139.1 degrees

[00Warpig00]

My Overheat flag has been tripping semi frequently lately, just enough to be annoying, yet the portal never shows any strange values.

I decided to set up some code that keeps a HighTemp value and displays it on my screen.

every time through my loop I read the temp probe (1) set it to a variable and if the value is higher than the last time through the loop update the variable to the new High Temp. Trying to find out what is going on. So it turns out about an hour ago I get a high temp flag for the first time in a week or so. I go over to the controller to look at the screen and although my temp is reading normal by the time I get to the controller my High temp display on the screen is showing me 139.1 degrees. I have not set up any other monitoring for probe 2 or 3 yet.

I'm sure I pulled the wires out of the temp probe molex plug trying to unplug the probe, as I'm pretty sure I have yanked them from all three of my probes at one time or another. I have wiggled my connections and wires where they enter the molex plug and cannot recreate this behavior. Seems to be at random. I'm not sure I am loosing connection at this point.

Any Ideas?

[rimai]

Does it do that if you remove probes 2 and 3 too?

[00Warpig00]

I will disconnect the other two to check and report back as soon as I have any clue. It could take anywhere from a few hours to over a week for definitive info.

Nick

[00Warpig00]

This weekend I unplugged my other two temp probes. Since I have removed the other two probes from the system the overheat temp has spiked to 137.4 twice in a period of only a couple of hours setting an overheat flag.

Nick

[rimai]

Has it done that in the past or just started now?
How long is your extension, if you have one?

[00Warpig00]

Probably started with an occasional Overheat trip about the last week of January. Would trip Hi Temp but no hi temp recorded on the portal and I never saw a hi temp on the android app or the controller. Would happen about once a week at first. by second week of feb it was tripping twice a week still with no clue as to why. So I added code to update the display on the controller in each loop with the High temp. Only updated the display if the new temp was higher than the highest temp recorded. Thats how I caught the 139 degree reading. More of the same with all three probes installed up until today when I unplugged 2 and 3 and reset the controller. 2/3 now read 0 and now #1 trips about every hour since I unplugged the others.

Nick

[rimai]

Are you using any extension at all?

[00Warpig00]

no extensions yet on any of my temp probes.

Nick

[rimai]

Humm...
Can you try only one probe, but not the one that is currently connected?
Use one of the other 2.

[00Warpig00]

Humm...
Can you try only one probe, but not the one that is currently connected?
Use one of the other 2.

I knew you were going to suggest this next. So before I went to work today I unplugged the #1 probe that's acting up. I plugged in another probe into the same connector that I unplugged the previous probe from. Rebooted the controller and let it run all day without an overheat. It was the only probe plugged in. I also put the probe end in my tank since the one that was acting up is my tank water temp probe.

Looks like I may have a bad probe.

Nick

[rimai]

Cool. Glad we found out.

[DrewPalmer04]

Your issue sounds like the name of a boy band hahahaha. Glad it's worked out.

[00Warpig00]

i spoke too soon. Leaving things the same probe two only plugged into controller into the port probe 1 was plugged into. No problems last night but came home from work today to a overheat flag and a high temp reading 137.4 starting to look like I spoke too soon yesterday.

Nick

[lnevo]

I would kill the overheat flag for now...set it to 200 so you could start tracking the temps and see where they go..does it stay at that temp? Is it just spikes?why don't you set an alert for overheat, and check it remotely...you can clear the overheat flag over the web...

[DrewPalmer04]

Does any relay turn on when you get the spike? Of course you'll have to track the spike to know but it could be related to a large amp draw, etc

[00Warpig00]

I have been considering killing the overheat flag for now. It's about 8F degrees outside here right now and my overheat flag obviously kills my heater in my sump. My tank normally runs a bit high in temp 79-82 degrees but with it being so cold out and having the heater tripped off due to the flag my tank was down to 75 degrees when I got home. Not a big deal I really don't start to worry about the water being too cold until it hits about 70.

As for the temp. It seems to be an instant spike. I have NEVER had an abnormal temp reported to the portal. I have also heard the skimmer and my powerheads shut off while I was sitting here. Got up and rushed to the tank and the temp was normal. So a few weeks ago I put some code in to monitor and display the High and low temps. Each time through my normal program loop the overheat probe temp is compared to a "High\Low" temp variable and if it is higher/or lower it updates my high\low temp display on my screen. This is the ONLY way I know that my temp was 139.1 degrees. I am not sure how long it stays high maybe I need to add some code to count the loops while it is high. Was hoping to resolve the issue without too much digging.

