Causes for low rejection rate in RO units?

[cabrego]

Hi,

I have been reading 0 TDS (after RO/DI) for the last couple of months, but now my TDS is 7 or so, as of yesterday.

Here is the breakdown of my TDS:

Tap>>>>164
RO>>>>>15
DI>>>>>>7


According to this my RO is only rejecting at ~91%.

My first thought is that my prefilters probably need to be replaced. At least some of them do. I knew this would occur at some point because I chose to buy a cheaper RO/DI so it is not suprising that they need to be replaced after about 1000 gallons or so.

IF the prefilters are going bad would that cause a decrease in the RO efficiency?

The membrame is advertised as a GE Desal membrane and should have around 97% rejection for my water pressure of 50psi.

[Randy Holmes-Farley]

Well, the DI is almost depleted and should be replaced at 1 ppm TDS, IMO. The prefilters will not impact the rejection rate or DI depletion rate significantly, unless the sediment filter is clogged and is causing low pressure on the RO membrane. Do you have a pressure gauge there?

The drop of about a factor of 10 is what I see with my unit, which is a new, top of the line Spectrapure (about 100 ppm TDS to 9 ppm TDS or so). How much of a drop you see depends a lot on what is in the tap water. Some things, like ammonia from chloramine and silicic acid from treatment of the water to boost pH at the water company can penetrate RO membranes more easily than most ions, since they have an uncharged form that can penetrate.

That said, if the line pressure is too low, you'll get poor rejection.

[virginiadiver69]

I just want to stress the point that Randy already made...
don't underestimate the effect of low pressure going into your filter.

You say you have 50 psi but where is that measured? Even if that's measured at the first stage...that is only minimum. You could consider getting a booster pump and give your RO/DI 80 psi. If that doesn't take care of the problem I would look at your membrane. I understand Dow Filmtec 75 or 100 gpd to be the ones that most people count on.

[cabrego]

<a href=showthread.php?s=&postid=13577930#post13577930 target=_blank>Originally posted</a> by Randy Holmes-Farley
Well, the DI is almost depleted and should be replaced at 1 ppm TDS, IMO. The prefilters will not impact the rejection rate or DI depletion rate significantly, unless the sediment filter is clogged and is causing low pressure on the RO membrane. Do you have a pressure gauge there?

The drop of about a factor of 10 is what I see with my unit, which is a new, top of the line Spectrapure (about 100 ppm TDS to 9 ppm TDS or so). How much of a drop you see depends a lot on what is in the tap water. Some things, like ammonia from chloramine and silicic acid from treatment of the water to boost pH at the water company can penetrate RO membranes more easily than most ions, since they have an uncharged form that can penetrate.

That said, if the line pressure is too low, you'll get poor rejection.

I am confused, you say your new Spectrapure only has a rejection rate of 91%. With that logic my GE and your Spectrapure happen to show the same rejection rate. And your DI is more efficient than mine, if you have 0 TDS. This would point to my DI being either exhausted, or just plain not very efficient. My unit is fairly new and my DI reason has only at 10-15% color changing.

That being said, I am trying to determine what I need to replace to 1) Not exhaust the Membrane prematurely,
2) Get me down to 0 TDS.
3) Improve the overall performance of my unit, if cost is not drastic to do this.

The pressure I measure is the input water pressure.

[Randy Holmes-Farley]

I am confused, you say your new Spectrapure only has a rejection rate of 91%. With that logic my GE and your Spectrapure happen to show the same rejection rate.

That is correct, I am saying it is the same and not necessarily something you are going to be able to change.

A DI should always give 0 ppm TDS until it becomes depleted and TDS starts to rise. Definitely replace the DI now. Other things may or may not need to be replaced.

I discuss RO/DI here:


Reverse Osmosis/Deionization Systems to Purify Tap Water for Reef Aquaria
http://reefkeeping.com/issues/2005-05/rhf/index.htm

[Randy Holmes-Farley]

How are you measuring the TDS? In line meters, or later? If later, how are you doing it? A little salt, (very, very little) will boost 0 ppm TDS RO/DI to the value you got. So unless it is inline, or in a clean cup collected directly, it may just be something that came later. :)

[cabrego]

<a href=showthread.php?s=&postid=13578255#post13578255 target=_blank>Originally posted</a> by Randy Holmes-Farley
How are you measuring the TDS? In line meters, or later? If later, how are you doing it? A little salt, (very, very little) will boost 0 ppm TDS RO/DI to the value you got. So unless it is inline, or in a clean cup collected directly, it may just be something that came later. :)

I am using a handheld meter to measure the TDS after the RO/DI. I typically use a clean glass to measure the sampled water. This method has worked pretty good, and I have been able to measure 0 TDS when the unit was new. When ever I get questionable results, I make sure to double or triple check my results, just incase there is some sort of residue.

Logic would lead me to believe that if I could get my membrane efficiency improved, that I would be able to get my TDS to zero.

