rf interference

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
v6a1a4
Posts: 26
Joined: Tue Apr 23, 2002 1:01 am
Location: Vancouver
Contact:

rf interference

Post by v6a1a4 » Wed Aug 10, 2005 10:36 pm

I recently built a circuit using a RF module from RF Digital. It uses a key fob for a transmitter (RFD28005) and a small board with a Linx module running 418Khz for a receiver (RFD28004). The circuit is simple enough, a darlington is triggered which in turn triggers a relay (which has a reverse biased diode across the coil). When the relay has nothing connected to the contacts the unit works like a dream with a range well over 75 feet (line of sight). When I connect a 12v automotive light bulb, again works like a dream. When I connect a 12v, 10.5 watt solenoid to the relay contacts my range drops to about 5 feet and the circuit is very flakey. At first I thought the problem was with the solenoid generating the noise that is wrecking havoc on the RF, but now I’m thinking that it is the load on the relay contacts caused by the solenoid that is causing arcing that is wrecking havoc on the RF. Anyone? I am currently reading the April 2005 issue of N&V (I don’t have a lot of free time to keep current on me reading :) ) which has an article about contact arcing. They suggest a resister/capacitor in series across the relay contacts. The problem I am facing is that I stuffed my circuit into a way to small of a box and there is not a lot of room to start adding parts. Before I go ahead and try this fix does anyone have some thoughts? Am I even on the right track?<p>Cheers,
James in Vancouver

rshayes
Posts: 1286
Joined: Tue Mar 04, 2003 1:01 am
Contact:

Re: rf interference

Post by rshayes » Thu Aug 11, 2005 12:46 am

There shouldn't be much arcing due to the solenoid when the contacts are closing. The current at this point is relative low due to the inductive character of a solenoid. It takes some time, possibly several milliseconds, before full current flows.<p>The problem will occur when a flowing current is broken. This is when you get an arc, since the voltage jumps up high enough to srrike an arc in a small air gap.<p>If the contacts bounce when they are closing, you can also get an arc, but it will be less severe than the arc created on opening the circuit.<p>A diode in the reverse direction (cathode to positive) across the solenoid will limit the voltage and prevent an arc from forming. It will probably be smaller and easier to install than an RC network. The diode should be rated at a higher value than the solenoid current and for a voltage higher than the solenoid voltage. A 50 volt, 3 amp diode should be adequate in your case. A 1 amp diode, such as a 1N4001, would probably work, since the current flow is only for a very short time.

User avatar
jwax
Posts: 2172
Joined: Mon Feb 09, 2004 1:01 am
Location: NY
Contact:

Re: rf interference

Post by jwax » Thu Aug 11, 2005 3:34 am

To dodge the problem, use a MOSFET in place of the relay?

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: rf interference

Post by Robert Reed » Thu Aug 11, 2005 9:26 am

Most solenoid circuits operate with a momentary quick punch of HEAVY current to move its actuator to new position. Once accomplished,it can it can sit there at 0 current until the next action is called for. If your circuit has a common power supply for all, solenoid current requirements even if momentary may be pulling your B plus voltage down far enough to srew up total circuit operation.

Gorgon
Posts: 325
Joined: Wed May 04, 2005 1:01 am
Location: Norway
Contact:

Re: rf interference

Post by Gorgon » Thu Aug 11, 2005 12:08 pm

Hi James,
If you power all electronic and the solenoid from the same power supply it may be that the solenoid kicking in, clamps the voltage some time before it stabilizes. To minimize the effect, try to isolate the electronics supply from the solenoids.<p>The easiest thing to do is to supply the elctronics via a diode and an extra reservoir capacitor. The diode will stop the discharge of the capacitor and the electronics will not suffer from brown out.<p>You dont't say anything of what type of power you use. If it is a labpower it might be that it has a too high internal resistance due to the current foldback circuits. Try to put an electrolytic capacitor of some size over your power supply.<p>But, a diode and an electrolytic capacitor on the electronics should do the trick, if that's the problem.<p>TOK ;)
Gorgon the Caretaker - Character in a childrens TV-show from 1968. ;)

User avatar
Chris Smith
Posts: 4325
Joined: Tue Dec 04, 2001 1:01 am
Location: Bieber Ca.

