RODI plumbed to sump - pros and cons??

[jtm235]

This was my thought, and use this with a float valve in the sump. Then add some sort of leak detection is always a plus.

Appreciate all the input on this thread.

In my pet programming project I plumb directly from DI to the sump...using a combination of pretty much all of the equipment mentioned in this thread. I think my setup is pretty secure (knock on wood). Like Panda Jerk mentioned, redundant valves definitely help you sleep at night because solenoid valves are destined to fail eventually. I took the valve redundancy a step further and use a motorized ball valve between the tap line and RODI input. The US Solid motorized ball valves seem pretty industrial and less likely to fail than a solenoid based on design. The truth of the matter is though...How would you know if a redundant valve is in a failed state? You wouldn't unless you test them each periodically, which I also do (once per month). 1 redundant valve essentially halves your risk...not eliminating it.

Like Gtinnel and everyone else said...TDS creep is a real problem if not purged properly. Definitely need to purge the RO water if it sat for a while. I purge it every fill cycle...probably more than necessary.

My ATO system is designed as follows:

Flow:
- Motorized ball valve between tap and RODI input
- Solenoid valve to purge RO output for 120 seconds
- Solenoid from DI to float valve in sump's return chamber
- Float valve lets water into return chamber

Sensors & Safety:
- eTape liquid level sensor in sump's return chamber to initiate (and end) ATO cycle
- conductivity meter in return line (ATO cannot start unless minimum salinity is met)
- water leak safety valve on floor of RODI room (before motorized ball valve)
- water leak safety valve on floor of sump (before float valve in sump)
- ATO is programmatically throttled...only checks/runs 4 times per day with a max fill time of 10 minutes per session. With a 75 GPD RODI filter this is max 2 gallons per day.
- inline TDS meter from DI output to alert if any TDS > 0 coming in.

I believe the float valve in the sump is the most critical safety implementation - and should prevent catastrophic overfill 100% of the time, if it's working properly. I thought of adding a second float valve too but don't quite have the room. Regardless this piece is definitely on my scheduled replacement list every 6-12 months.

Also, I was using solenoid valves on the RODI side because US Solids didn't have any RODI-safe (non-metal) motorized ball valves at the time...But I think they make them now...will probably upgrade the solenoids to plastic MBVs. One big reason to use MBVs is they don't require constant power (less heat and less deterioration, no coils which will eventually go bad).

All in all, definitely no complaints so far...but redundant solenoids/valves are a must, along with 1 or 2 overfill protection mechanisms. Scheduled maintenance, testing and parts replacement are also vital. A lot to deal with but still beats manual IMO.

 

[gbroadbridge]

In my pet programming project I plumb directly from DI to the sump...using a combination of pretty much all of the equipment mentioned in this thread. I think my setup is pretty secure (knock on wood). Like Panda Jerk mentioned, redundant valves definitely help you sleep at night because solenoid valves are destined to fail eventually. I took the valve redundancy a step further and use a motorized ball valve between the tap line and RODI input. The US Solid motorized ball valves seem pretty industrial and less likely to fail than a solenoid based on design. The truth of the matter is though...How would you know if a redundant valve is in a failed state? You wouldn't unless you test them each periodically, which I also do (once per month). 1 redundant valve essentially halves your risk...not eliminating it.

Like Gtinnel and everyone else said...TDS creep is a real problem if not purged properly. Definitely need to purge the RO water if it sat for a while. I purge it every fill cycle...probably more than necessary.

My ATO system is designed as follows:

Flow:
- Motorized ball valve between tap and RODI input
- Solenoid valve to purge RO output for 120 seconds
- Solenoid from DI to float valve in sump's return chamber
- Float valve lets water into return chamber

Sensors & Safety:
- eTape liquid level sensor in sump's return chamber to initiate (and end) ATO cycle
- conductivity meter in return line (ATO cannot start unless minimum salinity is met)
- water leak safety valve on floor of RODI room (before motorized ball valve)
- water leak safety valve on floor of sump (before float valve in sump)
- ATO is programmatically throttled...only checks/runs 4 times per day with a max fill time of 10 minutes per session. With a 75 GPD RODI filter this is max 2 gallons per day.
- inline TDS meter from DI output to alert if any TDS > 0 coming in.

