RO membrane rejection

[Chilliton]

Looking for some help calculating RO membrane rejection percentage. My source water is 42 TDS measured with a handheld TDS meter. After RO only I get 1 TDS. Following the calculation 42 - 1 = 41 then 41/42 = .976 x 100 = 97.61%

How does low source water TDS affect membrane rejection calculation? Does this mean that the best performance I can expect from my membrane is 97.61% since I am at 1 TDS after RO only water? Or does my RO only reading need to be 0 TDS?

I am looking for some guidance in order to properly monitor my RODI system.

System is Spectrapure MaxCap manual flush 90 GPD .5 micron sediment, .5 micron carbon block, MaxCap super DI and SilicaBuster Super DI. water temperature is 86 F house pressure is 35 psi so i use a booster pump at 77 psi.

I understand that 97.61 is good rejection rate, but the average rate for my spectraselectplus 90 membrane is 98%. Is rounding up Ok in this case? if that's the case i am right where I need to be so this will serve to confirm my system's performance. Thanks in advance!

 

[mcarroll]

DI is what takes you to zero - not RO. You're right!

 

[Chilliton]

I am getting 0 TDS product water. Just wondering about the calculation pre DI.

 

[AZDesertRat]

You are testing at the lower limits of your TDS meter. Most are +/-2% of full scale accuracy and 1 is as low as it reads besides 0 so you are kind of in the dark. This is why I recommend to HM Digital COM-100 handheld which reads down to tenths of a part accurately. Your 1 TDS could actually be something like 0.1 or 0.6 or whatever which is much more accurate. You are not 0 but you may not be 1 either.

 

[Chilliton]

I am currently using the HM digital TDS 4TM, but given my source water I will be investing in the HM COM 100 in order to get the best readings as you suggest.

Thanks for your input. I have incorporated many of the steps you have explained for RODI system maintenance like the tee and ball valve for pre DI water samples, hand held TDS meter, and sanitation and filter replacement steps.

Also, my manual says that I need to change flow restrictor only if I change membrane size. So it's ok to replace my membrane and keep my restrictor as long as it is the same GPD? Again thanks for the advise my reef and I really appreciate it!

 

[AZDesertRat]

Yes but always check your waste ratio periodically as flow restrictors can and do plug up and change. I find it good practice to change it with the membrane myself, its only $6 and it gives me a chance to adjust it to exactly where I want it to be.

 

[mcarroll]

I agree about keeping an eye on the waste ratio....keep the restrictor as long as you monitor it.

Personally, I had terrible luck with the little tubes clogging and for some reason they were a real bear to get out of my system. Pliers and a lot of force was required every time. Plus, my waste ratio changes pretty dramatically when cold weather sets in, so I need to change it seasonally.

What a pain!!

So I use two "external" solutions now instead of the little tubes and haven't had any more clogs....and making changes is a snap!

The main restrictor is a "fast flush ball valve plus restrictor" from airwaterice.com...but I think I've seen the for sale elsewhere too:

$15...but it does more than a little tube restrictor.

For my seasonal changes I have one of these Mini Needle Valves from Buckey Field Supply that I pop onto the end of my waste line in the fall:

$5...I can dial in the waste flow rate, lock the adjustment knob with the locking collar, and the setting doesn't change for the whole cold season.

Been doing it this way for 5-8 years now....so much better than the old way!

$0.02

 

[Chilliton]

Great information. I will be checking my waste ratio next.

 

[AZDesertRat]

I had an expensive needle valve on my RO/DI and it cost me a new membrane when it clogged up and slowed the waste dramatically. I had the waste line where I could not visibly see it which will never happen again. Asr as the flush valve, they have no documented proven value other than to lighten your wallet and give you a warm fuzzy feeling. I have never seen even one independent study that shows otherwise.
I keep two capillary tubes, one timmed for summer and one for winter but found my water temp doesn't change drastically which is hard to believe in Phoenix, so I quit changing it each season and have over 8.5 years on my current RO membrane still at 99.43% rejection rate. 650-850+ TDS going in and its never ever been flushed.

 

[mcarroll]

the flush valve, they have no documented proven value

I'm not sure if I understood your statement correctly, but are you suggesting TDS creep doesn't exist?

Just in case...(or for anyone who hasn't run into this)

I monitor TDS of the product water (permeate) coming out of my RO membrane as well as coming out of my DI (final) stage.

TDS in the permeate water (non-waste line of the membrane housing) is always high after the filter sits not running for a while and pressure equalizes from the permeate to the waste side of the membrane.

That's TDS creep.

It may not be the biggest deal in the world, but I don't want to drop 150 TDS water from the RODI system into my barrel that's supposed to have 0 TDS water in it.

I use that "fast flush" valve to rapidly run water through the system until I see normal numbers on the TDS meter (usually just a few to several minutes). Once the numbers are back inline I put the restrictor in the "fast flush" valve back online by turning the handle, which once again forces water through the membrane, DI chambers and to my barrel. (I hope that makes sense!)

Works like a charm and I didn't get it for any other purpose.

I'm curious if this wasn't what you were talking about though....

P.S. RODI systems work best when the run continuously....in theory I should have a lower GPD membrane so it could just run and run and run in between uses. This would not allow TDS creep to happen. I'm just not sure there's a (good) way to engineer it like that for our purposes since we tend to use the water in batches and sometimes unpredictably.

 

[KStatefan]

Wouldn't flushing the permeate side accomplish this much faster than flushing the waste side?

 

[mcarroll]

Wouldn't flushing the permeate side accomplish this much faster than flushing the waste side?

Maybe theoretically, but I'm not sure how one would go about flushing the permeate side. Good idea tho!

I guess a valve would do it...wonder if there would be enough volume to notice the savings though?

 

[AZDesertRat]

Flush valves are on the waste side of the membrane so do not flush out TDS creep which migrates to the treated side of the RO membrane. A DI bypass valve, very different from a flush valve of flush kit is easy to install on the treated side between the RO membrane and the DI filter and can manually flush out the TDS creep water on start up. People get the two confused but they are not the same thing. One bypasses the flow restrictor in the waste line and theoretically increases the velocity and volume of water enough to scour the membrane surface but in reality does not since you are not increasing the incoming pressure or volume enough to do any good. The other sends the water on the treated side of the RO membrane to the drain or wherever you plumb it to until you close it, usually based on readings from a handheld TDS meter, and actually have proven value.

TDS is very real and something inherent with all membranes. You need to remember, as you mentioned, it is on the permeate or treated side of the membrane though. What happens is when you shut off the RO/DI system, the tap water containing XX TDS migrates or osmoses to the treated side. The contaminants or ions on the treated side have already been removed, your 98% rejection rate, so the water is agressively trying to attract everything it can to get back to the way it was in nature so pulls or attracts the TDS on the tap(waste) side until it balances out.

 

[mcarroll]

I understand what you're saying and can't disagree with your logic, but I can't say for sure that describes the whole picture.

What I know for sure is that the TDS output reading is high when I turn on the system and it seems to drop a lot faster when the restrictor is offline. So it "seems" like I may be doing my DI a favor. I am watching a real-time reading from a TDS meter vs "guessing" that this is happening, for what that's worth.

That valve also makes it easy to take the restrictor offline for the fall/winter when I use that needle valve on the waste line instead. I'm still happy if this were my only use-case - so much more convenient than those little tubes. And $15 spent usefully is an insignificant sum in this hobby. At the same time, saving a few bucks on something that causes you headaches is no value at all.

Apparently there's an even more-complete flush system than "fast flush + DI bypass"...huh.
http://theh2oguru.com/media/pdf_files/t/h/the_valve_positions_on_typhoon_iii.pdf

I guess I don't feel like my DI stage is exhausting quickly or this might seem more interesting. I will keep it in mind as usual though!

I've gotta add that my fast-flush/restrictor valve seems like it must be clog-proof. I've never had an issue with clogging in the ~8 years I've had it. (Knock on wood!)

 

[Chilliton]

So finally I measured my waste ratio. One minute run product was 310 mls. One minute run waste was 630 mls. So I am almost right on 2:1 ratio which is what my flow restrictor installed is rated for.

Thanks again for the info. [emoji2]

Tested at 77 psi and 46 TDS source water and 84.5f temp

 

[AZDesertRat]

You have pretty warm water. Make sure to check it again this winter, if the water temperature drops much the waste ratio will change a lot. Water becomes less viscous or more dense as it gets colder so the waste ratio will go down since it is harder for the molecules to pass or osmose through the membrane.

 

[Buckeye Hydro Verified Sponsor]

TDS Creep is easy to address with a 3-Way Valve on the permeate tube. Temporarily turn it one way to send TDS creep water to the drain, and then flip the valve the other direction to send the permeate on the the DI stage.

Russ