Re: rf interference

Post by Chris Smith » Thu Aug 11, 2005 9:40 pm

Your coils on the relay may be absorbing too much RF. Seperate them by at least one foot. <p>Try wrapping that section [the relay] first in paper, then in foil to shield this possibility. <p>Also make sure you dont have any sloppy wiring going from one to the other.

v6a1a4
Posts: 26
Joined: Tue Apr 23, 2002 1:01 am
Location: Vancouver
Contact:

Re: rf interference

Post by v6a1a4 » Thu Aug 11, 2005 10:55 pm

Thanks for all the input, the solenoid power is completely separate (9v battery for the rx and a gel cell for the solenoid) from the other electronics, I do have diodes on both the relay coils and across the solenoids and yes the solenoids are within a foot of the receiver. This is almost unavoidable with the setup.I will try shielding the RX and place the RC network across the relay contacts and let you know when/if anything works. This is a problem we often read about and it seems like such a no brainer, but until it rears its ugly head with a project you're working on you don't realize what a pain it really is.

Gorgon
Posts: 325
Joined: Wed May 04, 2005 1:01 am
Location: Norway
Contact:

Re: rf interference

Post by Gorgon » Fri Aug 12, 2005 3:45 am

Hi James,
Have you tried to move the solenoids away from the receiver, just to confirm the close-is-bad theory?<p>This may be far off and out, but have you tried to turn the solenoids in all three axis, to minimize the magnetic flux influence?<p>TOK ;)
Gorgon the Caretaker - Character in a childrens TV-show from 1968. ;)

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: rf interference

Post by Robert Reed » Fri Aug 12, 2005 9:24 am

Treebear
one thought has occurred to me and that is the noise you may be creating would fall under the general heading of "manmade noise". The interference produced is usually strongest iin the MF & HF portion of the frequency spectrum (0.3 to 30 Mhz).with rapidly diminishing power as we move into the VHF & UHF regions. At 418 Mhz You should be fairly immune to this type of interference from a purely RF standpoint. However, not knowing your reciever specs (and the reeiiver is everything in a communication system) such as r.f. amp's 1 db compression point (in DBm), and overall front end selectivity, its hard to say the level of interference that manmade noise would cause. Still even in a poorly designed receiver, I think you would maintain fairly good immunity at your operating frequency. My point being ,that the interference may be doing its ' dirty wok ' at some other point in the circuit via some other path. With RF work---GOOD---grounding is essential.
Also in setting up your additional solenoid/apparatus, you didn't unintensionally block your teenee antenna did you?
These frequencys are very line of sight.
Solenoid should have little effect on your setup,unless of course it was resonant at 418 Mhz. Hardly likely.
[ August 12, 2005: Message edited by: ROBERT REED ]<p>[ August 12, 2005: Message edited by: ROBERT REED ]<p>[ August 12, 2005: Message edited by: ROBERT REED ]</p>

rshayes
Posts: 1286
Joined: Tue Mar 04, 2003 1:01 am
Contact:

Re: rf interference

Post by rshayes » Sat Aug 13, 2005 7:33 am

If you don't need continuous operation of the solenoid, you might consider triggering a monostable with the receiver output, and use the output pulse of the monostable to to drive the solenoid (through buffering and relays). Once the monostable starts, the output of the receiver will make no difference until the cycle is complete. You may need a second delay to gate out any noise the solenoid generates when it drops out.<p>Another possibility is an optoisolator between the receiver and the relay (assuming that the receiver operates on 9 volts and the relay operates on the 12 volt battery. This will allow breaking both the power and ground leads between the two batteries. Shielding the solenoid wouldn't hurt.<p>At 400 MHz a couple of inches of lead and a few pF of capacitance can form a resonant circuit. Sparks can generate small amounts of energy up into the GHz range. There once was a military signal generator (early fifties) that used a buzzer to excite a tuned cavity. A small signal could be obtained between 3 and 10 GHz. This was probably used for field testing radar receivers. Some optimist put a 0 to 100 dB variable attenuator on the output, but I doubt if anyone ever used that attenuator past half scale.

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: rf interference

Post by Robert Reed » Sat Aug 13, 2005 9:56 am

The highest frequencys generated by impulse type noise is related to the rise time of the step pulse that is produced in this situation. Unless a device is specifically designed for this purpose, the rise times are relatively slow concentrating most of their power in the low frequency portion of the spectrum.
Case in point--your neighbopr running an electric driill--Horrible interference on TV ch. 3 (54-60 Mhz) but UHF- probably nothing. And these channels are 4.5 Mhz wide making them more prone to interference. Your system should have about 25 Khz band width, which would allow lot less noise to pass through.<p>[ August 13, 2005: Message edited by: ROBERT REED ]</p>