I believe the float valve in the sump is the most critical safety implementation - and should prevent catastrophic overfill 100% of the time, if it's working properly. I thought of adding a second float valve too but don't quite have the room. Regardless this piece is definitely on my scheduled replacement list every 6-12 months.

Also, I was using solenoid valves on the RODI side because US Solids didn't have any RODI-safe (non-metal) motorized ball valves at the time...But I think they make them now...will probably upgrade the solenoids to plastic MBVs. One big reason to use MBVs is they don't require constant power (less heat and less deterioration, no coils which will eventually go bad).

All in all, definitely no complaints so far...but redundant solenoids/valves are a must, along with 1 or 2 overfill protection mechanisms. Scheduled maintenance, testing and parts replacement are also vital. A lot to deal with but still beats manual IMO.

The more pieces you have, the more likely that one will fail.

It also means you need to consider and test every possible failure mode which increases exponentially the number of tests that need to be performed any time you make any change, no matter how small.

It is definitely possible to over engineer a solution which results in an increase in cost and complexity which may not be cost effective.

It's all fun, but at some point a line has to be drawn dependant on risk appetite - after all this is not human life support project.

 

[jtm235]

Over engineered is a subjective term. Well-thought-out design, implementation and quality parts are what matter. This is my expression of direct-plumbed ATO system with minimal failsafe mechanisms. TDS and conductivity meter are not necessary…just practical for me integrate. My point is, with a properly functioning and installed float valve (also critical for a gravity-fed ATO)… a direct plumbed ATO system cannot overfill, and the most likely point of failure is the solenoids/relays. In fact it’s assumed solenoids (and MBVs) will fail eventually. It’s a testable and replaceable part…redundancy and periodic testing of solenoids is not that difficult. And if one is going through all this trouble to automate ATO...then it shouldn't be too difficult to implement an additional alerting mechanism when the sump level gets too high (if all else fails). To answer OP's question from my experience:

Pros of direct plumbed ATO:

1. no lugging/pouring water
2. no need for RODI storage tank or ATO reservoir (which also may have a float valve that needs to be periodically tested and replaced)
3. no need to ever think about topping off
4. Stable salinity.
5. More time to focus on other aspects of tank.
6. Fun pretending you’re better than the peasants who have to carry RODI water to their fish overlords.

Cons of direct plumbed ATO:

1. periodic testing (5 minutes per month) and maintenance (replacement if necessary) of solenoids, float valve and water level sensor/s.
2. higher risk of overfill (depending on quality of design and implementation).
3. need to have well-engineered equipment/controller.
4. I promise it’s not cost effective…haha.

 

[PetitReef]

I love it.
I travel a lot and having the RODI directly plumbed means I dont have to worry about tank top off's.

I use the Aqua One Smart ATO RO.

It refills the RODI reservoir when it is almost empty (10 gal reservoir).
The ATO itself pumps from that reservoir.

It has two optical sensors (one at the bottom of the reservoir to turn on the solenoid valve attached to the line from the RODI filter and another at the top to determine when full). Because of this design there is no additional TDS creep.

The RODI line is attached to a float valve in the reservoir which will prevent the system from flooding if there is an electronics or sensor failure.

Been running for almost two years now with no issues.

AUTOAQUA

www.autoaqua.com.tw

Never put your trust on a solenoid alone, even this one that just got stuck open even without electricity.
so my GHL program and extra level sensor didn't do anything as the solenoid was technically off.

thanks fully I had a float valve as redondancy and it just saved my tank and probably my floor.
I wish i could have another reservoir for RODI as i had in my previous appartment, but no place on this one to add a gravity fed one. need to try to find a solution, other wise i will serial solenoid on the RODI feedline for redondancy.