RFI Noise??

54d18

I have had a Rogue 3024 for about 2 years now, and find it to be rock solid.
The only problem I have encountered is the interferance that occurs on
our Marine Band VHF radios. The main frequency is channel 16 at 156.8 Mhz.

I have wired the solar panels using armoured cable (Teck) hoping that the
armour would offer some shielding, also, I have installed a metal torroid of about
2" inside diameter with many turns of both pos and neg leads going in the same
direction immediately outside the Rogue controller inside of a 8x8 metal junction box
using a metal nipple between boxes with ground bushings on both ends of the nipple.

I would like to try ferrites, I have seen them in the Online Store of this site, but would like
some insight on using them, mainly:

Should both pos and neg go in the same ferrite?
Should both pos and neg have sepparate ferrites?
Should I buy the largest dia and put as many turns as possible it?

Any insight on this would be greatly appreciated,

Troy

BB.

Re: RFI Noise??

Yes, the + and - cables should go through together (and wrap together, if room). Larger DC current with "saturate" the iron/ferrite core and make, essentially, non-magnetic in terms of killing your RFI.

When you pick a ferrite--you need to look at the material itself and make sure it is "active" in the frequency range of interest.

-Bill

inetdog

Re: RFI Noise??

BB. wrote: »

Yes, the + and - cables should go through together (and wrap together, if room). Larger DC current with "saturate" the iron/ferrite core and make, essentially, non-magnetic in terms of killing your RFI.

When you pick a ferrite--you need to look at the material itself and make sure it is "active" in the frequency range of interest.

-Bill

One small addition to that one: Putting both + and - through the same ferrite will cut back the "common mode" interference which is equal in amplitude and phase on both + and - outputs. But if the - lead is grounded, all that the ferrite will do is make the same amount of RF current flow in both the + and - wires.
If you can work out a situation with a reduced number of turns which can avoid saturation of the ferrite (the same ferrite which is good for 20 Amps AC should also be able to handle 20 Amps DC) then you should also put a second ferrite on the ungrounded (if there is one) lead only. Luckily this will only have to handle the panel current and not the battery output current.


Probably the best solution available to you on the battery side is to keep the wires to the battery as short and large as possible and twisted, and perhaps put a capacitor (of the appropriate voltage and sized for the frequency range of interest) from + to - just on the battery side of a single ferrite. At lower frequencies the chances of radiation on the battery side are low but at 156 MHz, you could be getting significant radiated noise unless the wires are only a few inches long. The batteries should act as a capacitor to keep interference from passing out the other side on the load wiring, but they may do nothing to reduce radiated RF from the connecting wires. Keep them short and twisted and maybe even put a shield around them.

I would not worry about radiation from the panel wires, as long as you can keep the noise generated by the CC from reaching them in the first place. Shielding them on the way to the panel will not help with radiation from the panel elements themselves.

And finally, if you have an amateur radio friend, ask him about designing a "trap" specific to the 156.8 MHz frequency you are having the most problems with. It can be a simple as single properly cut piece (exactly 1/4 wavelength at 156.8Hz) of coaxial cable with the end left open and properly weather-sealed.

niel

Re: RFI Noise??

first of all do you know if the interference is coming from the leads that you put the toroid on or not? it could radiate out of the pvs leads as well as the battery leads and a small marine ht held to the leads may disclose the source area. also be very sure of connections and check for any corrosion or even bad solder jobs on any wires as this could make things far worse. the ht can be used to sniff out the interference and if it is too strong then try removing the antenna to better pinpoint it.

don't rule out a problem with the radio's antenna or coax connections. rfi is tough to nail down, but perseverance may pay off. a question i have is does the interference go down with less solar output or not?

Vic

Re: RFI Noise??

There is this exhaustive (or perhaps exhausting?) Thread:
http://forum.solar-electric.com/showthread.php?5887-Radio-Noise-from-PV-System

Second, if the main radio that you are using is a base station type set, you could try removing the antenna coax at the radio, open the Sqelch, and note if the noise is; reduced, gone, or about the same. This will reveal weather the interference is Radiated, or Conducted noise. If radiated, it might be possible to move the antenna, altho, if on a boat, the common vertical antenna it is probable that the antenna cannot be moved far enough away from the electronics and wiring to make a difference.

And, guess that you KNOW that the interference you are sleuthing, is actually coming from the CC (?). Obviously, switching off the input and output breakers should tell this story ... know that you know this.

Sometimes, conducted noise can enter the radio itself (if it is a base station radio) via the power leads, so in this case you might try filtering the power leads to the radio, where these leads connect to the power supply, or battery.