Does the water contact time with the DI make a big difference? I am just thinking that my RO membrane feeds a 4gallon tank. And when I open my RO/DI valve water from the tank then pumps from the holding tank, through, the DI resin fairly quick and into my fish tank. The water is flowing at around 1-2 gallons per minute at this time. After that water then begins to flow at the rate the membrane is rated at for a given pressure, temperature, etc. and the contact time with the resin is significantly longer. This seems to make sense what do you think?

IF this is true unstored water would have a better TDS reading than the initial 4 gallons.

[AZDesertRat]

Are you taking samples directly from the RO membrane and DI filters or from the holding tank?
Try sampling directly from the membrane for the RO only TDS. Next take the holding tank out of the loop and make some DI directly from the RO into the DI filter. This may not work now since your DI resin is exhausted but try it with fresh DI.
I think your holding tank may have high TDS due to TDS creep since it is fairly small and probably prone to frequent starts and stops. This is common with pressure or bladder tank setups. With my previous system I first started with a single 5 gallon pressure tank and found the TDS was high so I added a second pressure tank so cycling was not as frequent. Eventually I bought a new 14 gallon pressure tank and eliminated the two small ones and got even better results. The other thing I did was to make sure I completely emptied the system at least every two weeks and started with fresh water to eliminate the build up of TDS. My DI lasted much longer that way.

A properly packed DI cartridge should be able to handle 2 GPM no problem. By properly packed I mean a full 20 oz vertical DI canister and cartridge packed tightly with no head space to channel or short circuit and hooked up with the normal bottom up DI filter configuration so all resin gets wetted.

What membrane do you have? What is your water pressure available at the membrane after the prefilter and carbons? What is your water temperature? What is your waste ratio?
How old are the filters in the unit now? Do you change them at 6 month intervals and do you disinfect the system at that time?

[cabrego]

<a href=showthread.php?s=&postid=13582034#post13582034 target=_blank>Originally posted</a> by AZDesertRat
Are you taking samples directly from the RO membrane and DI filters or from the holding tank?
Try sampling directly from the membrane for the RO only TDS. Next take the holding tank out of the loop and make some DI directly from the RO into the DI filter. This may not work now since your DI resin is exhausted but try it with fresh DI.
I think your holding tank may have high TDS due to TDS creep since it is fairly small and probably prone to frequent starts and stops. This is common with pressure or bladder tank setups. With my previous system I first started with a single 5 gallon pressure tank and found the TDS was high so I added a second pressure tank so cycling was not as frequent. Eventually I bought a new 14 gallon pressure tank and eliminated the two small ones and got even better results. The other thing I did was to make sure I completely emptied the system at least every two weeks and started with fresh water to eliminate the build up of TDS. My DI lasted much longer that way.

A properly packed DI cartridge should be able to handle 2 GPM no problem. By properly packed I mean a full 20 oz vertical DI canister and cartridge packed tightly with no head space to channel or short circuit and hooked up with the normal bottom up DI filter configuration so all resin gets wetted.

What membrane do you have? What is your water pressure available at the membrane after the prefilter and carbons? What is your water temperature? What is your waste ratio?
How old are the filters in the unit now? Do you change them at 6 month intervals and do you disinfect the system at that time?

Ok,

To answer your questions,

1. With respect to the holding tank, right now any water that I test originates from the holding tank. Another, troubleshooting technique would be to actually remove the tank from the loop.

2. The membrane is a GE desal 100 gpm.

3. I am not sure about the water pressure at the membrame. i will have to investigate that. I am also not sure about waste ratio, what is the best way to measure that?

4. The system is only a couple of months old, and I have used an estimated 3-400 gallons of RO. Assuming, my pressure switch is working correctly.

5. The DI resin does not meet your requiements for being properly packed. It is a horizontal placed 18oz tube.

Is the DI resin exhausted without question? Based on the measurements it is still working, but not as effectively. based on color it seems to be 10% gone.

[Randy Holmes-Farley]

Color is not a good way to gauge a DI. If the output really were half of the input, and it read 0 ppm TDS at one time using the same method and it is not contaminated by salt in the collection container, it is past time to replace it.

[AZDesertRat]

Testing from the tank will give you readings of the water in the tank with accumulated TDS not what the membrane is actually producing on long runs. Try disconnecting the line coming directly out of the membarne housing and test from that point. I think you will find it is much lower than what you are seeing in the tank.
Color changing resin is a poor indicator of resin condition, especially if not packed well since it will seperate and stratify.
To improve the DI system you presently have, take it off the top of the membrane bracket and mount it to the wall next to the RO/DI so the water enters from the bottom and has to flow out the top. First make sure the resin is packed tightly, disconnect the DI, fill it to the top with fresh resin, tap it on the table or counter several time, fill it again, tap again etc. until there is no space left and the sponge insert compresses slightly when you put it back togeteher. Now mount it vertically with bottom up flow and you will see a big improvement since all resin will come into contact with water and there is no channelling or short circuiting. Next make sure you drain the 4 gallon tank completely every week or two to use up and high TDS water in the system and the DI will last much longer.

