .001 uf (voltage is not critical) capacitor across the LFE input jack of the sub.

[Mr. Widget]

I had an RFI problem with a couple of tube amps I was running. I tried filters, ferrite cores etc... none of these "tricks" completely solved the problem. It turned out that finding a better system ground and tying EVERYTHING in the system to that ONE ground took care of the problem.

I assume your powered sub is plugged into a separate circuit from the rest of your system. If this is the case, you should try running a BEEFY extension cord from the same outlet as your system and see if that helps.


Widget

[doyall]

I've tried four different interconnects (all well shielded and one very expensive) all in the 6 - 8 foot range and run as far away from power cords as possible to no avail. This leaves me with a ferrite core, plumber's tape and a capacitor in the line.

Ferrite core and plumber's tape would be simple if they work.

If I have to go with capacitors, should they be in series as opposed to parallel? If in series, on one or both sides of the line? I do not have any .001 uF caps handy but do have some .01 uF's. What will those do?

[Don C]

If you want to use capacitors they would go between the signal line and ground. Remember that caps pass AC and block DC. So the paralell connection would shunt the RF to ground. I'd go with Widget's suggestions regarding grounding first. Disconnect everything and hook everything in the system to to one power outlet. Build a bare bones system, disconnect everything from the pre-amp except for the CD player and see if that helps. Then add things one at a time and see where the noise begins.

[Chas]

I've tried four different interconnects (all well shielded and one very expensive) all in the 6 - 8 foot range and run as far away from power cords as possible to no avail. This leaves me with a ferrite core, plumber's tape and a capacitor in the line.

Ferrite core and plumber's tape would be simple if they work.

If I have to go with capacitors, should they be in series as opposed to parallel? If in series, on one or both sides of the line? I do not have any .001 uF caps handy but do have some .01 uF's. What will those do?

You could try .01 since the signal is all below 100 Hz or so, it shouldn't affect anything since you are feeding from a dedicated LFE output. The idea is to shunt all the high frequency noise at the sub input to ground. It's an old amateur radio trick used to deal with neighbors who complain about RFI.

[Thom]

There's two reasons that the speaker level inputs wouldn't pick up the noise. One is signal level. The noise is probably nothing compaired to the signal on the speaker wire and the other is impedence the input for the speaker level leads is probably so lo that it's basicly a short circuit to the noise signal. If you can lower the impedence of the lo level input without effecting the quality of the sound that might take care of it right there. (in some circuits this will be as simple as jumpering with a resistor, in others this may sound like crap). The bypass capacitor, already mentioned, may be quite effective, depending on the offending signal. If it's actually rfi, since you are only interested in the complete other end (20-60 or 20-100) you can even go a little heavey with the capacitance, if it does something you don't like you'll know it. If the problem goes away and nothing sounds bad, put the textbook down and wait for problems to come to you. You could try your own little filter just as close to the input of the subwoofer as possible (inside really isn't too close.) Put a small cap across the line then interupt the signal wire with a small choke then put a second cap across the line. (by line I really mean input) You also might make your own cable. Get a length of shielded twisted pair. Ground the shield at your preamp only. Use one from the pair as your hot and the other as your ground. This way you will have the advantage of a twisted pair, You will have a grounded shielded . But current won't flow through your sheild and be induced onto your signal wire. This is done frequently in instrumentation, both isolated and not. If none of this works there are many things you could have done that day that would have cost you much more. Seriously, somewhere in there I believe your problem will go away.

At this point if you still have the offending signal, I'd want to know what is wrong with using the speaker level inputs. There are still options but they might involve transformers or some other signal conversion so unless your reason for not using the speaker level inputs is more than theoretical it might be a good place to punt.

You could move the subwoofer amplifier,but I'm guessing that if you weren't opposed to that it probably would have already occured to you. I expect you will have abated your problem at this point.

[doyall]

Good suggestions all. Moving the sub didn't help the RFI. I did make a LFE cable with stranded, twisted wire with a drain and some EMI was tamed but it did nothing for the RFI. Got lots of stuff to try next. Now I only need to find the time... Thanks everyone.

