Homemade RO Not Working as Anticipated...Help Please!

[Lbiijjs]

Couple of second year sugar makers here. Thank you to all on this forum for freely sharing a tremendous amount of know-how by everyone here who, like ourselves, has caught the incurable 'maple bug'.

So, we were ready to take a big leap in our productions this year with the introduction of our new homemade RO unit into our small time operation. It consists of a cheapy Chinese 100GPD booster pump, a 5 micron pleated pre-filter, and (3) TFM-100 RO membranes (100 GPD) plumbed in series. (Hope I attached the photo below correctly)

http://img.tapatalk.com/d/14/03/16/6edy9anu.jpg

On our initial flush/rinse on the membranes, the pump made 85 PSI and moved 1 gallon (permeate & concentrate combined) through the system in 6 minutes with a 3:1 ratio. We thought that was not too shabby. We expected some drop off in performance with sap but not sure how much.

Anyway, here is what we ran into with our first attempt to run sap through...

Sap was approximately 2.75% and around 50 degrees F. Our system ramped up to 165 PSI and held steady, processing only 1 gallon of sap in 22-23 minutes with a 55-60% removal rate. We adjusted the needle valve quite a few times because our concentrate flow petered out to a fast drip several times. The pressure briefly fell to 145 PSI when adjusting the needle valve to correct concentrate flow but then quickly climbed back to 165 PSI and stayed. We thought maybe our system got a blockage of some sort. But, after moving the pressure gauge to test inlet pressure of all membrane housings, all were the same at 165 PSI.

What is wrong with our setup?

Our UNEDUCATED thoughts on this are...
1.) that we could have warmed the sap more before RO, and maybe that would help flow rate
2.) that we could have the wrong membranes for this ($25 each on eBay)

Any insight on this would be greatly appreciated.






2013 - Bit by maple bug...23 taps on buckets, 11 gallons gold
2014 - 28 taps on buckets

[not_for_sale]

What do you use in your prefilter?

[not_for_sale]

Also, you concentrate flow needs to be flowing as fast as the pump will allow. Otherwise your membranes will foul.

[bowhunter]

I don't think you have enough concentrate flow through the membranes. These membranes won't last very long if you pull out too much permeate on each pass. The best place to run them is between 10 and 15 % recovery. Recovery is the % of flow through each membrane that's being removed as permeate. The ratio of concentrate out of the membrane should be 5:1. So for a 1/2 gallon per hour of permeate you need to have 2 1/2 gallons of concentrate. There are two ways you can run your unit. 1) reduce the pressure so that you only remove about 30-35% as permeate per pass and recirculate concentrate back to your feed tank. or 2) add a second needle valve and recirculate concentrate from the last membrane back to the inlet to the pump. The recovery is pretty important and on my system it seems to be very important when processing sap.

Two other factors are working against you. Your sugar content is pretty high at 2.7%. Sugar content has a huge impact on the amount of permeate the membrane will pass because the opposing force to permeation is the osmotic pressure of sugar which increase as sugar content increase. Second, membranes are size using 77 F water. The colder the sap the slow the lower the permeation rate.

I think you membranes are probably fouled, but able to flush out most of the material with a permeate wash.

[Lbiijjs]

Pre-filter is a 5 micron pleated polyester, GE model FXWPC. How can concentrate flow as fast as the pump will allow and still maintain enough back pressure on the membranes? Isn't that the purpose of the needle valve, to somewhat restrict the concentrate flow?


2013 - Bit by maple bug...23 taps on buckets, 11 gallons gold
2014 - 28 taps on buckets

[bowhunter]

You do have to partially close the needle valve so the pump will generate pressure. As the pressure increases the flow through pump will decrease depending on the pump characteristics. If you have 100 GPD membranes they were designed for a feed rate of 600 gallons per day or 25 gallons per hour. You need a pump capable of supplying 25 gallons per hour at your operating pressure. It sounds like your pump was only doing 10 gallons per hour at 85 PSI so your pump is too small for the size membranes you have. Home RO membranes are rated at 50 PSI and 77 F. On cold sap at 200 PSI you have to derate them about 50%. So a 100 GPD membrane will permeate about 50 GPD on sap at those conditions.

2013 - Bit by maple bug...23 taps on buckets, 11 gallons gold
2014 - 28 taps on buckets[/QUOTE]

[Clinkis]

I have a similar RO. A few things....

With around 3% sap your should be able to remove 50-60% water on a single pass. Anymore then that and membranes will foul. Also your pressure seems a high. My rig runs around 125psi this might be causing membranes to foul prematurely. When my system is running optimally I usually proccess 7-9 gallons per hour of cold raw sap. System needs to be thoroughly flushed after every 45 gallons with permeate and needle valve wide open. I'm not familiar with the brand of membrane so not sure if that might be part of your problem too.

[Clinkis]

This is the specs on aquatech pump that most people, including myself, use on these types of RO's. It seems to have similar specs with respect to flow rate. Don't know all specs on this Chinese pump but sounds similar and should work.

http://www.cwwltd.com/product_files/704_Aquatec%2BBooster%2BPumps.pdf

[Clinkis]

This is the specs on aquatech pump that most people, including myself, use on these types of RO's. It seems to have similar specs with respect to flow rate. Don't know all specs on this Chinese pump but sounds similar and should work.

http://www.cwwltd.com/product_files/704_Aquatec%2BBooster%2BPumps.pdf

Actually...after looking again......I was looking at the specs for the 6800 pump specs. Most use the 8800 pump and it has flow rate of 500 GDP (1.4LPMx60x24/4=504gpd).

So in retrospect...very likely the flow rate of your pump is way too low. May need to upgrade your pump

[hodorskib]

Having run my system for 3 years now I can say two things you want to check are:
1. Your feed tank must be above your pump - these pumps are not designed to draw liquid they are designed to boost/push. So by having your feed tank above your pump gravity will solve the problem.
2. Most residential membranes are designed to handle a maximum of 125psi beyond that you could actually be destroying them. Some folks have run them beyond that pressure but 165psi seems really high.

The highest % I have managed to obtain is 9% and at that level they foul around every 40-50 gallons. I am able to run over 100 gallons at around 8% with 150gpd membranes at around 115psi without any issues.

Good luck hope this helps.

[Clinkis]

Having run my system for 3 years now I can say two things you want to check are:
1. Your feed tank must be above your pump - these pumps are not designed to draw liquid they are designed to boost/push. So by having your feed tank above your pump gravity will solve the problem.
2. Most residential membranes are designed to handle a maximum of 125psi beyond that you could actually be destroying them. Some folks have run them beyond that pressure but 165psi seems really high.
The highest % I have managed to obtain is 9% and at that level they foul around every 40-50 gallons. I am able to run over 100 gallons at around 8% with 150gpd membranes at around 115psi without any issues.

Good luck hope this helps.

I agree having your feedtank above your RO will help boost your pressure. That said, my RO is mounted above a 45 gallon drum and my aquatech 8802 will suck that drum dry and function properly. I think, as this quote mentions, the pressure may be too high and causing problems.

[Lbiijjs]

We thought 165 PSI was too high as well. We're were hoping it would be around 125 PSI. Our pump is gravity fed from our sap barrel mounted above, so that should be fine.

With the needle valve all the way open there was very little pressure in the system and concentrate flow seemed strong. Probably it was just zipping by the first membrane in the series and out. As soon as the valve was tweaked down a bit, permeate flow started but was never more than a trickle. Concentrate flow was nearly equal, a trickle. I agree that our membranes seem like they could be fouled. That makes sense. Would changing to 150 GPD membranes help matters?

I'm trying to wrap my head around the pump end of the equation though. If the pump is able to move water through the membranes at 85 PSI with good flow, it seems to me that I should be able to find a point between 85 PSI and 165 PSI where I get some separation of water from sap with the same pump. It certainly is building more than enough pressure against the membranes. Or, am I looking at this all wrong? Would using a higher flow pump allow me to open the needle valve more, causing higher concentrate flow and lower system pressure, while still pushing some of the water through the membranes?

Sorry for machine-gunning you guys with all the questions. Thanks again to you all.
By the way, our first boil of the season was complete today. We got almost 1.5 gallons of sweetness! :-)


2013 - Bit by maple bug...23 taps on buckets, 11 gallons gold
2014 - 28 taps on buckets

[Clinkis]

I won't pretend to know the math that will prove or disprove if the pump is the problem. All I know is what works and the Aquatec pump with lower pressure and higher flow rate works. On my pump there is a little Allen screw on the diaphragm housing that can be turned to adjust the pressure. You may want to check to see if your pump has something similar. I'm using 100gpd membranes and others have had success using 150gpd so either one should work.

I did a ton of research before I built my RO and almost everyone was using the aquatech pump and filmtec membranes so that is what I used. Unfortunately you have created a bunch on new variables by using different parts so it may take some tinkering to get it to work.

I would do a really good flush with luke warm permeate and peroxide with needle valve wide open to see if it is just membranes fouling and also see if you can adjust pressure down on your pump. Another thing to try which will help lower your pressure is put your feed tank on the ground and make the pump pull the sap up into your RO.

Another thing I just notices is you should be checking your pressure after your last membrane and before your needle valve. This will give you a truer read of watch pressure is being applied to the membrane. This will also give you an indication if your pre filter is plugged so you may want to move it or add another one.

I also had grief last year with certain types of pre filters. The ones that look like then tightly wrapped string, for whatever reason, would not work.

Hope this helps. Good luck!