For either conducted or radiated, Common-Mode Chokes, and shielding of conductors should help. As noted, a small disc ceramic capacitor across the PV In terminals and another of ht Battery terminals right at the CC could help. In the case of the PV in capacitor, make certain that the capacitro can tolerate the max possible PV voltage. Vic

niel

Re: RFI Noise??

vic,
just so there's no confusion here, he did not say he was using an ht (handie talkie) and is most likely using a full sized marine radio. my suggestion was to use an ht if he has one to use to sniff out the interference. this would also say if it was a conducted interference or not as well if the ht did not get any interference. the ht should be running on its own internal battery of course so as to not pick it up off of the battery wires if that's the source.

the suggestion of disconnecting the antenna was a good one too.

the question i asked before is also an indicator of whether or not solar, aka the cc, is the cause or not.
"a question i have is does the interference go down with less solar output or not?"
this is because when a cc handles more current the noise generated should also be stronger. if it would be a case of not being dependent on the pv current to generate the noise from the cc then disconnecting the pvs while the cc is still connected to the batteries and on would indicate that. i would normally expect it to disappear when the pvs are disconnected.

Vic

Re: RFI Noise??

Yes, niel,

I had NO confusion what YOU were saying, YES, HTs and Hand-held RXes are great tools, as are portable AM batt operated radios -- great sniffers.

In my limited experience, PV cables are the dominant RFI source, as, traditionally, these leads are the longest and generally most exposed. Batteries are low Z, and leads generally short, but at VHF, of course, the batts are not teriffic capacitors.

YMMV, Good Luck, Vic

54d18

Re: RFI Noise??

Wow, as usual, many excellent responses, gotta love this forum.

Just to add to the information I had listed intially, the interferance
disappears after the sun goes down and the controller goes into sleep mode.

The radio being used is a base type unit, and the interferance also exists
on my handheld radio, even when I am outside, and a significant distance
away from the pole where the panels are mounted, probably 100-150' at least.

I have ordered some ferrites from the online store here, and would like some insight
on how to configure them, they are listed up to 300Mhz, I hope that I can
get rid of that nasty noise.

Someone here mentioned or questioned if the interferance was worse when the
unit was at a lower solar input, and the answer is yes, when the unit is at max output,
there seems to be less interferance.

Troy

niel

Re: RFI Noise??

"Someone here mentioned or questioned if the interferance was worse when the
unit was at a lower solar input, and the answer is yes, when the unit is at max output,
there seems to be less interferance."

that's odd as i would've expected it to be less at lower solar input or am i reading you wrongly? anyway, read the other thread on the mx60 noise and that should help some. do know you may or may not cure it. also, it may not hurt to get a hold of marc at rogue and see if he can be of assistance too.

RCinFLA

Re: RFI Noise??

Looks like their name may be appropriate.

I did several related word searches in their manual for items relating to FCC certification or UL certification and found no hits.

Are there any labels on the unit's case indicating either type of certification?

2manytoyz

Re: RFI Noise??

A quick check online and your controller appears to be a MPPT type. I'm using a couple of ProStar 30M controllers, which are PWM. These too produce some RF interference. I tried to capture the video with a low resolution video camera, but it did a poor job of showing the noise on the computer screen of one of my security cameras. I used a ferrite to greatly reduce the noise. Initially, I used a split ferrite to test the fix first. It actually caused the two pieces to buzz in my hand, making a lot of audible noise.

[video=youtube_share;ffXBcXPQ9gc]http://youtu.be/ffXBcXPQ9gc[/video]

I later installed a one piece ferrite on both the positive and negative cables as close to the controller as possible. This didn't fix the issue 100%, but the result was satisfactory.



Good luck with your gremlin. Please post any results, good or bad.

niel

Re: RFI Noise??

robert,
i never gave it a thought that you may hear the 2 separate halves vibrate audibly, but as the pwm cc is typically functioning in the audio frequency range it can very well do this and you proved it. the solution was a one piece ferrite or toroid as you've indicated. even if the vibrating would not bother you, the fact is they would be hitting each other at at xxx hertz and could eventually weaken and crack the ferrite halves.

i couldn't see the results you got on the screen too well.

54d18

Re: RFI Noise??

2ManyToys, I see from your photos that you are using 2 ferrites
each installed separately on the pos and neg wires, is this what
I should be trying when mine come in, or should I try both pos and neg
through the same ferrite?

I am anxious to get to the bottom of this, I've procrastinated long enough on this issue.

T

inetdog

Re: RFI Noise??

54d18 wrote: »

2ManyToys, I see from your photos that you are using 2 ferrites
each installed separately on the pos and neg wires, is this what
I should be trying when mine come in, or should I try both pos and neg
through the same ferrite?

I am anxious to get to the bottom of this, I've procrastinated long enough on this issue.

