350 PPM TDS before RO, 10 PPM after RO, 0 PPM after DI - chewing through resin

[TinyChocobo]

Can I use my Saltwater ALK/PH tests on my tap water to use the calculator at https://www.hamzasreef.com/Contents/Calculators/CO2DIDepletion.php to determine CO2 or do I have to use fresh water test kits? @Randy Holmes-Farley

 

[TinyChocobo]

Just read through the entire thread and I don’t believe I saw it addressed... do you you use a 3 way DI bypass valve to protect the DI resin from TDS creep when you first fire up your RODI unit? Also, it sounds like your using color changing DI resin. I rely on my inline TDS meters to tell me when my DI resin exhausts rather than the color change. I’ve had what looked to be an exhausted DI cartridge continue to produce 0 TDS water for moths after a full color change.

I am watching the TDS with an in-line TDS meter and won't be replacing it until it starts reading above 0. That said - thanks for the tip on the color change running out before the resin.

I do have the auto-flush valve but I think I'm going to plumb in a line between the membrane and the resin to dump to the drain for a few minutes when I start up the system. The auto flush only dumps to drain for 25 seconds or so - which is probably not enough time?

Lower the water pressure, keeping it around 55-65 psi , i know the RODI's say they can handle up to 90 or whatever. but keep it at that level and even though it'll be slow output, it'll keep your resin alive longer

I thought about that - lowering the pressure. The membranes I have should be 99% rejection. I still think 550 ml/min restrictor is too little and am waiting on the 800 to show up - hopefully in the next couple of days.

 

[Water Dog]

I do have the auto-flush valve but I think I'm going to plumb in a line between the membrane and the resin to dump to the drain for a few minutes when I start up the system. The auto flush only dumps to drain for 25 seconds or so - which is probably not enough time?

Your auto flush valve is different from a DI bypass valve. The auto flush feature just flushes the rejection side of the RO membrane. The DI bypass valve lets you drain the built up TDS creep that would otherwise go directly into your DI cartridge, vastly shortening its life. Install this 3 way valve before your DI cartridge and flush out the TDS creep water for a few minutes before redirecting flow back through your DI resin.

https://www.bulkreefsupply.com/3-way-ball-valve-with-john-guest-1-4-push-connect.html

 

[TinyChocobo]

Your auto flush valve is different from a DI bypass valve. The auto flush feature just flushes the rejection side of the RO membrane. The DI bypass valve lets you drain the built up TDS creep that would otherwise go directly into your DI cartridge, vastly shortening its life. Install this 3 way valve before your DI cartridge and flush out the TDS creep water for a few minutes before redirecting flow back through your DI resin.

https://www.bulkreefsupply.com/3-way-ball-valve-with-john-guest-1-4-push-connect.html

I don't have a 3-way valve but I installed a T with two regular valves - so I can divert flow down the drain instead of through the DI. I ordered a 3-way valve and will swap out the Tee and two valves for the 3-way once it arrives.

Right now I'm running 80 PSI through the membrane and watching TDS before DI - and then I will turn off the pressure pump and watch TDS to see if there is a change.

 

[TinyChocobo]

It took about 5 minutes to get from ~20PPM to 9PPM coming out of the membranes with the booster pump at 80PSI. That's as low as I can go with the pump on. With the pump off I'm getting ~62 PSI - and I am watching to see if TDS creeps back up or goes down with the pump off.

 

[TinyChocobo]

I'm seeing the same TDS at 60 and 80 PSI. It starts at ~20 and settles down to 9~10 after start-up. I think the only difference between 60 PSI and 80 PSI is that I'll make water a bit faster at 80.

 

[TinyChocobo]

TDS Creep takes far less time to settle at 80 PSI than 60. About half the time it seems.

