[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [TCML] Treating RF feedback



On 3/17/13 3:01 PM, w5als wrote:
HI Brandon Being a ham operator and tesla coil, I will tell you what
I would do.

Bear in mind that what's effective at 2-30 MHz (HF radio) and 100kHz or 100 MHz might be quite different.

If you can get at lease 4ft in the ground that will work
but do 2 ground rods at lease 4 ft apart. you have to be the length
of the ground rod apart from each other min. More the 4 ft. would be
better. Done not run one wire and go from one to another. Run 2
separate wire's to each one, with them coming to a common spot. I
mounted a wall outlet in my shop with nothing hook to it but the
ground wire's only. Than when I use my Tesla coil I plug my ground in
that outlet only. If you use 3 short ground rod's use 3 separate
wire's. But the deeper the ground rod the better. alton w5als

Are you saying that the bottom of the secondary coil is connected to the "greenwire" ground pin at that receptacle, which is then connected to your three rods?

Or do you have 3 rods driven surrounding your TC, you have wires from the rods to the TC location, and then hook that to the bottom of the secondary coil.

And have a dedicated outlet for the TC power.

I'm not sure there's any good physics behind this suggestion. Not that it doesn't work, but I suspect that if it does work, it's because of factors other than the ground rods, per se.

WIthout knowing more about how your house is wired and where your TC is relative to those ground rods (and more important the wires going to them) it's hard to say whether this is an effective strategy. Or, more to the point, measuring the RF current in those ground wires.


For EMI/EMC and tesla coils, there's two factors to bear in mind..

there's the 100-300 kHz frequency from the coil itself.. but that doesn't propagate very well, and the "antenna" (e.g. the coil) is a very tiny fraction of a wavelength. So all the 100kHz fields tend to be close to the coil, and you don't have to get very far away before they are "small". (if you're 10 coil heights away.. e.g. 30 feet from a 3 foot high coil) I'll bet they're almost negligible.


For EMI, I think a bigger problem is the transients from the actual sparks, and those radiated from the wires going to the spark gap. The latter typically radiates at VHF sorts of frequencies (think of the wires going to the spark gap as a dipole radiator, with the spark gap at the center). Several coilers have measured substantial radiated fields from this source, and since VHF receivers are pretty sensitive, it's easy to jam them. Whether it could "false trigger" is another story..

Then there's the wide band transients from when the spark "fires". You have a very high di/dt, so you get a fairly big magnetic field radiated from a loop that's fairly large (ground, secondary coil, topload, spark, wherever the spark hits, conductor from spark end to ground). This is why you *don't* want to run a coil indoors with a 50 foot wire to the ground outside. You're making the transient transmitter more effective by making the antenna bigger.

some analysis from a few years ago is here
http://home.earthlink.net/~jimlux/hv/tcemi.htm

The problem with things like garage door openers is that they have unshielded wires going to the photosensors and limit switches, so you could easily get a pretty big receiver loop (with "capacitive coupling" being part of the loop). In my garage case, the opener is about 10 ft from the door, and there's a wire that goes down to each side of the door where the photosensors are, and then a ground through the plug in the ceiling for the opener. So there's this huge 8 footx15foot rectangular loop to pick up signals (the wire from the photosensor runs along side the metal track of the door, and has significant capacitance to "earth ground" since that track is connected at the slab with an anchor bolt. It's kind of like it was designed for poor emi performance.

Sprinkler systems are worse. They are very cost sensitive, so the output is typically a bunch of triacs connected right to the microprocessor switching the 24VAC to the valves, and there's a common ground wire for all valves. It's easy to get a big loop between "valve hot" and "common ground" depending on how it's wired. And if there's much induced voltage on the control wires, then the triac will switch on (or, simply that the triac is a 50V part, and it just avalanches when there's a spike). Your TC is emitting pulses synchronized with the power line, so it's optimally timed to trigger the triacs with an induced gate pulse.

Since people want to save money, the valve wiring is often sort of a point to point daisy chain, rather than a single twisted pair to each valve (the latter would be ideal for EMI), and the wires are buried and often have some conductitivty to the surrounding soil. I had terrible trouble with EMI on my first sprinkler timer.




On 3/17/2013 3:45 PM, Brandon Hendershot wrote:
Hi All,

With the renovation of my coil nearly complete, I'll need to
address the old issue of electrical appliances going haywire
whenever I run my coil. At the time I didn't have a proper RF
ground, only a counterpoise. On separate occasions, I'd have the
jacuzzi tub turning on or the sprinkler system switching the
sprinkler heads like pinball paddles. My hunch was RF interference
which the list seemed to support. I'm currently attempting to
install a 10' ground rod but that's proving difficult with my dry
rocky soil. Since my deadline is fast approaching I may need to
rely on the counterpoise once again and I want to make sure I don't
go fizzling all the fancy electronics we've acquired in the past
years since I retired my coil.

On a separate topic, any bright ideas on getting a 3/4" copper pipe
 backed with city water pressure through a boulder?

Thanks, Brandon H. _______________________________________________
Tesla mailing list Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla

_______________________________________________ Tesla mailing list
Tesla@xxxxxxxxxx http://www.pupman.com/mailman/listinfo/tesla


_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla