Condx changes values when lights turn on

[toddrtrex]

Setting up a new 210, got everything running on Monday. Tuesday I calibrated the condx probe and the reading matched my refrac.

Over the following days I noticed it was changing from around 40, to around 35, and this wasn't over a period of time, a quick spike. After looking at the graphs, I noticed the spike happens right when the lights turn on. Just watched it on the display, and the spike went from 40.7 to 34.9.

The CondX probe is plugged into a PM2. I have moved the probe's wire to the front of the sump to the PM2, so that is doesn't cross paths with the light's cord (( ATI, 8*80 )). Just before the 6 lamps turned on I moved the aquabus cable from the EB8 bar that the lights were plugged into, I moved it in the the LDM, on the opposite side of the sump, in the front, not crosses any power cords. No change.

With looking at the graph for the last couple of days, there is a minor spike/drop off when the 2 lamps turn on/off, but the major spike/drop off occurs with the 6. The CondX values stay pretty much the same once the lights are on/off.

Any help would be great, thanks.

Graph



Extra info,
APEX gold
3 EB8s
WMX
ALD
PM2
LSM
2 20 Amp circuits
The lights, light's fan, main pump, one MP40, ATO power supply, and canopy clip on fan are on one EB8, into one of the 20 Amp circuits, nothing else on that circuit.

[aquamanic]

Try moving the power cords for the lights to a wall outlet. Get an extension cord if you have to. See if the problem goes away when you do that (get's the lights off the Apex system all together).

[toddrtrex]

Try moving the power cords for the lights to a wall outlet. Get an extension cord if you have to. See if the problem goes away when you do that (get's the lights off the Apex system all together).

Will be able to unplug them, but using them on a wall outlet long term won't really be able to work -- from both a control aspect and aesthetics.

What is odd, is that when the lights are off, is when the CondX jumps up to 40, which is about 5 points off from my refrac which I just calibrated, using the pinpoint solution and the same overhead lights that I test with.

Just did it, moved the lights to a completely different circuit. Right when I unplugged them, the CondX jumped up, plugging them back in made it jump back down --- again, completely different 20 amp circuit.

If it matters, the PM2 is on the same side of the stand as the light's cord, but they are NOT crossing. Right now the ATI cords are outside of the stand, and the PM2 is inside it.

While not ideal, I could try to move the PM2 to the opposite side of the stand, just would make a long travel for the probe cords. (( my sump has a built in probe holder ))

[alb]

Given your description, I'd be suspicious that the lights aren't actually the culprit and instead you've got some other piece of equipment leaking stray voltage into the tank. If it were me, I'd turn everything off, then one by one turn each piece of equipment on, pausing after each one for a couple of minutes. If you see a spike on the condx probe, that's your bad actor.

Not uncommon to see a cracked housing on a pump or a heater cause such a thing, and stray voltage is...weird, thus the methodology described above.

[toddrtrex]

Thanks, but I am pretty sure it is the lights, just from watching the CondX on the display change right when the lights turn on/off --- the spikes on the graph match up with the programmed time for the lights to go on/off.

Since the tank has only water/rocks in it, I can (( somewhat )) easily unplug everything and see what happens with the lights go on/off.

Question --- would it "count" if I just turned the outlets off from Fusion, or actually unplug them all? (( not a huge issue, just would be a lot easier to just turn off with fusion.

Thanks again.

[alb]

Just turning them off would be fine. TBH, if it is a stray voltage situation, the less you touch plugs, the better.

[toddrtrex]

I ended up unplugging everything (( wanted to make sure, plus I wanted to switch out one of the heaters )), and it is the lights --- biggest drop/spike from when the 6 lamps (( out of 8 )) turn on/off.

Tell ya, sure glad that I put labels on each of the plugs, made it a lot easier to plug everything back in.

Now, I guess I move the PM2 to the other side of the stand? I guess one good thing, the readings seem to be correct when the lights are on (( not that that makes sense to me )), which is when I would be concerned about the levels.

[aquamanic]

I would think it's less EMI and more about some sort of ground issue between your lights and your controller. That's why I wanted you to plug the lights in someplace other than your Apex system, just to see.

You can move your PM2 and see what happens.

[toddrtrex]

I would think it's less EMI and more about some sort of ground issue between your lights and your controller. That's why I wanted you to plug the lights in someplace other than your Apex system, just to see.

You can move your PM2 and see what happens.

Thanks.

Last night I plugged the lights into a separate 20 amp circuit, (( and with nothing plugged into the APEX at all )) and the spike/drop still happened.

Will try to move the PM2 to the other side, but starting to wonder if this is just going to be the way it is going to act.

What makes me wonder the most is why the reading is "wrong" when the lights are off.

[aquamanic]

You say it happens worse when you get 6 lights turned ON but can you isolate it to a fixture? Like when fixture #6 turns ON whether that's the first fixture or the last, the problem gets worse?

[RussM]

You say it happens worse when you get 6 lights turned ON but can you isolate it to a fixture? Like when fixture #6 turns ON whether that's the first fixture or the last, the problem gets worse?

If the OP has the ATI 8xT5HO lights I'm thinking of, it is one fixture. 6 lamps are on one power cord and two lamps are on another, giving three overall levels of lighting.

[toddrtrex]

If the OP has the ATI 8xT5HO lights I'm thinking of, it is one fixture. 6 lamps are on one power cord and two lamps are on another, giving three overall levels of lighting.

Exactly.

There is a much smaller spike/drop off when just 2 of the 8 turn on/off, the major spike/drop off happens when the remaining 6 turn on/off. Can see the the when the 2 lamps turn on/off in the graph above, it is much smaller, but still there.

[aquamanic]

I see. Are there multiple ballasts inside it or just one? Are the lamps new?

[toddrtrex]

I see. Are there multiple ballasts inside it or just one? Are the lamps new?

Brand new fixture and lamps. Been running (( excluding a few times to see how the tank is looking )) for only 2 weeks now.

To the best of my knowledge, there are 2 ballasts, but I am not 100% sure. The 2 power cords and fan cord all "meet" in a box (( about the size of a Vortech power supply )), with one cord then running
to the fixture.

Have some errands to run today, but will try to move the PM2 later on, to see if that changes anything.

ETA, this is the fixture -- 8*80.

http://atinorthamerica.com/sunpower.php

[zombie]

When you tried it on a seperate circuit, was the outlet physically close to the original outlet? If so, try it in another outlet in a different room and let us know what happens. If it continues in a different room, then there is significant insulation breakdown in one of the components of the ballast that is not safe to operate, and could potentially start a fire. I would return or repair it under warranty if that is the case. If the behavior stops in a different room, there may still be an issue with capacitive leakage, but the problem could lie in your outlet grounding. If that is the case, I would have an electrician (or someone very familiar with household wiring) test and inspect your outlet grounds.

You might be an engineer if...You have no life and can PROVE it mathematically.

[toddrtrex]

When you tried it on a seperate circuit, was the outlet physically close to the original outlet? If so, try it in another outlet in a different room and let us know what happens. If it continues in a different room, then there is significant insulation breakdown in one of the components of the ballast that is not safe to operate, and could potentially start a fire. I would return or repair it under warranty if that is the case. If the behavior stops in a different room, there may still be an issue with capacitive leakage, but the problem could lie in your outlet grounding. If that is the case, I would have an electrician (or someone very familiar with household wiring) test and inspect your outlet grounds.

You might be an engineer if...You have no life and can PROVE it mathematically.

The outlet was within 8 feet of the ones that are powering the tank --- but is a separate circuit. I had my electrician (( career, not just a weekend one )) install the 2 separate circuits for this tank. Have had do other work in the house, and watched him every time --- more for curiosity --- and feel great about his work. Someone who is more OCD than I am.

I moved the PM2 to the other part of the stand, and ran the probe wires to the front of the sump and over --- no other wires (( probe or power )) near by. The 6 lamps are not on it, but the 2 are. It appears that that the small drop, that happens with the 2 come on, has reversed. Will see what happens around 1230 when the 6 come on.

But, what is still odd, the values only appeared correct when the lights were on. However, will worry about that once (( hopefully )) the spikes/drops stop.