[Thom]

We use twisted pair shielded for plc communication and where it has a major amount of noise to deal with (larg DC motors) 4000hp wound rotor ac motors, 2500hp motors on vfd's sometimes we resort to double shields. Not all shields are equal and guaranteing that your shield does not carry current can be a big deal. If possible grohnd the shield at earth ground and only connect signal ground to one of the twisted pair. you're not interested in any high frequency at all. If this signal is seperate from anyware else get adventures with rour bypas caps. Does it have a high level lo level switch? Are you doing anything different at all when you run it lo level, besides pluging in the line? if it's coming in on that line, you sure would think a pi filter would shut it down. There has to be an impedence that will kill it. You could find that and see whether it's one you can live with or not. There is no doubt that carried to extreme the things mentiond will get rid of the signal. It shouldn't be too hard to find out if this happens while you have a signal you like, or not.

[Ian Mackenzie]

If you can post a schematic of the sub amp it may prove useful.

Ian

[doyall]

"If you can post a schematic of the sub amp it may prove useful."

It is a JBL E250P. Can't find a schematic in the Links forum.

"Does it have a high level lo level switch?"

Not really ... it does have a switch that tells it if the low level is LFE or normal stereo low level imput though.

"Are you doing anything different at all when you run it lo level, besides pluging in the line?"

Nothing except flipping the switch to LFE operation.

I am not going to let this get the best of me. I have enough suggestions to keep me busy until I find it.

[duaneage]

I'd go for a ground rod. Get 6 - 10 feet of copper pipe and drive it into the ground, taking care to avoid water lines, power, and phone wires. Run 10 or 12 gauge wire to the equipment. You can get a pipe clamp for grounding at Home Despot or Lowes

[Thom]

I really hate showing my ignorance, even if you would think I'd be used to it by now. I'm drawing a complete blank on "LFE".

As far as driving your own ground rod, sure if it hasn't already been done for you but I guarantee it has. If it wasn't done intentionaly, your water oipe almost for sure goes underground more than your ground rod was going to. I can see it being fixed but in July it starts messing up again and you have to irrigate.

[Chas]

I really hate showing my ignorance, even if you would think I'd be used to it by now. I'm drawing a complete blank on "LFE".

As far as driving your own ground rod, sure if it hasn't already been done for you but I guarantee it has. If it wasn't done intentionaly, your water oipe almost for sure goes underground more than your ground rod was going to. I can see it being fixed but in July it starts messing up again and you have to irrigate.

LFE = Low Frequency Effects

[duaneage]

As far as driving your own ground rod, sure if it hasn't already been done for you but I guarantee it has. If it wasn't done intentionaly, your water oipe almost for sure goes underground more than your ground rod was going to. I can see it being fixed but in July it starts messing up again and you have to irrigate.
The traditional cold water pipe grounds are good for safety but if your getting RFI then it's not good enough. If fact, the cold water pipe could be part of the problem. An isolated ground of good quality prevents eddy currents and radio-phonics caused by other equipment sharing it. Motors, flourescent lights, electric heaters, computers and the like can modulate voltages on a ground line.

Take it from someone who set up sensitive comm gear in the Air Force, a good separate ground is best.

[Thom]

He,s not having problems with those things. He's trying to keep one radio station out as I understand it.

as far as LFE I still don't know what that means. Is that the encoded chanell for the 6.1 or 7.1 or whatever? Sorry I mostly play music.

I know if you have a station comming in on your phono ( no mine's packed away someware) and your using a 47k input, switch to a moving coil with about 20 ohms and even though you add a stage of amplification, by by radio station. Now if it's close enough and powerful enough that it comes in on your door knob, that's different.

[Chas]

as far as LFE I still don't know what that means. Is that the encoded channel for the 6.1 or 7.1 or whatever? Sorry I mostly play music.



Yes, it's the ".1" the dedicated sub output for 5.1, 6.1 or 7.1.

[doyall]

If all else fails, a ferrite core installed on the LFE cable might help, for example, this one from Radio Shack;

http://www.radioshack.com/product/index.jsp?productId=2103979&cp=&pg=2&y=3&origkw=rfi&kw=rfi&kwCatId=2032058&x=18&numProdsPerPage=100&parentPage=search

This did the trick. Simple, cheap and effective. Couldn't ask for anything more. I may try some of the other stuff just for fun when I get the time. And I will probably get some more of these for other equipment as well. Thanks.

[Thom]

The only reason for buying the expensive snap together one is if it would be a hassle to break the line. Otherwise you should be able to buy a sack full of doughnuts for that price. Glad your problem's fixed.