[cabrego]

Thanks for the info, I will do some of this trouble shooting after work today.

Thanks!

[cabrego]

<a href=showthread.php?s=&postid=13582493#post13582493 target=_blank>Originally posted</a> by cabrego
Thanks for the info, I will do some of this trouble shooting after work today.

Thanks!

Hey guys here is an update, I was a bit busy yesterday, so the only thing I was able to do was verify where the pressure gauge is placed. It is actually placed before the RO membrane (after the prefilters), so my ~50 psi is directly to the membrane.

I also observed a couple of interesting things a few minutes ago.


Tap TDS>>>>>>170
RO/DI>>>>>>>>10
RO/DI (bypassing holding tank) >>>>>>2
RO (bypassing holding tank)>>>>>>>>>7 ~96% rejection rate.

I think when u consider my pressure is just below 50 psi, a rejection rate of 97% is very respectable and closer to the manufacturer's specifications of the RO membrane, so I pleased about that.

I suppose there is an argument that can be made that, the rejection rate is probably a significantly lower if the reported numbers are based on a system without prefilters, which cut into the efficiency of the membrane.

As previously mentioned it appears my DI resin could benefit from being replaced. I picked some up last night, so I will most likely replace it sometime this week.

There is also a taste filter in my system could this contribute to TDS creep as well? It maybe interesting to see what my results are when I remove the taste filter.

[AZDesertRat]

Absolutely remove the taste and odor filter. If you use the unit for drinking water install the T&O filter close to the drinking water faucet so only that portion of water passes through it, the ydo contribute to TDS since they are granular carbon which crumbles and turns to dust with time.
97% is about as good as you can expect from an off the shelf membrane with any use on it. You may find the TDS improves when the membrane has run longer. Take some samples after it has run for a hour or so and you will probably find it lower still.
What is your waste ratio? 4:1 ? Colder winter water raises the waste ratio some which lowers the psi at the membrane thus lowering production. Keep it at 4:1.

[cabrego]

<a href=showthread.php?s=&postid=13599872#post13599872 target=_blank>Originally posted</a> by AZDesertRat
Absolutely remove the taste and odor filter. If you use the unit for drinking water install the T&O filter close to the drinking water faucet so only that portion of water passes through it, the ydo contribute to TDS since they are granular carbon which crumbles and turns to dust with time.
97% is about as good as you can expect from an off the shelf membrane with any use on it. You may find the TDS improves when the membrane has run longer. Take some samples after it has run for a hour or so and you will probably find it lower still.
What is your waste ratio? 4:1 ? Colder winter water raises the waste ratio some which lowers the psi at the membrane thus lowering production. Keep it at 4:1.

The T&O filter can not be any closer to the faucet.

Water flows from the pressurized holding tank to either the T&O filter and then faucet, or through the DI membrane. Water will take the path of least resistance, i.e. valve opening.

I have not yet measured the waste ratio, is it as simple as measuring the waste water, and the RO water simultaneously, and then taking there ratios after some time?

[AZDesertRat]

Thats good as long as its RO only at the faucet. If you are taking the RO only samples from that point though you might want to pick another location to sample from as the readings will be high due to the carbon. I installed a tee and ball valve right off my RO so I can test directly from the membrane myself.

[cabrego]

<a href=showthread.php?s=&postid=13600377#post13600377 target=_blank>Originally posted</a> by AZDesertRat
Thats good as long as its RO only at the faucet. If you are taking the RO only samples from that point though you might want to pick another location to sample from as the readings will be high due to the carbon. I installed a tee and ball valve right off my RO so I can test directly from the membrane myself.


I have not reported RO results thorugh the faucet. I have Teed into the RO output line to get the results I have reported so far. The best results are when I bypass the RO Tank and the water I sample is "freshly made" into the cup.

So, in summary, I believe we have concluded the following:

1. my RO membrane is working properly. ~97% rejection.
2. Water from the holding tank increases the output TDS significatly. 2ppm vs. 10ppm.
3.Pressure at my RO input is ~50 psi.

Did my method to determine waste ratio check out okay with you?

[Buckeye Hydro]

We recommend against providing water from an RO pressure tank to the DI. I read through this thread quickly so I'm not certain if that is the way your system is configured.

Russ

[cabrego]

<a href=showthread.php?s=&postid=13617265#post13617265 target=_blank>Originally posted</a> by BuckeyeFS
We recommend against providing water from an RO pressure tank to the DI. I read through this thread quickly so I'm not certain if that is the way your system is configured.

Russ


Russ the RO tank does feed either the DI resin otuput or the faucet output. I see why you would recommend not using it to feed the resin.


It seems that if I would like the most efficient use of my filters, I need to bypass the RO tank. Which means I can not use it for drinking water.

[Buckeye Hydro]

Feel free to email us and request a plumbing diagram to our reef/residential series systems. They use a pressure tank to feed the faucet, but not the DI.

Russ