Comprehensive RO/DI efficiency study, especially e-bay units.

[AZDesertRat]

Its not a hard and fast mathematical rule but it just seems from years of bench testing the 2% increase doubles DI life no matter what the starting point is. Charles Mitsis at Spectrapure has worked on this for probably 20 years now and has found it to be very accurate.

 

[jdieck]

<a href=showthread.php?s=&postid=12960767#post12960767 target=_blank>Originally posted</a> by old salty
Say you have 100ppm and your membrane removes 96%; this leaves you with 4ppm. If you go to 98% which is a 2% increase, you end up with 2ppm. Your DI resin should last twice as long filtering 2ppm as it would 4ppm.

Now, say you have 100pm and you remove 90%; this leaves you with 10ppm. If you increase RI efficiency by 2% to 92%, you end up with 8ppm. Wouldn't you obtain a 20% increase in DI lifespan rather than doubling the lifespan?

Yes your calculations are right and follow the way the industry measures membrane rejection.
The statement should be "For every halving of the inefficiency your DI cartridge will last twice".
You will double your DI capacity going from:
90% to 95%; or from 95% to 97.5%; or from 97.5% to 98.75%; or from 98.75% to 99.38% percent and so on.

In my opinion all this discussion is not a matter of which unit is more efficient but one of cost benefit.
If you can get the same TDS with any system, you need to see the other trade offs to make a decision. Some will be interested in the initial cost of the unit versus the cost of DI replacement while some business may want to tie it to the labor cost of replacing cartridges when you pay an employee by the hour and so on.
Also as mentioned by Randy you also need to consider many other factors other than just TDS in the water you need to filter to really determine which unit to use. Things like not only inlet temperature and pressure but amount of dissolved CO2 and/or sediment, or if it is treated with chlorine, chloramine, sodium hydroxide, initial PH and content level of silicate, nitrate, and so on. Variations in all these things can really make a huge difference in the performance of two different systems.

 

[carlso63]

<a href=showthread.php?s=&postid=12957292#post12957292 target=_blank>Originally posted</a> by Gravesj1s
Carlos- I didnt see anywhere AZdesertrat made any claim of bashing. I think the point the original poster was looking to gain was in general consistientcy . It could actually be beneficial if enough posters responded.Especially if those from the same area (same water supply)using different units were seeing different results.Lets give it a chance.
Maybe we could include pressure and temp for those that are able. I will post mine as soon as I get a chance.
-Graves

For the record, I was not referring to AZ but to the OP of the thread and the (deceptive) difference between the title of the thread and how the thread progressed

 

[carlso63]

<a href=showthread.php?s=&postid=12959336#post12959336 target=_blank>Originally posted</a> by ukroylisa
"And, when the dust settles, the 0 ppm I am getting out of my unit is the SAME as the 0 ppm you are getting out of your unit - no matter what psuedo scientific terminolgy you pull out of your whazoo - ZERO TDS IS ZERO TDS. PERIOD"

Is this true? With an accuracy of 2% of full scale deflection (FSD) for a typical TDS meter, and a FSD of 1999 ppm, your "zero" might be 39ppm. Even with a FSD of 999, your TDS could be 19ppm. How many people calibrate their TDS meters? (Their efficiciency should stay the same of course)
Roy

Actually, BOTH our statements are true. Your comment about uncalibrated meters possibly being "off" by a significant amount is true. My comment that a "0" TDS reading from one unit is the same as a "0" reading from another unit (assuming we were checking both units output with the same TDS meter) is also true.

In the past it has been inferred by some that certain units "0" water is somehow "better" than other units "0" water - which is basically not true. (There may be minor differences in parameters like elec. conductivity readings but that is really not relevant to our purpose of purifying the water for use in an aquarium...)

 

[carlso63]

My only other qualm with the OPs methodolgy is he is focusing on the ebay units advertised claims of either 90 or 93% (different websites, I would assume) instead of comparing using the real world data that he asked us to provide.

IMO he is doing this because the advertisements poorly written "data" can "prove" his point - while the actual real world data posted here puts his claims in jeopardy.

Because, skimming through the results it looks like:




"Name Brand results"(in percent):

99.4 & 99.1 = Spectrapure MaxCap
97.8 & 99.0 = Ocean Reef
99.3 = Typhoon
97.6 = Kent
93.8 = Melev


Average = 98.0


"Ebay brand results"(in percent):

97.2, 97.4, 99.0, 98.5, 90

Average = 96.4 with the 90% 100GPD unit, 98.0 without


Interestingly enough, if you remove "Bubble girl" and her 90% rated 100 GPD unit from the ebay side, the "average" efficiency (at least from those who posted results here) is 98.0% whether you bought a name brand or an Ebay unit. (I did not include jflecks results only because he uses dual RO cartridges on his unit, which is not the norm here)

Although (obviously) this small sampling could hardly be considered "proof" of anything, I do feel that it echoes the general point that I made earlier, which was - - -

The efficiency of the ebay units is very very close to most of your more expensive name brand units (the main exception, IMO, would be the MaxCap - but that unit is a level of quality above everything else, name brand or not; that's why you pay a premium price even over other name brands)

In fact, for the most "accurate" numbers comparo here , IMO, I would remove both the premium MaxCaps and the questionable Melevs from the "name brand" table and "bubblegirls" from the ebay table and you would end with 98.04% for the ebay units and 98.42% for the name brands - or an average difference in efficiency of around 0.4%

And for the "average" person (with, say, 250 ppm tap water), that 0.4% difference means that with the ebay unit your after RO TDS will be 5 ppm and with the name brands it would be 4.25 ppm , a difference of less than 1 ppm = meaning IN THE REAL WORLD your DI resin would last approx. 16% longer.

In my real world case my DI lasts around 9 months (270 days) ; same usage and water supply with the name brand unit my DI would last approx 315 days (16% longer). In 10 years of use I will need to change my DI approx 13.5 times; if I had the "average" name brand unit I would be changing the DI 11.5 times over the same 10 year period.

It costs me $14 every time I change my DI resin.

So, DI resin costs for the ebay unit in my real world situation are $28 more than the "average" name brand here - spread out over 10 years...

Or a whopping $2.80 per year more.

So much for the "hidden costs"... I'll take spending 3 bucks a year more in costs vs. ponying up 75 - 100 bucks more at the time of purchase and then "saving" 3 bucks a year...

 

[iairj84]

I tend to agree with the posts that say it's in the filters and it's more or less the same with the exception of MaxCap...

Mine's a Water General 100gpd from ebay.
Filters are probably 9ish months old.
RO is 9 months
Haven't replaced the DI Resin yet and it's spent pretty much as I make a good amount of water. and my results without the DI is fine with me for now.

Pre RO hovers anywhere from 140-250 but lately it's been the lowest I've seen 140... As tested today.

At 140 my post RO TDS is 3... Giving me a .978 efficiency rating. I haven't replaced the DI lately and my output water is 3... which I'm okay with seeing as our earth's topoff is anywhere from 8-12 from the sources I've seen... I'm okay with anything under 5 and I am under that.

I will replace prefilters and DI resin within the month.

 

[old salty]

<a href=showthread.php?s=&postid=12961210#post12961210 target=_blank>Originally posted</a> by jdieck
Yes your calculations are right and follow the way the industry measures membrane rejection.
The statement should be "For every halving of the inefficiency your DI cartridge will last twice".
You will double your DI capacity going from:
90% to 95%; or from 95% to 97.5%; or from 97.5% to 98.75%; or from 98.75% to 99.38% percent and so on.

Ok. I was just wanting to make sure my math was right. Once a geek...


If it wasn't for AZ's input on RO/DI units, I wouldn't have nearly the system I have now. He's Mr. Jones and I'm trying to keep up. I go through way less DI resin from following his advice.

 

[Tang Salad]

AZDesert: Do you know if cheap membranes generally last as long as the expensive membranes?

 

[AZDesertRat]

The life of the membrane is determined mostly by three things, tap water quality, quality of the filters preceding it and how well it stays flushed.
Tap water quality controls everything. The worse it is the harder it is going to be on the membrane. Note almost all manufacturers performance claims state the used softened water in their testing.

The quality of the prefilter and carbon block have a direct effect on membrane life. If the contaminates never reach the membrane in the first place they do not have to be removed. A 10 micron nominal rated prefilter is not going to do the same job as a 0.5 micron absolute rated prefilter, or even a 1 or 5 micron nominal rated prefilter. Unless you have high sediment loading or neglect to change your prefilters on a regular schedule there is no reason not to go to a better quality prefilter, even with much better filtreation the difference in pressure drop or head loss is minimal between a 0.5 micron and a 10 micron. I use a 0.2 micron myself and push the change schedule out to a year with no negative results.

Same goes with the carbon filter. Most people don't realize the granular activated carbon that somes with some less expensive units has a lifespan of about 300 total gallons. Thats 60 gallons of product or good water and 240 gallons of waste at a 4:1 waste ratio. I can't stress enough how important a high quality carbon block is. The 20,000 gallon Chlorine Guzzlers are a great filter for just about any situation. Thats 4,000 gallons of good water and 16,000 gallons of waste at 4:1. They are also a 0.5 or 0.6 micron near absolute rated filter so trap any particulates the prefilter misses. Its important to have a prefilter rated the same as or smaller than the carbon block though or you are plugging the carbons pores and rendering it useless. The block is made up of billions of tiny pores or pockets and they need to be kept clean to work. Put a 10 or 5 micron nominal rated prefilter in front of it and you have just defeated the purpose of having a good carbon and probably cut its life drastically.

Continous wasting is very important to a membrane and keeping the waste ratio at close to 4:1 is the best way to accomplish this. Reducing the waste ratio reduces the velocity and does not carry the accumulated solids away from the membrane so they begin to solidify on the membranes surface. If you have never seen this, it looks like salt brine or table salt which flakes off when you cut it open. No amout of using a "flush valve" will have any effect once this accumulation is present. Large municipal and commercial units get around this by utilizing chemical cleaning but that is not common in hobbyist units. Keep it at 4:1 and it should last.

So, use good filters, keep them serviced and keep it flushed.

 

[Tang Salad]

Thanks AZ.

I can't see that this informal pool served any purpose, other than to ignite a few tempers, because it simply does not (cannot) account for all the variables.

 

[areze]

because it didnt support your preconcieved notions? seems like it did serve a purpose, it showed ebay units are not far and away complete wastes of money.

it still doesnt account for the other "benefit", you are getting the skeleton of the unit for less money also. your ongoing 3$(if thats accurate) annual loss can be erased by simply replacing the filters with something else. however, at 3$ a year, the ebay filters are significantly higher savings than 3$.

 

[aquaman67]

I don't understand how you can throw out one person's results and then say

<a href=showthread.php?s=&postid=12961754#post12961754 target=_blank>Originally posted</a> by carlso63
instead of comparing using the real world data that he asked us to provide.

I think I did use the data proved. According to my calculations 80% of the people who replied appeared to have gotten a pretty good deal.

20% did not.

And you want to throw the 20% out when it's the 20% that is the entire point of this whole thread.

It's not the 80% I'm concerned about, it's the 20%.

Any time I mention, "Hey look at what you're getting off ebay, make sure you test it and got what you paid for." I get people who chime in and say "Well, mine's good so they all must be."

There are still advertisements that show pictures of Dow membranes. Do you get a Dow membrane when your unit comes? Maybe, maybe not.

I guess they didn't have a picture of the membrane you'll actually be getting, so that's ok. It's not misleading in anyway...

 

[Esquare]

Original system from Filter Direct on Ebay
Installed 3/08 on my 120 gallon total system (DT 54 sump approx 70)
Water changes approx 10% to 17 % weekly

Water General RD-102 6-Stage RO+DI system, 100 gpd, single output.

1st stage 5 micron sediment filter, 2 ¾” x 9 7/8”
2nd and 3rd stage carbon filter, 2 ¾” x 9 ¾” high performance coconut shell carbon
4th stage TFC/TFM type membrane, 100gpd 1 ¾” x 11 ¾”
5th stage DI filter
6th stage DI filter

cheap hand held TDS meter from filter direct
Dual inline TDS monitor Model DM-1 from HM digital

Tap water 140
RO output 1
DI output 0

RO efficiency 99%


Yesterday I noticed that my DI water was up to 3 ppm so I figured it was time to change the resin. After researching it, I found the SpectraPure MaxCap D2 for $100 and figured I might as well go that route as replace the resin.

New system
Reconfigured yesterday

Same as original except the DI portion has been replaced with the MaxCap D2

HM com-100 hand held TDS meter
2 Dual inline TDS monitor Model DM-1 from HM digital

Tap water 140
RO output 1
First DI output 0
Second DI output 0

RO efficiency 99%

The reservoir is reading 0.7 ppm (the new hand held unit is the only one I have the measures that low).

Question for Azdessertrat, you mentioned your DI output is less than 0. Would you mind telling me what it is so I can compare my .7? I just want to see how well my frankenfilter is performing. Also, I plan on replacing the sediment and carbon filters when they die with some better quality filters, will the spectrapure filters fit in my canisters (the current filter measurements are above)?

Thanks

 

[chrismunn]

spectrapure 70 gpd MPDI RO/DI. purchased from LFS in cal. 5 years old, never changed any filters. 3 gallons per week.

hm digital dm-1 dual in-line tds meter. purchased from marine depot.

in = 410 ppm

out before DI = 8 ppm

99.512 % efficient.

 

[AZDesertRat]

The HM Digital COM-100 handheld meter will measure to 000.01 accuracy unlike other meters which only read to 001. . The COM-100 can also be used as a conductivity/resistivity meter which is much more sensitive by at least a factor of 10. Besides that meter when I was doing my long term testing I borrowed a lab grade benchtop Thornton resistivity meter for my tests which will read down to the full 18.2 megaohms most units will never produce.

 

[aquaman67]

<a href=showthread.php?s=&postid=13571340#post13571340 target=_blank>Originally posted</a> by leeweber85
I bought that unit from filterdirect a couple years ago. It had a lot of good reviews here on RC, but personally I don't think it is that great of a unit. The RO membrane in mine only removed about 90% not anywhere close to the stated 97-98%. I would burn through the DI resin after about 50 gallons.... it should have lasted for somewhere around 800-1000 gallons. In case you don't know, DI resin is pretty expensive too.

I would recommend going with one of the units from the filterguyz or buckeye field supply, both sponsors here on reef central and are both VERY helpful in answering ANY questions.

I would also recommend a 75gpd membrane and a vertical DI chamber.

Where were you about 2 months ago?

LOL! It's all good.

BTW the e-bay average is down to 95.35 now with TWO units scoring in the 90 percent efficiency. And only 66.6% got a good deal...