Best setup for RO/DI unit

[Reefmaiden]

I've got a 100gpd 5 stage RO/DI unit. It came with the following cartridges (in order): Sediment filter, Carbon Filter, Carbon Filter, DI resin, and then the RO membrane.

I've been wondering if it would be better to replace the 2nd carbon filter with another DI resin instead? Or is the current setup the most efficient?

Opinions?

[JHardman]

Take a reading of your TDS for your tap water and RO product.

[mx_tang]

Erin,

Are you sure that DI resin is coming right after the carbon filter? That seems very strange, since DI is used as a final polish in water purification. Here's a picture I found of an AWI unit online:


http://imagehost.auctionwatch.com/bin/imageserver.x/00000000/theh2oguru/ReefkeeperLabeldRODIV.JPG

I would recheck the lines and make sure they aren't going in this order:

1.) feed water
2.) sediment
3.) carbon #1
4.) carbon #2
5.) RO membrane
6.)fast flush and DI bypass are attached on or near the membrane.
7.) DI resin

If Walter at AWI connected the lines wrong, then I'm sure you can call him and he'd send you out a new DI cartridge and some other stuff, as he's a great guy to work with as far as customer service issues.

This is what I would do to optimize my RO/DI unit:

1.) check the sediment and carbon to make sure they are still good. Usually when the sediment starts to look like as reddish/orange as a pumpkin, it's time to change them both out.

2.) take out the RO membrane and make sure it's truly a 100gpd or 75gpd unit. The model numer for the 100gpd unit should read "TW30-1812-100" and the 75gpd unit would be the same except for the 75 instead of the 100 at the very end. If it's a 100gpd unit, then when it comes time to switch out the membranes, I'd get a 75gpd membrane. These are the ones I just got from another vendor, but they're all good. In Az, our water temp and pressure is usually higher than normal, giving us more than the rated output on the membrane by 5-25%. Also, you're going from a 90% rejection rate for the 100gpd unit up to a 98% rejection rate for the 75gpd. On 1000ppm TDS water, this is the difference between 100ppm TDS and 20ppm TDS. It ends up being a huge factor when you start using your DI on a regular basis because the savings alone is worth swapping out a new membrane.

3.) Flush your membrane frequently. You don't need to flush after every gallon, but try to be consistent and flush before, after, or or both if you have the time. It only takes 30 seconds, but could bring your TDS levels down 1-10ppm, especially if you're working with high TDS off the get go.

4.) Bypass all product water going to the DI stage for at least one gallon. I bypass for 5g and then I turn on the DI stage. This is because you'll see an increase in TDS the moment you turn your RO/DI unit on from the initial low pressure to high pressure change. Some things just slip past the RO membrane during this changeover.

5.) Use a higher permeate to product ratio if you have an adjustable flow restrictor. If not, then you will have a capillary tube somewhere before the flush kit. If you cut this a bit, you can increase your wastewater ratio, but it's sometimes difficult to gauge whether you need to cut more or if you've cut too much. DM fitting has an adjustable flow restrictor that I am going to try to get for the locals because there were a lot of people who wanted them for their new 100gpd to 75gpd RO membrane switch. Even if you don't switch membranes, increasing this ratio will give you overall lower TDS in the water because the dilution of the waste is extended to a larger volume of water.

6.) change the resin cartridge when it starts climbing to nearly the same levels as the water coming after the RO membrane. There is a small window when your DI will adsorb 18, 19, 20, 200ppm TDS! Once it hits an adsorption threshold, the cation and anion beds still want to exchange ions, so they begin to leach compounds back into the water from highest to lowest atomic weight (such as silicate, etc....hydrogen would be last, obviously). This is even worse on a dual or triple stage DI because it becomes a domino effect and the second one prematurely fails, which starts leaching the same stuff to the 3rd stage. It's problematic if you don't catch it soon enough.

7.) use clean buckets. I suppose this is obvious, but don't go out and mix concrete in a bucket and expect the DI water in the exact same bucket to be pure...or even nontoxic for that matter. The water is only as clean as the vessels they are stored in, so keep them clean. I find the trashcans with lids work best. I just drill a small hole for the RO tubing and wait until it fills up.

That's really about it. I guess you only wanted to know whether your unit was hooked up correctly, which I honestly don't know. However, if you want to bring it to the DMS meeting, I'd be happy to take a look at it.

[mx_tang]

Originally posted by JHardman
Take a reading of your TDS for your tap water and RO product.

If the DI is, indeed, before the RO membrane, then taking a measurement after the RO membrane won't necessarily prove that the system is hooked up incorrectly. Although, after just one use, that DI should be shot. Okay, that previous statement was a lie...after about 3-6g of 500-1000ppm TDS water, the DI would be completely exhausted. :eek1:

The actual fittings are either DM fittings, John Guest, Jaco, or some other type of speedfit where you can actually push the ferrel/collet in while pulling the tube out. From here, you can test the DI output, RO output, feed output, and wastewater output. If you tell me what your initial TDS readings are, I can pretty much figure everything else out, up to and including your wastewater ratio.

OR

You can just follow the tubing and visually check where each line connects to in series.

OR

I can take a look at it.

OR

You can e-mail or call Walter at AWI and see what's going on.

OR

Just run out and get some maltballs and some cotton candy and be done with it. It won't solve the efficacy issue, but cotton candy is some good stuff.

[JHardman]

Originally posted by mx_tang
If the DI is, indeed, before the RO membrane, then taking a measurement after the RO membrane won't necessarily prove that the system is hooked up incorrectly. Although, after just one use, that DI should be shot. Okay, that previous statement was a lie...after about 3-6g of 500-1000ppm TDS water, the DI would be completely exhausted. :eek1:



Actually I assumed she had it plumbed correctly, but instead was interested if she had the super high TDS like you have or the lower like I have and how well the RO was working for her. If it was high I was going to suggest that she look at a 75gpd membrane instead of the 100gpd, thereby extending her DI resin.

BTW has the RO/DI stuff you ordered come in yet?

[Reefmaiden]

Ok, I guess I stated my problem (and the order of my canisters) wrong. So sorry! I guess I was just going in order of the canisters on the unit, not the actual order the water flows. I checked the tubes and it is actually going in the order of: sediment, carbon, carbon, RO, and then the DI.

I just changed out the DI resin on Friday and have made approx. 30 gallons of water off of it so far. Tonight before turning it on to make some more water I changed out the sediment filter and one of the carbon filters. I took a TDS reading after a few gallons had run through and I went from 863ppm TDS from the tap to 1ppm. So I think it is working ok.

My question was really if having two carbon filters is necessary? Or would it be better to use that canister for something else?

[mx_tang]

John,

The membranes, carbon block filter, sediment, and resin are here.

The adjustable flow restrictors and filter housing were backordered, but in stock as of Friday afternoon in Florida. This means that it either shipped Friday or Monday for Wednesday/Thursday delivery. You got to love our vendors :lol:

BTW, I ended up getting a case of them because the original $20 filter housing didn't have a bracket and the speedfittings. The final cost was $25 shipped, so I thought that was like getting the bracket and fittings for free. So, if you want to run any experiments, I've got 3 reserved (one is yours, of course), and 9 available. Perhaps people will want the nuclear grade resin? I don't know...I've talked to some people, including you, who say that they are running <100g per refill cartridge. That seems like either really high TDS or there's something wrong with the batches. I'm still on my first cartridge, but I probably don't make as much water as I used to make.

I actually ran another experiment for no reason whatsoever with an actual dual membrane system getting the same feedwater, but hooked them up to be parallel. The 75gpd membrane gave me 9ppm TDS out of 842ppm TDS and the 100gpd membrane (4 months old) gave me 59ppm TDS. They were both set at a 6:1 ratio, but what surprised me was how much water production I was getting out of the 100gpd membrane; I was getting close to 7 gallons/hour x 24hrs/day = 168gpd! Perhaps this might explain the not so great rejection rate. The 75gpd unit still produced more water, but only about 3.6g/hr x 24hrs/day = 86.5gpd. Even though the 75gpd is making 25% less water, in reality, it was making about half as much water production as the 100gpd membrane. However, I can live with making water half as fast if I maintain my TDS at sub 10ppm TDS.

BTW, I do ever possible mod to my RO/DI unit, which includes adding silicone to every single o-ring in the system (which = 11) to give everything a tight seal. You really need a tight seal on the membrane o-rings going into the membrane housing to ensure that all of the water traveling after the prefilters actually goes through the membrane and not around it (bypassing it).

[mx_tang]

Originally posted by Reefmaiden
Ok, I guess I stated my problem (and the order of my canisters) wrong. So sorry! I guess I was just going in order of the canisters on the unit, not the actual order the water flows. I checked the tubes and it is actually going in the order of: sediment, carbon, carbon, RO, and then the DI.

I just changed out the DI resin on Friday and have made approx. 30 gallons of water off of it so far. Tonight before turning it on to make some more water I changed out the sediment filter and one of the carbon filters. I took a TDS reading after a few gallons had run through and I went from 863ppm TDS from the tap to 1ppm. So I think it is working ok.

My question was really if having two carbon filters is necessary? Or would it be better to use that canister for something else?

Hmm, this is a very debatable topic to add two carbon stages instead of one. Unlike the DI cartridges, the carbon block actually acts as a sediment filter and through a very thin layer of carbon. It's very possible that some chlorine gets through the membrane, especially with high flowrates and pressures. Here's the carbon filters that I've been using as well as other locals.

The Chlorine Guzzler (http://www.kxindustries.com/matrikx/matrikx_1.htm)

It's interesting if you read the spec sheet where it has a maximum flowrate of 1g/min. People normally running RO units within normal TDS ranges won't ever go past a 6:1 ratio, but I know of at least 2 other people that use the same if not higher waste reject to product ratio. My dual 75gpd unit makes about 173gpd = 7.21gph = .12gpm. Since I'm running a 6:1 ratio, I make a total of 7volumes of water per unit time. So, .12gpm x 7 = .84gpm. That's cutting it kind of close. On the 100gpd unit I was using before, though, I went through the same wastewater to product ratio, but made about twice as much water. This would clearly put me over the 1gpm maximum flowrate that the manufacturer recommends.

For non dual membranes, running lower wastewater to product ratios, etc, this isn't a problem. I think it's a problem for people with high pressure, dual membranes exceeding 75gpd, and moderate to high wastewater to product ratios. The carbon is just for insurance, but if I were going to have 2 carbon stages, I wouldn't have 2 carbon blocks. I'd probably run carbon through one of the refillable cartridges that I have to give the water more contact time between the carbon. There was some discussion about chloramines and people talked about how to eliminate it from the tank. Some people ended up with 2-3 carbon stages on their RO/DI units.

If you want to use the canister, it has definitely been done before and only requires minor adjustmenest of the tubing. Although, I'd probably leave the unit as is and get an add-on DI filter housing/canister. I ended up with 9 extra filter housings, so lmk if you'd want me to bring one over to the meeting, etc (if they arrive on time).