Nick

[00Warpig00]

I guess it could be something like a higher draw or maybe leakage into the tank, but nothing has changed in my system for a while. Just all the sudden I started getting over temp flags. I try to insure one thing is stable and working before I move to my next project. Been trying to get everything tuned up for the past several months and never seem to make much progress. It's always one thing or another. I'd like to someday be able to say my RA+ is "Officially Installed"

Nick

[DrewPalmer04]

Just have it track the high spike and the time when it occurs? This will allow you to see if it's relay related or whatever you can match up in your code related to time.

[DrewPalmer04]

I guess it could be something like a higher draw or maybe leakage into the tank, but nothing has changed in my system for a while. Just all the sudden I started getting over temp flags. I try to insure one thing is stable and working before I move to my next project. Been trying to get everything tuned up for the past several months and never seem to make much progress. It's always one thing or another. I'd like to someday be able to say my RA+ is "Officially Installed"

Nick

Just not sure if it's the RA fault or equipment aging/failure fault. Worth looking into though.

[00Warpig00]

I'm not sure I am concerned with the exact time it occurs. It would be different if the controller logged constantly to somewhere (a file etc) but with the delay of pushing data to the portal and/or my android app pulling data it's comparable to trying to snap a picture of a lightning bolt. Just seems unless it all occurs at the same instant and that same instant is pushed to the portal or pulled by my droid that it would be impossible to see. I think I may be more interested in the how long this condition exists more so than the time it happens.

Ultimately I just want everything to work as I planned it to. As I said I have had it running several months and I still haven't gotten to the point that I have everything set up yet. If it's not a flaky float switch it's the wifi mod loosing connection or the IO module causing lockups or a brand new symptom of high temp flags. Not sure what will be next. I'm starting to loose my will to want to fight with it constantly. I just want it to be DONE. lol

Not saying that all of my issues are the RA's fault either. I know for a fact that my float switch issue has nothing to do with the controller it self. I have put a meter on the switch and the controller is reporting it's status accurately. Why the float switch reports active when it clearly should not is another matter I still need to figure out. I think i may have some huge magnetic field that only screw with one of my float switch locations lol and I have tried two switches in the same spot. The both do the same thing but i have no clue as to why...

Other things like the controller locking in a reboot loop while my IO module is plugged in I think may be a RA issue. The temp high flag the jury is still out. The Wifi mod... Who knows! Read the forum MANY people are having issues with wifi. Glad to report my Ethernet setup has been PERFECT thus far. So happy I went to wired Ethernet

Nick

[DrewPalmer04]

That's quite the load you have going on. I hate frustration but sure do love when I figure it out

[00Warpig00]

So I think I *may* have resolved my funky float switch issue... After my RANT here last night I spent a good two hours really digging into my odd float switch behavior on that singular float switch. I don't want to jinx myself but after Roberto already sent me a replacement float switch that behaved the same way as the first and confirmation with my DMM I knew it had to be something other then the float switch, but I had no clue what as there was no visible evidence. I think I figured it out last night. A few more days monitoring should reassure me I got it. Turns out the issue may have been being caused due to the float switch being too close to a corner in my sump. What I think was happening that I could not see is that the return water flowing into my sump was hitting the sump wall\corner and then flowing down causing a down current near the sump walls\corners causing the float on the switch to catch the current and ride down the shaft to activate the reed switch. I could not see the float doing this because I have my float switches all encased in white PVC (think pill capsules) with slats cut into both ends to allow water through but not much else. So it seems I had downward current forming inside the "pill capsule" containing the switch. I spent a bunch of time moving the switch/capsule around in my sump and found a spot without downward current flow. So far so good! my switch has not unexplainable shown active when it shouldn't yet.

BTW does anyone know of a method to log the status of my IO module ports? Something like the relay activity on the portal is what I am thinking or a way to push some sort of info to me in the event of an IO port condition change. This would help me verify things are staying in the correct condition without having to be at home to look at the controller.

Also while I was working in the tank I decided I was finally going to get my PH probe working correctly. Why not introduce one more variable... lol So that seems to be working now too.