[alb]

If they appear correct when the lights are on, then that would tend to indicate that you calibrated the probe with the lights on, and thus it's essentially calibrated to the level of EMI that the lights put out.

Given that the drop reverses when it's moved (presumably) to the other side of the EMI field, I'd suspect that to be an unusually large amount of EMI, and so I'd recommend paying close attention to what zombie states about a possible ballast issue above. There's always some amount of field present from any electrical device, but if it's affecting a probe submerged in water at a significant distance from the lights, then personally, I'd be quite concerned.

[toddrtrex]

If they appear correct when the lights are on, then that would tend to indicate that you calibrated the probe with the lights on, and thus it's essentially calibrated to the level of EMI that the lights put out.

Right, that makes so much sense. Clear as day now, just couldn't wrap my brain around that part --- thanks.

Given that the drop reverses when it's moved (presumably) to the other side of the EMI field, I'd suspect that to be an unusually large amount of EMI, and so I'd recommend paying close attention to what zombie states about a possible ballast issue above. There's always some amount of field present from any electrical device, but if it's affecting a probe submerged in water at a significant distance from the lights, then personally, I'd be quite concerned.

So, I don't know much about EMI fields, so not sure if that is what the cause is, after doing some experiments.

I took the CondX probe and place it in a glass (( filled with tank water )), and moved about 10 feet away from the tank. The readings were back to the "lights off" number. I slowly moved closer to the tank -- about 3 feet at a time, and waited for a minute to see what the readings were at that spot. Each time, they stayed at the "Lights off" number. I finally placed the glass on top of the probe holder (( to get as close as possible to the everyday spot )), and the readings held.
Within 10 seconds of placing the probe in the water, the reading dropped.
I unplugged the 6 lamp cord and the readings went back up.
Plugged the 6 lamp cord into an extension cord, and plugged that into an outlet 20 feet away -- the readings dropped.

So, there is something about the lights that is causing the change, but I would think if it was EMI, it would have still happened when the probe was in the glass. If that part is correct, then it would seem that there is a ground issue with the lights, but not sure how much (( if at all )) voltage is leaking. I will need to pick up a voltmeter, but can say that from having my hands in the tank numerous times, I have never felt anything.

Thoughts about this new info?

Thanks

[zombie]

From what you are seeing, you have narrowed it down to two possibilities, capacitive leakage through the ground connection of the lights, or magnetic induction from the ballast to the water. You can narrow it down to one or the other by taking the lights off the tank and placing them on the ground at least 5 feet from any water. Then turn them on and see if the numbers spike. If they still spike, then there is insulation breakdown in the lights and it needs to be serviced or replaced. If it does not spike when the lights are moved away from the water, then there is an inherent magnetic field from the ballast that will be present regardless of repair or replacement. If that is the case, the only thing that would fix it is to return the fixture and get a different brand/model.

You might be an engineer if...You have no life and can PROVE it mathematically.

[toddrtrex]

From what you are seeing, you have narrowed it down to two possibilities, capacitive leakage through the ground connection of the lights, or magnetic induction from the ballast to the water. You can narrow it down to one or the other by taking the lights off the tank and placing them on the ground at least 5 feet from any water. Then turn them on and see if the numbers spike. If they still spike, then there is insulation breakdown in the lights and it needs to be serviced or replaced. If it does not spike when the lights are moved away from the water, then there is an inherent magnetic field from the ballast that will be present regardless of repair or replacement. If that is the case, the only thing that would fix it is to return the fixture and get a different brand/model.

You might be an engineer if...You have no life and can PROVE it mathematically.

I guess the good news is that when I plugged them into the EB8, with the lights off the tank, there was no drop/spike. Of course it returned when I put them back over the tank.

So, it would seem (( and correct me if I am wrong )), that this will just be an annoyance, and really won't harm anything. One good thing, (( if there can be )) is that drop/spike is somewhat predictable. Though, I do believe that I will recalibrate the probe and have the lights off when I do that. That way I can get a true reading, and can easily turn the lights off when I want that true readings.

While that was a bit of a pain in the butt, the help was great. So, thanks guys.