T

MY recommendation would be separate ferrites which can take high current (high magnetic field) on each wire, and possibly a larger ferrite with both leads run together through it, with more than one turn if possible, since the net DC magnetic field will be zero.

nsaspook

Re: RFI Noise??

niel wrote: »

robert,
i never gave it a thought that you may hear the 2 separate halves vibrate audibly, but as the pwm cc is typically functioning in the audio frequency range it can very well do this and you proved it. the solution was a one piece ferrite or toroid as you've indicated. even if the vibrating would not bother you, the fact is they would be hitting each other at at xxx hertz and could eventually weaken and crack the ferrite halves.
.

My solution for the split shell ferrite vibration noise was to inject RTV into them to fill the voids. Another product that might help is a cable wrap. http://www.kitagawa-ind.com/rs_up/pdf/ECBR-CF_shohin3.pdf

I have some scrap (10 feet plus) from a deinstalled tool at work that I could send to someone with a RFI/EMI problem to try. http://farm9.staticflickr.com/8038/8031700393_711885b601_c.jpg

PM me.

BB.

Re: RFI Noise??

ideally, the two halves should touch (bundle tie, snap housing, etc.)--Air gaps reduce the amount of flux going through the ferrite.

-Bill

nsaspook

Re: RFI Noise??

BB. wrote: »

ideally, the two halves should touch (bundle tie, snap housing, etc.)--Air gaps reduce the amount of flux going through the ferrite.

-Bill

Ideally, Yes...

The small gap between has little effect on the magnetic performance as the total impedance in hardly effected because the flux lines will deflect into the cores away from the gap unless the gap is a sizable fraction (10%+) of the wavelength of the signal. (assuming the overall coverage is close to 100%)

2manytoyz

Re: RFI Noise??

54d18 wrote: »

2ManyToys, I see from your photos that you are using 2 ferrites
each installed separately on the pos and neg wires, is this what
I should be trying when mine come in, or should I try both pos and neg
through the same ferrite?

I am anxious to get to the bottom of this, I've procrastinated long enough on this issue.

T

RF is one of those black magic topics. I've had to deal with a bit of it at work when working on radio gear. Even my lead engineer, who had a lot of experience in this topic (and formal training), still found that it was a trial and error type thing.

Just recently, we installed a noise cancelling system inside a radio used by propellant handlers. Very noisy inside the suits from the airpack being worn. The small circuit eliminates the transmitted noise on the communication equipment. The wires from the noise cancelling board were run through a ferrite. On some radios, it worked great, others not so much. The radio case would act like an antenna, and would do weird things depending on how close it was to a body. Upon further examination, the ferrites that had the wires wrapped in a messy fashion worked great. Those that were neat, had the wires color coded and paired together (grounds next to each other, same with positive power), did the weird antenna thing. Turns out, it worked out better to randomize the wire going around the ferrite. Less likely to induce energy onto other wires.

Only reason I share this is to give a reason to the answer of... "I just don't know". Without trying it, can't guarantee the results. The good news is ferrites are cheap. I'd probably start with the one piece ones that slip over each of your two cables right at the charge controller.

The split ferrites I initially tried came with a small metal clip that holds to two halves together. Obviously the wrong product for this application.

Here's the exact ones that worked with my setup: http://www.newark.com/fair-rite/2643102002/solid-round-emi-suppression-core/dp/24C4674 Just keep in mind that mine is a PWM type, but I still believe these would likely improve your issue. Unless you have the ability to measure the frequency of your noise, and buy the specific ferrite for that frequency range, it's back to the trial and error method.

Good luck!

Vic

Re: RFI Noise??

It is generally considered that Type 43 material is the best for the RF frequency range of the OP's birdie/noise.

In my experience, 43 is also the most common, especially for these cable style large beads. Unfortunately, there are no specific makings on any of these parts, FWIW.
Opinions, Vic

2manytoyz

Re: RFI Noise??

Vic wrote: »

It is generally considered that Type 43 material is the best for the RF frequency range of the OP's birdie/noise.

In my experience, 43 is also the most common, especially for these cable style large beads. Unfortunately, there are no specific makings on any of these parts, FWIW.
Opinions, Vic

Hey Vic, turns out the ones I'm using (and posted a link to) are type 43 material. Specifically for "Broadband Frequencies 25-300 MHz". There's a P/N on my ferrites, so I was able to find out the specs.

I have several of the engineering kits by Fair-Rite at work, which made trying various types easier. I don't have enough background on this topic to have picked type 43 material based on prior experience, but I often get lucky. Such was this case. :cool:

BB.

Re: RFI Noise??

"I'd rather be lucky than good."
Lefty Gomez - Wikipedia, the free encyclopedia

-Bill

Vic

Re: RFI Noise??

It IS good to be lucky.

Believe tha Eng Kits hves P/Ns on individual parts.. Have never seen any #s at all on production Ferrites. But, of course the "Iron" and other low frequency power toroids are often color-coded.

Perhaps things have changed, would seem that an inkjet, or even laser could put identifying marks on ferrites.

Keep up the good luck, Vic

2manytoyz

Re: RFI Noise??

When I ordered replacement parts for the kits, this is how they came:



I haven't ordered them in bulk, maybe 10 at a time. The smaller ferrites didn't have numbers on the parts, just on the packaging.

'tis good to be lucky! :cool:

54d18

Re: RFI Noise??

OK, I'm back from the camp, and tried the ferrites on my system.
When using a handheld radio as a sniffer, I was able to track it
down to the solar panel leads.

I used 2 ferrites with 4 turns through them, 1 on pos and 1 on neg,
installed right at the controller, then the wires go through a short
nipple into a metal enclosure, where they do the same thing through
another pair of ferrites, then they both go through the same ferrite
and then through another the same.

The result was not conclusive, at first it seemed ok, then more interference,
only on channel 16 (156.8 Mhz)

So, long story short, no success yet.

T

niel

Re: RFI Noise??

do you have access to ceramic disc capacitors either from a store like radio shack or an old junk tv set? if so try placing a cap across the controller input leads and try with and without the ferrites. .001uf to .005uf should do well. even 500pf is ok to try. just remember you don't want to short across the leads as you have the full pv power across there in a tight space so be careful.

Vic

Re: RFI Noise??

Smallish value caps are often better at VHF. Something like 50 to 100 pF, Keep leads as short as possible. Connect them as close to the noise source as possible -- in this case the CC Vin leads. Just make certain that the voltage raging of the capacitors is several times the Maximum PV voltage you would ever expect to see. Many current vintage Disc Ceramic capacitors have very low voltage ratings. It might be possible to solder the capacitor's leads across any MOVs that happen to be on the Vin terminals.

RFI suppression has a bit of mystery to it, as until one knows just what is going on, suppression can be difficult.

Much of this has been noted before.

Normally, moving antennas as far from the noise source works well, but believe that this is not much of an option in this case.

Good Luck, Vic

54d18

Re: RFI Noise??

I tried a 50uf Eelectrolytic cap, but I suppose
a ceramic type would be better to short the AC.

The values indicated in the earlier posts seem small, but
then again I really do not know, I just thought that larger
would have been better, but having read the posts, I will be
getting some to try for the weekend coming up.

How about the ferrites? Should I try the same configuration?
I am thinking of getting a scope to check the interference,
I could probably use one from work.

T

Vic

Re: RFI Noise??

Hi T,

Yes, an electrolytic is very very inductive, which makes these parts useless at HF, and escpecilally at VHF. Even high value Ceramics do not have good performance at VHF. The material that allows higher C value ceramic caps performs poorly at VHF in my experience. These are generalizations that apply to lower cost ceramic caps used in mass-produced products. There are some high quality high value ceramics that have very low inductance and will pass large currents, but they are much more expensive than Radio Shack style ceramics.

I would leave the Ferrites as you have them, and if a 30 to 100 pF cap does no good, then perhaps ditzing around with the ferrites may help you figure out what is needed. Any scope will need to have fairly good bandwidth. You will probably be trying to look at the edges of some AC signal riding on the PV In leads. It seems that what you really have is a Birdie, not just broad-band noise. Trying to see this on a Lab Scope will probably be difficult. There could even be an Image problem within the radio in question. These things are not always straightforward. Forget if you have tried a different radio Transciever to see. This stuff is fun, but can be a challenge.
Best of luck, Vic

BB.

Re: RFI Noise??

I would try putting both + and - leads in parallel through the ferrite (with a couple of turns, if possible).

You might be saturating the ferrites with multiple turns of DC current (through one leg).

Putting the +/- leads in parallel cancels out the DC magnetic field and will prevent the ferrites from saturating (and hopefully work better at killing common mode RF energy).

-Bill

niel

Re: RFI Noise??

i agree it is most likely a birdie just as the mx60 had one in that area as well. a scope may help to give direct images of the birdie strength as components are tried.

btw vic. i was going by a 0 ohms impedance rating for the cap and would roughly fall on a value near .001uf and would represent a short circuit at about 150mhz when across the leads. (graph reference to older arrl amateur handbook) it may be necessary to use even more than 1 or 2 of these caps to offer some good attenuation of this birdie. the scope may prove invaluable if he knows how to use it.

54d18

Re: RFI Noise??

OK, a quick check around and in an old parts pail,
I have a .01K and a .005u

I'll check for some more on some old boards at work.

Thanks for all the pointers, hope to get to the bottom of this...
This site is awesome.