If I can only get my controller to never lock up with a flashing screen (IO Module?) I might make some progress into uncharted waters some day...

Also had a situation today where my High\Low temp monitoring code registered a 130+ degree temp from the probe, but it appeared that the spike was fast enough that the overheat flag code DID NOT register it. So a 130+ temp and NO OVERHEAT. Im not sure how often the overheat code is looked at but it seems that this spike is rather quick. My code in my loop catches it but at least this time the overheat flag code didn't

Nick

[rimai]

Does this chart help any?
http://forum.reefangel.com/status/chart ... &filter=io

[00Warpig00]

Does this chart help any?
http://forum.reefangel.com/status/chart ... &filter=io

Hmm,

I'm not sure to be honest. Seems that chart shows variance of 61/62/63 Im not really sure what those numbers on that chart are telling me.

Nick

[rimai]

The numbers in there represent the 6 bits of the I/O module.
61 decimal = 111101 - Indicates Channel 1 was activated
62 decimal = 111110 - Indicates Channel 0 was activated
63 decimal = 111111 - Indicates no Channel was activated

[00Warpig00]

The numbers in there represent the 6 bits of the I/O module.
61 decimal = 111101 - Indicates Channel 1 was activated
62 decimal = 111110 - Indicates Channel 0 was activated
63 decimal = 111111 - Indicates no Channel was activated

Ahh.. Duh.. It never even crossed my mind to do a decimal to binary conversion on the data

Nick

[00Warpig00]

Does this chart help any?
http://forum.reefangel.com/status/chart ... &filter=io

hmm

seems at 12:40 ish this morning my IO module detected two float switches toggling active then inactive several times over the next few hours ending after 2:00 am.

Interesting if I'm reading it correctly the 54's on the chart indicate that both my "Sump Empty" & "SW Topoff Full" switches were flipping back and forth. this is interesting because the topoff float switch is in empty air in my ATO Brute can and the sump empty switch is in the bottom of my sump. It even appears that during one report to the portal only my ATO full switch was active. Not sure what to make of this... lol

Nick

[rimai]

Did you do anything around that time?
How did it go back to being normal again, which I assume is 63, right?

[00Warpig00]

Did you do anything around that time?
How did it go back to being normal again, which I assume is 63, right?

Not sure how any of it happened... lol I could see the sump empty switch doing strange stuff since I just put it in it's new sump location a few days ago. I have not touched my topoff float in forever. the topoff brute is only half full so the entire switch assembly was in open (dry) air at this time. Unless a Chupacabra http://en.wikipedia.org/wiki/Chupacabra broke into my place in the middle of the night and was playing around with my tank. lol I was sound asleep in bed ...and yeah 63 is the normal state (all switches inactive) Seems odd both switches were active at the same time and both seemed to cycle synchronously for all cycles except one. weird... I never noticed any behavior like this before, but I never really looked at my IO history either. I wonder if some false data is somehow being transmitted on the I2C bus. Maybe related to the strangeness with the IO module locking up. Not sure... just thinking out loud. the RA has not locked up yet since this occurred that I am aware of.

Nick

[00Warpig00]

Does this chart help any?
http://forum.reefangel.com/status/chart ... &filter=io

WOW! the plot with the IO Module thickens.

at 07:38am this morning my controller reported that my IO Mod data went from a byte 63 to a byte 18.
a byte 18 from my IO mod (010010) means that 4 of my float switches went active (according to the IO module) since the previous push to the portal 5 minutes earlier. "Sump Empty" "SW Topoff Empty" "SW Topoff Full" "RO Topoff Full" now for the really interesting part... They stayed in this state for the next 12 data pushes to the portal. Then... POW! locked up controller (repeated flashing blank white screen & flashing green\red leds on controller) YES controller locked up exactly an hour after this state was established at 08:38 was my LAST PUSH to the portal. So I get home to a locked up controller and pull the usb cable from the IO Mod and the controller reboots and is again running normally.

Im sure you are already on to this Roberto but it appears as if the IO Mod is flaking out and locking up and when the controller tries to read data from it over I2C and cant it's flaking the controller out too. once I remove the data connection to the io mod removing power and resetting the IO mod and removing the lack of response to the RA itself the RA seems to jolly well pick up with it's chores and gets back to work.

Nick