EasyTouch/Solar Issue

[merlin309]

Hi all,
First time poster, long time reader!

(Long, TL/DR summary below).

I've got a issue and could use some guidance. We have an EasyTouch system with solar which is about 2 years old. Been working great.

About a month ago, we added the iChlor30 SWG. That's working great (after a couple of teething issues).

But about 2 weeks ago, the Electronics breaker popped late one afternoon. Discovered it about 1 1/2 days later and reset it and all was well. Popped again a couple days after that. Noticed it right away and tried to reset it then, but couldn't (wouldn't latch closed at all). My thinking at the time was that it was a very hot afternoon (about 105), and the box in the sits direct sunlight facing west, the corner with the breakers on the south side, so maybe it is just getting too hot (those are thermal circuit breakers). Later that evening, I reset it with no problems.

It worked fine for the next couple of days, but then I noticed the solar wasn't working! A little troubleshooting later (thanks to a forum post that suggested switching the sensors around), I discovered the solar temp sensor was dead (open fault). Ordered a couple new ones (and switched the solar/air sensor around to get the solar going again). The new sensor arrived and I installed it. All was fine again.

Until Tuesday when the breaker popped again. Reset it, and all was good until I realized yesterday the solar was dead again. Got out my trusty multimeter and discovered the new sensor is open now. Put in my spare sensor (good thing I got 2) and all is good now, but for who knows how long?

TL/DR: the solar temp sensor keeps open faulting (twice) and the Electronics breaker keeps popping

Finally to my questions:
- Could a bad (open) sensor cause the circuit breaker to pop?
- Could the breaker popping somehow destroy the sensor?
- Could there be something going on the board that causing both to happen? Maybe temperature related after all. (my slightly educated guess)
- Is the SWG just a red herring and not really involved (since it's the iChlor and does it's own thing, and just has two-way communication to the EasyTouch).
- Had anyone seen anything like this and might have any suggestions?

Thanks for any feedback and information you can supply!

Mark

 

[Jimrahbe]

Mark,

Let's see what Tom has to say.. He is on top of what power supply voltages go where.. Calling @ogdento

My gut says, when the electronics shut off, that just shuts down the SWCG.. I doubt the SWCG is bad.

Thanks,

Jim R.

 

[ogdento]

Hey guys,
I don't think that the swg or the breaker tripping would hurt the sensor - or at least I can't think of how that would happen. The swg "could" cause the breaker to trip if the comms were fried, but it doesn't seem like that's the case.

I don't have a board in front of me but if you look on the opposite side from the sensor connector, you'll see the traces from the two solar sensor pins go to some transistors and resistors which then go to the microcontroller... it's possible one of those components got damaged. A bad component could cause too much current to go through the sensor... which you can really think of as a low-power resistor that varies with temp (10k matches to about 70 degrees if i recall).

I would first inspect that area of the board to see if any parts connecting to the solar sensor pins are damaged/blackened, have poor or broken solder joints, or if any traces are bad. Then I would remove the solar sensor, connect a 10k resistor to it, and measure the current going through the resistor... I can't imagine it would be much more than a few mA, but if it is that could explain why a couple have blown. You could also measure the voltage across the pins for each sensor and compare them - I haven't done this so I'm not sure what they're supposed to be, but my guess is they should all be about the same.

Tom

 

[merlin309]

Hey guys,
I don't think that the swg or the breaker tripping would hurt the sensor - or at least I can't think of how that would happen. The swg "could" cause the breaker to trip if the comms were fried, but it doesn't seem like that's the case.

I don't have a board in front of me but if you look on the opposite side from the sensor connector, you'll see the traces from the two solar sensor pins go to some transistors and resistors which then go to the microcontroller... it's possible one of those components got damaged. A bad component could cause too much current to go through the sensor... which you can really think of as a low-power resistor that varies with temp (10k matches to about 70 degrees if i recall).

I would first inspect that area of the board to see if any parts connecting to the solar sensor pins are damaged/blackened, have poor or broken solder joints, or if any traces are bad. Then I would remove the solar sensor, connect a 10k resistor to it, and measure the current going through the resistor... I can't imagine it would be much more than a few mA, but if it is that could explain why a couple have blown. You could also measure the voltage across the pins for each sensor and compare them - I haven't done this so I'm not sure what they're supposed to be, but my guess is they should all be about the same.

