Results 1 to 6 of 6

Thread: Programming philosophy

  1. #1
    Frequent Visitor
    Join Date
    Apr 2017
    Location
    Dubai
    Posts
    71

    Programming philosophy

    Four months into my experience with Apex 2016 and the reefkeeping hobby and spending $$$ on extras such as FMK (with 3 flowmeters), FMM (with three optical sensors and leak detector), 1LINK (powering WAV and PMUP), DOS+DDR and BoB (with 3 on/off switches and 3 floatswitches). I was about to buy PMK.

    I have had a lot of fun learning how to program all sorts of automation based on water levels, switch positions, temp, pH, time and flow rates. It all worked beautifully.

    Until this morning I felt I had my rig dialed in with stable temp, Alk, Ca, Mg, pH, extremely low Phosphates, Ammonia and just starting to dial in trace elements. Livestock looking great! Even SPS is happy. As a newbie my learning curve is steep but it's been rewarding with only a few bumps.

    This morning Apex took a dump (documented elsewhere under "Apex 2016").

    This event has profoundly shaken my confidence in automating and controlling parameters with the Apex 2016, and especially in the use of flow monitoring for flow control. Switching the 24V ACC ports is also startlingly unreliable, especially the Neptune Solenoid Valve (these two matters are under an open ticket with Neptune).

    My philosophy now will be to strip away as much primary control from the Apex and use stand alone and independent controllers to monitor and maintain temp, water level and flow parameters. Lighting was already independent and for sure will stay that way (AI Hydra 26HDs are not Apex compatible).

    Apex is just not reliable enough to stand as the primary or sole means of control. This is an appalling insight to have to learn given the $ and considerable time I have invested! I wanted to have nothing but success with Apex.

    With today's event the Apex gave away its "right" to temp control, return pump flow control and power head control. For now I will leave refugeum light, skimmer and dosing under Apex control but will implement alternatives over the coming weeks.

    How do others use Apex? For me it's no more a trustworthy reefkeeping controller. Monitor, maybe. We will see.



    Sent from my iPhone using Tapatalk

  2. #2
    Frequent Visitor
    Join Date
    Jul 2016
    Location
    US, Central Time
    Posts
    106
    I view it like I view most products. It's a god send most of the time, but it's not perfect. I watch it daily and double check it on a regular basis. What they are doing is exceptionally complex, and they've done amazing things. Hardware and software are so hard to get right. So like so many things in life, I trust but verify. I hope whatever happened to you is recoverable and not a tremendous loss.

  3. #3
    Frequent Visitor
    Join Date
    Apr 2017
    Location
    Dubai
    Posts
    71
    No loss. I am planning a long trip away for a couple of weeks and had automated control to keep parameters stable. Will simplify and replicate with redundancy and test for a few weeks.


    Sent from my iPhone using Tapatalk

  4. #4
    Frequent Visitor
    Join Date
    Jul 2016
    Location
    US, Central Time
    Posts
    106
    Quote Originally Posted by Charles0dxb View Post
    No loss. I am planning a long trip away for a couple of weeks and had automated control to keep parameters stable. Will simplify and replicate with redundancy and test for a few weeks.
    Glad nothing was lost! I always seem to learn something new I should have done when I have one of these failures. I'm hoping it made some piece of programming pop out or pointed at a weak link.

    Cheers

  5. #5
    Frequent Contributor zombie's Avatar
    Join Date
    Dec 2013
    Location
    Denver, CO
    Posts
    13,176
    My philosophy with apex is the same philosophy that I use in real life protection engineering protecting multimillion dollar pieces of electrical equipment (tlines, transformers, generators, etc.): Anything can fail so you always have a backup if it's important.

    The "important" things in a fish tank are heat, flow, and ATO.

    For heat, apex controls the heaters. If for some reason the apex fails, the heaters internal thermostat is set 2 degrees higher to prevent overheating the tank. On larger tanks, I run a third heater independent of the apex set 2 degrees below apex temp.

    For Flow, I use a UPS only for protection because I have my pumps running from wall power controlled by a VDM, and the return on an eb8 fallback ON, so 2 devices would need to fail simultaneously for a problem to occur there (exception being power outages, which the UPS covers)

    For ATO, I run redundant floats and use OSC based code so if a float fails, it's both protected against overfilling, and protected from underfilling (using stuck switch detection). I also use a manual float valve as a line of last resort when I have room for one.

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

  6. #6
    Frequent Visitor
    Join Date
    Apr 2017
    Location
    Dubai
    Posts
    71
    Thanks, Zombie. I am in a similar field and did factor in backups on the assumption that Apex would have the lower failure rate and would therefore be the primary controller of temp, ATO, return flow, dosing, pump heads, refugium lighting and skimmer.

    Heater, ATO and return pump are backed up with secondary hardwired control assuming "fallback" works as expected (which it didn't today). For the return pump there is a Jacod backup pump that is supposed to fall back on when the 10V signal is lost to the primary VarioS.

    However today the EB6 also errored and Jacod did not turn on, heater did not turn off (heater thermostat caught that one!) and ATO - thankfully - failed off.

    I will switch ATO so Apex is the back up to the simple two float switch ATO that I added recently after reading horror stories on the Forum. Likewise I will revert to a home automation solution using zigbee wireless switches from Devolo to detect when the primary VarioS pump isn't drawing current (varioS will still be controlled by Apex 10V) and then turn on the Jecod after a delay. I will reinstall the Maxspect controller and let it run the gyres autonomously (so two 10V interface modules will become ewaste). The WAVs will stay under Apex control. I will remove 24V accessories as the 1LINK ACC power supply failed on today. Not a trainsmash as I don't use PMUP for ATO but do use it to circulate water in the refugeum.

    After today my observation is that in my context the Apex is less reliable than the individual controllers and I will alter control design accordingly to reduce the demand rate on the Apex to mostly monitoring. This also means that flowmeter based control is a nonstarter; it was a lovely idea that worked well when flow sensing worked. But the way the FMM fails over and the way the registers reset and lose their identity in programs is a fatal flaw. When/ if I get the FMM reconnected it will purely be for monitoring.


    Sent from my iPad using Tapatalk

Similar Threads

  1. Question: Radion programming issue for Advanced programming vs. lighting wizard
    By Afm32607 in forum AquaBus Modules, Probes, and Breakout Boxes
    Replies: 1
    Last Post: 07-08-2017, 18:03
  2. Question: Radion programming issue for Advanced programming vs. lighting wizard
    By Afm32607 in forum Apex Programming for Lighting
    Replies: 1
    Last Post: 07-08-2017, 18:03

Bookmarks

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •