Your Universal Remote Control Center
RemoteCentral.com
Complete Control by URC Forum - View Post
Previous section Next section Previous page Next page Up level
Up level
The following page was printed from RemoteCentral.com:

Login:
Pass:
 
 

Page 1 of 4
Topic:
*THE* solution for RF interference on MRF repeaters
This thread has 59 replies. Displaying posts 1 through 15.
Post 1 made on Sunday February 4, 2007 at 15:25
HTBruceM
Long Time Member
Joined:
Posts:
February 2007
24
Long story short: put FERRITE COREs on your IR blaster wires, close to the MRF.

(Duplicate posting over in the other URC forum)

Here's my sad story:

I've had severe RF interference for some time now on my MRF-250. Seems like no matter where I placed the MRF, there would be some new interference cropping up over time causing intermittent operation. To make matters worse, I have the Motorola DVR from Comcast which is notorious for periodically going non-repsonsive to input commands (either IR or front panel) while queuing up those commands and then finally getter around to executing them more or less all at once.

It was so bad that I had decided to just forget about the whole RF thing and go back to IR exclusively. I just purchased the Spearkercraft IR repeater kit with the intention of converting everything back to IR. Being the engineer that I am, I just couldn't admit defeat and let this damn RFI problem kick my butt. It's just RFI, people have been effectively dealing with RFI for decades. So before I ripped out my existing MRF blasters, I thought I'd give one more shot to isolating the source of the RF in my case.

I first eliminated the power supply path by wiring up a 9V battery on a short cable into the MRF - no change there. Still a bunch of RF interference.

I even tried the whole tin foil thing with the intent of reducing the effective sensitivity of the MRF RF receiver. Figuring the remote control RF signal was hot enough to overcome the attenuation from the foil. Very marginal results here as well.

Then it finally occurred to me... what about all those long IR blaster wires running all over the place? So I began to disconnect them one by one while watching the MRF RF activity LED (in the channel-0 position). Sure enough, there was a direct correlation between those blaster cables and the degree of RF interference. The long IR blaster cables are acting like a bunch of antenae, picking up all kinds of RF noise and apparently injecting that noise into the MRF unit.

OK, I thought, now I'm getting somewhere. So, how do you reduce RF being introduced on these cables? Filters. Either capacitor to ground or series coil. I first thought about getting a small enough capacitor that would act as a high-pass filter, and just effectively short all that high frequency to ground. Probably would work. But then I realized the URC engineers must surely have thought about that and include those small and inexpensive parts internally in the MRF. Right? Anyway, after talking to some other colleagues, they suggested ferrite cores on the cables.

Duh. Simple and cheap solution that wouldn't require ugly soldering and stuff. So I just found a few of these <$1 goodies and looped each IR blaster wire around each core a couple times. This creates a series inductor. BAM - RF interference is gone completely. It's actually hard to get any RF interference now. The whole thing just became rock solid. While your mileage may vary, this certainly has worked for me. And it's a mystery why the talented engineers at URC haven't thought of such a simple solution for their products.

So I"m passing this info along to the experts in this industry. I urge you to try this simple and CHEAP solution. You can find cores like this at your local electronics parts store, or online at places like digikey:
[Link: dkc3.digikey.com] (clamshell type at bottom of page).


URC - if you happen to be reading this and would like to thank me for solving your design problem, I would be happy to accept upgradeable software for my MX-850. I purchased LOCALLY (not over the internet) from a custom installer who I later discovered is not an authorized dealer. This only recently became an issue for me when you made your liveupdate policy change :)

Last edited by HTBruceM on February 5, 2007 00:07.
Post 2 made on Sunday February 4, 2007 at 18:06
cma
Super Member
Joined:
Posts:
August 2003
2,694
I thought this was the solution:

[Link: remotecentral.com]

I can't seem to keep up with what the real solution is....
OP | Post 3 made on Sunday February 4, 2007 at 20:24
HTBruceM
Long Time Member
Joined:
Posts:
February 2007
24
Well it wasn't hard at all for me to isolate the source of my problems. Put the MRF into it's RF "sniffing" mode, where the activity LED lights up on all RF signals in the range it works in. You just put the ID switch to 0. Then place it where you get your worst interference (i.e. the LED lights up brightest). Then disconnect the IR blaster cables one at a time. That caused the LED brightness to reduce significantly - which indicates the blaster cables are the root cause.

