EMI - Halides - Controllers

[oreo54]

using two halides the interference was barely noticeable;

Maybe it is just that one ballast.. Suggest replacing it..
Some of my links list known "better" ballasts though what one can and cannot get (and works) is ???

You can try different "pairs" of ballasts and go from there.


The way I understood it is one needs to wire every end to an earth ground.
That would be 4 wires per ballast...

Honestly this leads me to my old computer peripheral mantra.. if it fails... switch parts. You could fight with this for a long time..

https://www.reef2reef.com/threads/metal-halide-and-apex.713697/

I had multiple issues early on with the pH probe but that was due to a metal halide ballast causing interference. Switched to a new ballast and the problem went away.

https://www.reef2reef.com/threads/h...ies-question-and-answer-thread.763552/page-37

 

[Syntax1235]

Thanks Oreo54, I have read those threads before.

I linked all the sleeves together with wire by daisy chaining then together, and then attached it to a grounding probe plugged into a power strip on a different outle - the interference got much worse.

I then took a two foot piece of copper tubing and pounded it into the ground almost all the way. I then attached the ground to the pipe - it didn’t help at all.

I think I have exhausted all my options. I’m out of ideas. I guess I will have to program the ATO to be off when the halides are on.

I tried turning on each ballast separately and they each caused interference. I guess my previous observation about two ballasts not causing interference was incorrect.

 

[oreo54]

Soo seems to be the ballast type.
Anyways had my own doubts about the braiding.*

Most that deal with it use some sort or mix of toroidal chokes of various material and types.

Switching power supplies can emit the same.

 

[A. grandis]

Won't the problem go away if you put the ballasts far from the controller? Maybe you could try that. I know the wire also could contribute to the EMI, but you already helped with the sleeves. I would try that to figure out if that could help somehow. Perhaps that's the easy solution for the problem?

Another solution is simply to get a EMF protection box. They are sold normally to place routers inside, in wired houses. I have that in my house to protect us from EMF. Yes, all the computers in my house are wired, no Wi-Fi! We sleep much better now!!!
Here is mine:

 

[JDCagley]

Control wiring is typically ran in shielded cables. One should also try to maintain 6" separation from line/control wiring and if they must cross, cross perpendicular. I do not know how Apex/others sensor wires are, but most all control wiring I use in the industry are shielded cable, grounded on ONE end only. If you ground both ends you can create a difference of potential across the shield, thus creating more interference as well.

 

[A. grandis]

Yup, we have Cat7 shielded cables in the house with protected heads for our computers. I hope his wire sleeves for protection are efficient and the box will do the trick.

 

[Syntax1235]

Thanks for the suggestions. The ballasts are around 6 feet from the controller and the sensors, maybe isolating them more could help.

is it possible for the emi to radiate from the ballast itself? I guess I could try some kind of box.

I need to take my mind off this for a while and enjoy the tank.

 

[JDCagley]

Are they electronic or magnetic? Interference can and will come from lots of AC devices. Hence why the control wiring should be shielded because controls are always around interference! I can't believe Apex sensors are not shielded? What type of sensor are you having issues with? Is it a 2-wire sensor? Could the wiring be modified/spliced into shielded cable? Or is it proprietary ends?

 

[A. grandis]

Maybe you should get a small of those boxes for EMF protection for the controller. It will be cheaper and the controller will be protected. easier to deal with. That way you won't need to worry about the sleeves doing the EMI. We bought our box on Amazon a while ago.

 

[Syntax1235]

Are they electronic or magnetic? Interference can and will come from lots of AC devices. Hence why the control wiring should be shielded because controls are always around interference! I can't believe Apex sensors are not shielded? What type of sensor are you having issues with? Is it a 2-wire sensor? Could the wiring be modified/spliced into shielded cable? Or is it proprietary ends?

Ballasts are electronic. I’m having issues with the Liquid Level Sensor and an optical eye.

 

[JDCagley]

That makes more sense as magnetic is typically a little easier sometimes to get rid of. Electronic ballasts are (typically cheaper ones) constructed in high frequency ranges which cause major disturbances.
Try to get the cords from ballast to light as short as possible. Make sure they are not coiled and keep the lamp cord as far away from the main line cord as possible to keep the frequncies from 'jumping' into the main lines.
You mentioned earlier about the copper pipe into the ground - this is a bad idea and against electrical code. You can set up a system of 'isolated ground' but it is more involved and would require some rewiring. The option of a 2nd ground rod will not really help in your case and cause a severe electrical issue if not done properly. Multiple ground rods are REQUIRED to be bonded together - often called a ring.
Isolated ground, though, is a shielded ground the whole distance from the ground source - i.e. panel. It should not touch any other metal, box, equipment etc. These are typically used in hosiptals and labs to ensure that the ground is 100% 'to ground'.

 

[JDCagley]

I would ensure the ground you have now is solid as well. Take a reading from black to ground and black to white and see if you have the exact same reading. (+/- a little would be acceptable) but in theory, you should have exact same reading from line to N and G.

 

[Syntax1235]

The wiring is a bit of a mess…. The length of the wires are fixed, they are fairly long. Here’s some pictures for reference.

 

[A. grandis]

Try not touch the wires, keeping them parallel apart as much as you can.
I wouldn't have the sensor, but it's very interesting to read what others are posting in this thread. I hope you find the solution!