Bernius1
Posts: 349
Joined: Sat Sep 21, 2002 1:01 am
Location: NY
Contact:

Re: rf interference

Post by Bernius1 » Mon Aug 15, 2005 12:33 pm

Just another thought; most FM receivers have a front end limiter. It's possible that the relay ckt is broadcasting noise loud enough to drop the gain in the front end. This would cause diminished range , while the ckt yet actually works. The only other thing I could think about is that the ground plane can sometimes be pushed by noise. You said that the two parts of the ckt are electrically isolated; grounds as well ? Regarding the cap, I'd think that a ceramic cap across the coil terminals would 'tank' any RF back into the coil. Am I right ? And any time delay in energizing the coil due to the ceramacap's charge time can be overcome with the electrolytic storage idea.
Can't we end all posts with a comical quip?

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: rf interference

Post by Robert Reed » Mon Aug 15, 2005 5:48 pm

Limiter stages in FM recievers follow the second IF stages and are there for one purpose--to remove any residual AM modulation at that point. Most quadrature detectors and discriminators are sucseptable to this, so it must be removed prior to demodulation. As a side benefit , limiters also allow a phenomena called capture effect (ability to lock on to the stronger of two signals).
There is no intentional "limiter" in the front of the receiver, but there is a circuit known as AGC (automatic gain control) which feeds back a DC sgnal to the front end to vary its gain. The stronger the incoming signal, the more the gain is lowered (AGC controlled gain is inversely proportional to signal from antenna).. The time constant of the AGC is not fast enough to react to one short burst of impulse noise and will therefore have no effect on front end gain.
The only "limiting" action that takes place in the front end is a signal from an adjacent channel(s) having enough energy to overload this amp rendering it inoperable. however this signal would never pass thru the IF stages, and neither would anthing else for that duration of time.
Which also brings up a point--is your system AM or FM.
I think most of us have been assuming FM .
I have no idea what"capacitor tanking back into coil" is, so I have no comment there.

Bernius1
Posts: 349
Joined: Sat Sep 21, 2002 1:01 am
Location: NY
Contact:

Re: rf interference

Post by Bernius1 » Tue Aug 16, 2005 9:22 am

Robert,
I'll take your word for it, but if the receiver is simpler, it may not have so complex an IF or AGC circuit, and I have seen earlier FM systems with AGC acting as a limiter early on to prevent front end overload.
Regarding the ground, PA systems get ground loops, and PCB traces ( even in the GND ckt) can develop a potential. Why do digital designers put a filter cap "as close to the power pins as possible" if not to prevent noise propagation ? I just figured, as rare as it may be, the the energizing coil was creating an RF potential or signal which was interfering with reception of the key fob. So, like filtering power pins, a cap across the coil terminals ( small, <.01uf ) might read any RF or reverse potential or spike during the time of inrush current and suppress it before it affects the detector circuits.
BTW, I'm not familiar with the named receiver card, are you ? And , if it's FM, is it susceptible to AM noise, as you suggested ? The interesting symptom to me is reduced range, but without failure. That tells me that either sensitivity is reduced , or eclipsed by noise.
Considering that RF is essentially monopolar until the detector adds a bias, the bias circuit's ability to reproduce an undistorted wave smells like rotten fish in Denmark.
Can't we end all posts with a comical quip?

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: rf interference

Post by Robert Reed » Tue Aug 16, 2005 6:17 pm

no-vice
I am not familiar with this receiver either which puts us at a definate disadvantage here.
Without proper design FM rec's are affected by AM modulation. An iteresting side note here is that production of Frequency Moduled RF is not possible without also creating Amplitude Modulation -- and vice/versa. I have seen the math to prove out this theorem, and beleive me I don't want to remember it. Just knowing the fact is sufficient.
As to most of your remarks, after reviewing, I guess we are both saying the same thing but in a different language.
One thing is still puzzling though:<p>Considering that RF is essentially monopolar until the detector adds a bias, the bias circuit's ability to reproduce an undistorted wave smells like rotten fish in Denmark.<p>Could you elaborate on this as I am not sure of what you mean?
What makes me suspicious about noise not being the problem is this:
If the sole purpose of this setup is merely to operate a relay driven solenoid, then once the receiver has initiated relay action , its job is done. This is before the solenoid is actuated or even before the relay contacts actually close as just the start of relay action is all that was neccessary. From this point on, all the RF noise in the world would make no difference as the "missile" has already been launched and there is no calling it back.
But,again I do not have the whole story on this circuit.

Post Reply

Who is online

Users browsing this forum: No registered users and 16 guests