There may be cases where the incoming DC from the MRF power supply also has RF noise. Putting a very small value cap across the incoming DC should attenuate any high frequency noise on that line. A ferrite core might also be useful here.

But as I indicated, my problem wasn't from the DC power cable. I confirmed this to a large degree by using a 9v battery in place of the power supply to eliminate any possibility of noise coming in from the AC power & through the transformer. That provided NO improvement - no change to my RF interference.

Adding a small value cap to the IR outputs was a possibility, but choosing the right value would be a trial & error process. You would have to find the right value that wouldn't affect the signal to the IR emitter. The ferrites have 0 effect.

One last note... After digging around in my parts drawer, I found the IR cables supplied with my Mitsubishi DLP. Mits has a feature where they can control other equipment via software in the TV. They drive IR blasters directly from the TV. Guess what those cables had on them? Yep, ferrite cores right at the connector to the TV.
Post 4 made on Sunday February 4, 2007 at 21:01
Dave E
Long Time Member
Joined:
Posts:
September 2005
282
Thanks for the information.
What receiver are you using with the RFX-250? The MRF-300, MRF-350 or the MSC-400?
Experience is what you get when you don't get what you want.
Any wire cut to length will be too short.
I must be a near GENIUS. All my teachers told me I was at the very PEAK of the bell curve!
OP | Post 5 made on Sunday February 4, 2007 at 22:45
HTBruceM
Long Time Member
Joined:
Posts:
February 2007
24
I am using the MX-850 remote with the MRF-250 RF Base Station. The remote is set to generate RF to all components except the TV which is IR only.
Post 6 made on Sunday February 4, 2007 at 22:53
JohnnyRose
Long Time Member
Joined:
Posts:
January 2007
37
Thanks!

I will give this a try.
Post 7 made on Sunday February 4, 2007 at 23:01
JohnnyRose
Long Time Member
Joined:
Posts:
January 2007
37
What type of cores did you use?

I found this site but its WAY over my head.

[Link: us.digikey.com]
OP | Post 8 made on Sunday February 4, 2007 at 23:36
HTBruceM
Long Time Member
Joined:
Posts:
February 2007
24
Ferrite cores are usually found on the ends of computer VGA monitor cable, and for the same reasons.

I'm using something like the digikey p/n 240-2143-ND. This is a solid piece that looks something like a ring or tube. With this "solid" type, you need to make sure you have big enough inner diameter to fit the blaster cable's entire connector through it a couple times, looping the the cable through it at least one loop. I wish I could post a picture here....

The better solution might be one of the clamp-on types. These come opened up like a clamshell, and you loop your wire around inside one half of the core, then close & lock the clamshell. You can even bundle a pair of your IR emitter cables into one of these. Model HF70RU10X20X5S is from TDK and is the clamshell type, but it doesn't include the plastic casing - you can easily just tape it together with electrical tape. Their web page is [Link: tdk.co.jp] (see the PDF on the line called: For EMI Suppression / for round cable / Separated Cylindrical).

Just pick one that fits. You might also try your local elect parts store. Remember you need to loop/wind the cable around the ferrite core at least one turn.

Last edited by HTBruceM on February 5, 2007 00:30.
Post 9 made on Monday February 5, 2007 at 12:54
Lee Kropp
Long Time Member
Joined:
Posts:
May 2004
262
HTBruceM email me when you have time.

Lee
Post 10 made on Monday February 5, 2007 at 21:11
splogue
Founding Member
Joined:
Posts:
November 2001
342
This is funny 'cause I always put ferrite cores on my equipment, including the IR cables to the MRF and MSC. My wife thought they were snake oil until I pointed out that all our IBM Thinkpad power supplies had them. Then I sent her a link to an online article that explains induction and how they work and she nodded off. When she woke up, she agreed they worked and politely asked me not to bore her in that particular way again. ;-)

Basically, they stop a long wire from being used as an antenna, either to transmit or receive. So, I put them near the component on all power cables (to stop the component from transmitting) and near the MSC to stop them from being used as an antenna. The more times you can wrap the cable, the more effective they are. This is an exponential equation, so twice is *a lot* better than once. Just make sure you wrap them in the same direction.