 

[TinyChocobo]

The biggest gain in DI life that I can see is still running the membranes in parallel - hopefully the parts to test that will be here tomorrow or the day after. I'm going to test the TDS of my tap water, TDS after the first membrane, TDS after the second Membrane [ in series ] and then TDS after both membranes in parallel.

That said before I move over to parallel, I am going to try the 800 ml/min restrictor instead of the 500 ml/min restrictor to see what kind of difference it makes on the rejection rate.

 

[TinyChocobo]

I just tested a single membrane and my TDS after the membrane once TDS creep settles is 7 TDS. With dual membranes in series it's 9 TDS which is 22% higher but not a huge amount of TDS overall.

I tested the source water just prior and it is ~247 TDS.

This works out to 97.1% Rejection Rate with a single membrane and 96.3% rejection with dual membranes if my math is right - so I am really starting to think the biggest issue here is not having a properly sized restrictor. The membranes I am using are Dow Filmtec High Rejection 99% Membranes - so I really should be getting closer to 99% IMHO.

As soon as I have an 800 ML/hour restrictor to test I will test dual-in-series membranes, single membrane, and will also test dual parallel for completeness.

 

[75pxatr]

I had a similar situation. My tds was over 400. I added a water softener and Spectrapure high rejection membrane. When I make a large batch of water, the TDS out of the membrane into the Resin is between 2-5. Huge savings.

 

[TinyChocobo]

I had a similar situation. My tds was over 400. I added a water softener and Spectrapure high rejection membrane. When I make a large batch of water, the TDS out of the membrane into the Resin is between 2-5. Huge savings.

The membrane I am using is a high rejection 99% just like the spectrapure.

I think a water softener would help a lot but I don’t have room in the utility room for one and I doubt the land lord would spring for one.

 

[Bruce Burnett]

iSpring 20 gallon tank with 14 gallon Storage Capacity Reserve Osmosis Water Storage Tank https://www.amazon.com/dp/B00439I86C/ref=cm_sw_r_cp_api_406HAbH54PZ8S

I have a 55 Gallon water barrel coming Tuesday I will be using with the RO float but for now this pressurized tank is what I have.

This is the same as having it go into ATO. You need to shut of water supply to RO unit each time you are done. You get bleed through. Best thing to do is have a line before DI that you turn on and let just ro water flow into bucket until until TDS drop then shut off and let flow through DI. When you first start the flow the TDS first coming out of RO section can be high. My TDS is around 400 even with soft water it is still high 300s. When first turning on TDS out of ro section is close to 200 after around 10 minutes it drops to 7-9 on soft water and 24-29 on unsoftened water. I figure when everything is good, softener working right and filters new I get 6-7 on RO so getting 98% rejection on 2-150 GPD membranes. I have been thinking of do the 3 chamber method discussed by BRS to see if I am going through one resin faster than the other. The typical rejection rate for the 75 GPD is 98-99% at 4 to one. At 3-1 it should still be 97% or better people with tap water less than 100 TDS do not realize how lucky they are.

 

[Bruce Burnett]

I've been told by both Melve's Reef and BRS that the pressure pump is suppose to go after the pre filters.

Always better as the prefilters are meant for normal tap pressure 50-70 psi. Mine is definitely after as I use a 125 PSI booster pump the smaller and lower pressure pumps do not work on a 300 gpd system.

 

[TinyChocobo]

This is the same as having it go into ATO. You need to shut of water supply to RO unit each time you are done. You get bleed through.

I am turning off the water supply and booster system as well as closing valves.

The storage tank fills and the system stops even if I don’t do that. There is simply nowhere for it to go.

I’m not draining a little from the tank at a time - when I use it - I empty the tank 100% and then let it refill.

Best thing to do is have a line before DI that you turn on and let just ro water flow into bucket until until TDS drop then shut off and let flow through DI.

I set this up today. It’s worth noting that none of the things I’m doing with this setup to save resin I had to do on my last setup. I can only imagine how much longer the resin on my last system would have lasted if I did this extra stuff!

When you first start the flow the TDS first coming out of RO section can be high. My TDS is around 400 even with soft water it is still high 300s. When first turning on TDS out of ro section is close to 200 after around 10 minutes it drops to 7-9 on soft water and 24-29 on unsoftened water.

I see 30-50 TDS for ~5 minutes and it drops to 9 with dual membrane in series and 7 with single membrane.

I figure when everything is good, softener working right and filters new I get 6-7 on RO so getting 98% rejection on 2-150 GPD membranes. I have been thinking of do the 3 chamber method discussed by BRS to see if I am going through one resin faster than the other.

I ordered a third chamber so I can do anion,cation,mix. I too am curious how fast I’m going through one or the other. It also will last longer as one gets used up faster than the other and I can just recharge that specific one or replace it rather than replacing both in the mix.

Just waiting on the resin to be in stock at BRS.

With as much water as I will make with this system over the next 5-15 years the investment in another state is tiny.

The typical rejection rate for the 75 GPD is 98-99% at 4 to one. At 3-1 it should still be 97% or better people with tap water less than 100 TDS do not realize how lucky they are.

True. My issue is that I’m not at 4:1 with my current restrictor. I really think using a properly sized restrictor will make a big difference.

Dual membrane. Proper restrictor. Proper pressure. TDS creep to drain before DI. Flushing the membrane. Running 3 chamber DI. At the end it should be much more efficient.

I am still going to try/test membranes in parallel just to see if I can do it and where my TDS falls. With proper restrictors and enough supply pressure I think it will give me the lowest TDS even if it has the most waste.

 

[TinyChocobo]

BRS is out of everything but the mixed bed resin. Any other good anion/cation resin suppliers?

I could use up the mixed bed stuff and then separate anion and cation when I recharge it - but would really prefer to just find it already separated and ready to go.

 

[Bruce Burnett]

I am turning off the water supply and booster system as well as closing valves.

The storage tank fills and the system stops even if I don’t do that. There is simply nowhere for it to go.

I’m not draining a little from the tank at a time - when I use it - I empty the tank 100% and then let it refill.

I set this up today. It’s worth noting that none of the things I’m doing with this setup to save resin I had to do on my last setup. I can only imagine how much longer the resin on my last system would have lasted if I did this extra stuff!

I see 30-50 TDS for ~5 minutes and it drops to 9 with dual membrane in series and 7 with single membrane.

I ordered a third chamber so I can do anion,cation,mix. I too am curious how fast I’m going through one or the other. It also will last longer as one gets used up faster than the other and I can just recharge that specific one or replace it rather than replacing both in the mix.

Just waiting on the resin to be in stock at BRS.

True. My issue is that I’m not at 4:1 with my current restrictor. I really think using a properly sized restrictor will make a big difference.

Dual membrane. Proper restrictor. Proper pressure. TDS creep to drain before DI. Flushing the membrane. Running 3 chamber DI. At the end it should be much more efficient.

I am still going to try/test membranes in parallel just to see if I can do it and where my TDS falls. With proper restrictors and enough supply pressure I think it will give me the lowest TDS even if it has the most waste.

With as much water as I will make with this system over the next 5-15 years the investment in another state is tiny.

Make sure you use you TDS meter and not the color change, good way of monitoring is have sensor before membranes and before each di section this way you monitor each section. I get 500-700 gallons gallon out of my resin and I still don't have line before resin to flush out creep but I make 70 gallons each time which last me 2 weeks for ATO

 

[TinyChocobo]

Yeah I got some extra meters. I plan on watching before first membrane. Before second. After second. Before DI 1, before DI 2, before DI3, before DI 4, and then after DI.

This way I can easily see if any part of the system is performing poorly. I am going to run cation, anion, mixed 1, mixed 2.

The reason I decided on 4 is because that will let me exhaust stage 3 100% and start using stage 4 before any TDS makes it to the final supply. This way I can pull stage 3 and swap stage 4 to 3 and put a fresh cartridge in 4.