Tom

Thanks for the suggestions Tom... It's worked fine for the last couple of days with the 2nd sensor (but it's also been slightly cooler out).

Good idea about checking the current with the resistor in place. I was wondering if somehow too much current was running through the solar temp sensor (it is a good 50ft away), that caused the breaker to blow, but not before frying the sensor a couple of times.

I've resisted unmounting the board, but it does sound like that might be then next step. I'll take a look (and pictures) Sunday morning when I think I have some free time before it gets 100+ outside.

Mark

 

[ogdento]

Hey you're welcome... it definitely sounds like you've got a board-level component problem rather than a setup/configuration problem. At this point most folks just swap it out for a new one but it sounds like you've got the skills to troubleshoot further, which is awesome.

When you said the sensor is 50' away I wondered if that could be playing a part... maybe a short happened at some point and damaged the input. Probably not a lightning issue since you're in Sac, not Texas or Florida

I haven't worked up a schematic of the sensor input but from a photo I can see each sensor input has a 10k and 3.3k resistor with a transistor. It looks like the 3.3 goes to the sensor, possibly to limit current. If the 10k connects to the transistor base, it's probably to turn the sensor on only when a reading is required... otherwise constant power can heat the sensor and skew the value. Anyway, I'd be interested to see what you find out.

 

[merlin309]

Finally an update: It worked fine for about 4 days and then the solar temp sensor died again! It was the day before we were to leave town for a week, so I didn't have time to dig into it more.

I took my last sensor I bought (glad I bought 3 at once) and just wired it directly to the panel instead of using the wires leading out to the solar panels. And it has worked ever since!

Admittedly it has cooled off a bunch (thanks to the fires leaving a smoky overcast) but I still have to wonder is something is injecting current into the wires leading up to the panels. I'm going to leave it like this for a few weeks to see if it continues to work. If so, I'll try replacing the wires leading out to the panel (they are in a conduit so it should be possible).

One thought just struck me: Could the wires be too close to the SWG transformer and somehow creating some sort of current into the temp sensor wires?

If it dies again, then it's time to dig into the board and see if anything is amiss.

Thanks again for the suggestions!

Mark

 

[ogdento]

Glad you have a sensor left and it's working when wired directly!

TLDR: you might possibly have a ghost voltage on your sensor wire... Switch a digital multimeter to ac and measure the voltage at the far end of the cable.

Interesting question about the SWG... to induce a meaningful current in a wire, the wire would have to be close to an alternating source (and ideally be a coil). How close is the sensor wire to the swg transformer? The swg transformer should be either inside the high-voltage section of your load center or inside it's own box, so the sensor wire shouldn't be all that close to it. I'd be concerned if your sensor wire was tightly coiled and within an inch or so of the swg xformer

But I wonder if you're seeing sort of a "ghost voltage" on your sensor wire... in your house you can often see "ghost voltages" on ac circuits, and they're caused by capacitive coupling - i.e. a cable carrying ac current is close to another cable on an unpowered circuit... but since they're long runs separated by an insulator, there's some small capacitance that can cause you to see ac volts (sometimes upwards of 100 volts!) on the cable that should be off. But as soon as you connect a load to it the voltage goes to zero.

So for your sensor wire... depending on how close the wire actually is to the swg transformer (which is of course AC), and given the length of your sensor wire (>50'), it's possible that you could see an ac voltage on that line which could actually interfere with the sensor. Switch a digital multimeter to ac and measure the voltage at the far end of the cable. It would be interesting to see what you find out.

 

[ogdento]

i just thought of something else... is the sensor wire in conduit (metal or plastic) and if so, is there anything else in the conduit with it?

 

[merlin309]

i just thought of something else... is the sensor wire in conduit (metal or plastic) and if so, is there anything else in the conduit with it?

Help if I actually said anything, hence the edit!

The solar sensor wire is in a PVC conduit, along with the wire for the remote transceiver.

The only thing it runs close to is the conduit containing the power for the IntelliFlo pump, and it's been like that since it was installed a couple years ago. It is nowhere near the SWG, so scratch that off the list. When I get more time, I'll take my meter out to look for stray AC voltage.

Thanks for the thoughts!