They also help organize cables, which is always a plus.

And no, they aren't the perfect solution to every problem. But, they're cheap, and effective, so why not? If nothing else, use them instead of velcro ties on the last leg of the run. If the client / guest / wife asks about them, you can show off a little on your knowledge of electronic components and inductive circuit transmission theory.

Sean
"If you can't win, change the rules."
[Link: invelos.com]
Post 11 made on Monday February 5, 2007 at 22:54
wire pulling monkey
Long Time Member
Joined:
Posts:
August 2005
100
Wow what great fixes, The additional power supply and caps, now Ferrite cores

Seems the end users are better at this than the designers of the products.

Additional power supply and caps.....$12.89

Ferrite cores for Emitter lines..........$5.94

finally a working RF system........Priceless and smiling

Glad to see some solutions
I can get 7 RG6 cables and 5 Cat 5e wires in a 1" hole and I saved a ton of $$$ by switching to GEICO www.geico.com
Post 12 made on Tuesday February 6, 2007 at 14:43
splogue
Founding Member
Joined:
Posts:
November 2001
342
Ha! I just ordered a bucket of caps and the soldering iron is ready -- wish me luck!
"If you can't win, change the rules."
[Link: invelos.com]
OP | Post 13 made on Tuesday February 6, 2007 at 15:59
HTBruceM
Long Time Member
Joined:
Posts:
February 2007
24
I think caps between the two emitter wires (or from each to ground if one isn't ground already) should work also. Just need to choose a value that is just small enough uF to not affect the <1mhz switching rate of the emitter signal. I would think something around .01uf maybe?

Be sure to post your results and in a particulary noisy RFI scenario.
Post 14 made on Tuesday February 6, 2007 at 23:55
JohnnyRose
Long Time Member
Joined:
Posts:
January 2007
37
Bruce,

It would appear that Ferrite Cores filter out different frequencies based on the size, shape and material of the core.

Is the RFI that causes problems with the MRF250 at the same frequency that the MX3000/MRF250 uses (other URC products)? Or is the MRF250 susceptible to frequencies in addition to the frequency of the MX3000 which is causing the interference problems?

If certain cores are capable of filtering specific frequencies, it would seem logical to select a core specific to the frequency causing the problem. Or if all RFI is to be eliminated selecting a core that does that (if one exists).

Im not an engineer and Im trying to make an informed decision about which cores to buy. Not that its a monetary concern at $1 a piece but if Im going to spend the time to do it, I want to do it correctly. Any information would be appreciated.

John
Post 15 made on Wednesday February 7, 2007 at 20:50
splogue
Founding Member
Joined:
Posts:
November 2001
342
Keep in mind that what you are really trying to do is stop the IR emitter lines from being used as antennas, either transmitting *or* receiving. So, ideally you would want to filter (attenuate) those frequencies that the MRF responds to so that it won't pick up noise or reflected signals on the IR lines that block out the real signal coming in through on the antenna.

Unfortunately there is almost no way to know if the cores you are buying will help filter out the frequency range you need, unless the seller provides specs or a model number so you could look it up. At least they will probably help clean up some of the extraneous noise, which should help. They filter out a wide band, not a narrow one, so that makes it more likely they will work for your application.

I've had very good luck buying batches on eBay so far. I don't remember exactly how much they were, but $1 each seems high. You may want to poke around on there a bit to get a feel for the prices and sellers before buying.
"If you can't win, change the rules."
[Link: invelos.com]
Find in this thread:
Page 1 of 4


Jump to


Protected Feature Before you can reply to a message...
You must first register for a Remote Central user account - it's fast and free! Or, if you already have an account, please login now.

Please read the following: Unsolicited commercial advertisements are absolutely not permitted on this forum. Other private buy & sell messages should be posted to our Marketplace. For information on how to advertise your service or product click here. Remote Central reserves the right to remove or modify any post that is deemed inappropriate.

Hosting Services by ipHouse