My utility room is quickly turning into a water purification plant.

 

[PhreeByrd]

Interesting read. A couple of things struck me while reading...

If the goal is to reduce waste water, then the original BRS setup with the waste water from the first R/O membrane feeding the second would be the correct setup. I didn't do the math to determine how significant it is, but this arrangement will reduce the final R/O purity. Since the second membrane is processing very dirty water, it is not going to produce the same purity output water as the first membrane, and D/I resin life will be affected by this. I also don't see how this would produce 150gpd from 75gpd membranes, but I would expect production above 75gpd, at least.

If the goal is to increase pure water production, then the parallel R/O membrane setup would be preferred, at the cost of higher D/I resin consumption and higher waste water rates. You should be able to get close to the 150gpd production with this arrangement. Basically it's two typical 75gpd systems running simultaneously.

If the goal is to maximize R/O output purity and therefore reduce D/I consumption, then the best arrangement is to have the second R/O membrane fed from the pure water output of the first membrane. The second membrane is then processing already purified water, producing purer water and leaving less cleanup for the D/I resin to do. Output rate would of course suffer, and you would effectively have a 75gpd system even with two membranes.

As evidenced by reading the thread, one can easily tie up the cost of a couple of year's worth of D/I resin trying to modify all of this stuff. Fun experiment, though.

 

[TinyChocobo]

If the goal is to reduce waste water, then the original BRS setup with the waste water from the first R/O membrane feeding the second would be the correct setup. I didn't do the math to determine how significant it is, but this arrangement will reduce the final R/O purity. Since the second membrane is processing very dirty water, it is not going to produce the same purity output water as the first membrane, and D/I resin life will be affected by this. I also don't see how this would produce 150gpd from 75gpd membranes, but I would expect production above 75gpd, at least.

The goal is not to save water or to reduce waste . The goal is to produce as much RO/DI as I can as quickly as I can with as little resin usage as possible. That said the output of a single 75 GPD would be sufficient for my needs except in an emergency of some kind - in which case I could enable a second membrane if needed via valving.

If the goal is to increase pure water production, then the parallel R/O membrane setup would be preferred, at the cost of higher D/I resin consumption and higher waste water rates. You should be able to get close to the 150gpd production with this arrangement. Basically it's two typical 75gpd systems running simultaneously.

Indeed. I think it depends on whether or not I can get enough pressure to run two membranes in parallel. I don't see why it would require any more pressure than two in series - but I'm not an engineer specialized in fluid dynamics so trial and error is the game .

If the goal is to maximize R/O output purity and therefore reduce D/I consumption, then the best arrangement is to have the second R/O membrane fed from the pure water output of the first membrane. The second membrane is then processing already purified water, producing purer water and leaving less cleanup for the D/I resin to do. Output rate would of course suffer, and you would effectively have a 75gpd system even with two membranes.

This is not something I had considered - but I will definitely give it a try for testing sake. It may even help eliminate TDS creep as the water that passes the membrane on the second membrane will only be ~7 TDS to begin with. I think the biggest issue is that I will likely be sending 80% of the permeate from the first system down the drain as waste from the second system. That said I may be able to get by with a much smaller restrictor on the second membrane if I go this route to keep pressure high and to force more of the permeate past the membrane.

As evidenced by reading the thread, one can easily tie up the cost of a couple of year's worth of D/I resin trying to modify all of this stuff. Fun experiment, though.

My goals are long-term [5~10 years]. I'll break even on DI cost eventually. That said - anything I have now that I don't use - will be put towards the next RO system I install so it's not going to waste. At some point I'm going to have a new house and will be setting up a new RO/DI system there . That said what I learn from all of this will also help me make sure that new setup is more efficient from the get-go.

 

[ColbyNTX]

Y’all are lucky. My tap waters TDS is usually around 740! .