[bowhunter]

These membranes are designed to see a minimum flow to prevent fouling. The pump is capable of enough pressure but not enough flow to get adequate velocity through the membrane. There's also a factor called recovery that you have to watch. You really can't permeate more than about 15% of the flow through the membrane per pass. If you try to go much higher the membrane bogs down quickly on maple sap. You have two choices. You can remove 15% as permeate per membrane per pass and recycle concentrate back to the sap tank or you can install a recycle from the permeate out of the last membrane back to the pump suction and use two needle valves. You must have about 600 gallons per day of total flow through the membrane to remove 100 gallons per day as permeate. It's just one of the limitations of RO. With three membranes in series the last membrane is not very effective because the %sugar is much higher going into it.

To your question a larger membrane will make the problem worse. You don't really have enough flow through the membrane with the current pump. You need an aquatec 8852 and you have to keep the pressure below 100 psi or the pump flow really drops off. I have a spreadsheet model of a small three membrane series set up like yours recycling concentrate back to the sap tank. If you have the aquatec 8852 and limit the pressure to 90 psi, you can increase 70 gallons of 2% sap to ~5.5 to 6% in about 7 hours. The flows would be ~0.2 gpm of concentrate and 0.1 gpm of permeate.

[Lbiijjs]

Thanks for your clear explanation, bowhunter. I would be interested in taking a look at that spreadsheet model you referred to.

Looks like its time to upgrade the pump to the Aquatec 8800 series that has been used successfully by others here. We looked for a pressure adjustment on our current pump, and it looks like there isn't one. I'm sure there is still going to be more trial and error, but probably a whole lot less error thanks to all you guys. Whenever we get our new pump hooked up, I'll post our results...Time to go pump shopping!


Sent from my iPhone using Tapatalk

[sugarphil]

Hello, guys, I'm really interesste in building my RO system this year.

I plan on using pretty much the same setup as displayed in this tread picture with slight modification:

- 2 membranes in parallel (1 pass- recirculation from holding tank)

- I'm thinking about going with a Nanofiltration membrane @ 200GPD (50$ each) + standard plastic house housing (12,50$ each) (The specs sheet shows this output for 15% recovery rate, 150PSI and 77F)

- Correct me if I'm wrong but: for 15% recovery rate of 400 GPD, I get 2666 GPD of feed flow rate (about 1,85 GPM) So I would need to find a pump that can reach closest to 150PSI while delivering 1,85 GPM (or 111 GPH).

- Now this brings me to the burning question of the pump.

1- The cheapest pump I could find for this output seems to be this one:

http://www.ebay.com/itm/DC-24V-8L-min-100W-High-Pressure-Micro-Water-Pump-Automatic-Reflux-Valve-Car-/262198248944?hash=item3d0c3bc5f0:g:kfcAAOSw54xUYdG A

Altought the pump states a flow rate of 8 lpm (pretty close to 2 GPM) I really doubt it can do this at it's max rated pressure (being a cheap chinese made pump I have yet to find it's pump pressure VS output curve)

2 - Another option I'm looking into would bbe this Procon pump on sale here :

http://www.wateranywhere.com/product_info.php?products_id=10089

Now with this pump and from the linked spec sheet on the website, it seems it can output 124 GPH at 150PSI, pretty close to what I'm looking for. The problem with this pump is that I'l need to find a 56C frame with at least 1/2 HP (lower HP would make the motor run at it's peak hp capacity all the time)

For the same price I could probably go for 2 cheap chinese pumps.

So what do you guys think ? Would anybody with experiences with other pumps think another option could meet my systems requirement ?

Thanks
Phil

[bowhunter]

Here are some thoughts:
1. I wouldn't go with the little car wash pump. The maximum working pressure is only 80 psi and I'm sure it's not designed for continuous service.
2. I'm not sure how much sap you're trying to process, but the membrane will deliver much lower performance processing cold sap than the rated capacity. I wouldn't be surprised if you averaged less than 200 gallons per day of permeate with this set up, maybe even less.
3. You have the flow rates correct.
4. You can recirculated back to the tank, but I think recirculating concentrate back to the pump suction works much better. The concentrated sap has a tendency to foul the filters pretty quickly. This avoids that problem.
4. The ProCon pumps are excellent little pumps.

Dave

[sugarphil]

Thanks for the info.

The setup is for my father in law's brand new sugar shack with 450taps under vaccum. Hes got a CDL evap rated at 300-450 taps but I dont know if thats for buckets or vaccum setup.

He used to operate a smaller sugarshack farther on is lot with 250 maples but he found it inconvenient because it was 15 minutes of ski-doo to go to and he didnt have electricity and a bathroom.

The 250 tap setup wasnt under vaccum. Last year we got 25gal, his best year was 40gal two years ago.

I wonder what kind of output we'll get with the 450 taps and vaccum ? Pounds per tap seem a little lower here in quebec city than in southern.

Back to the RO, I take it from other posters setup that the white plastic cassing can take continuous 150+ psi ?

Another question I would have is about the nanofiltration vs RO membrane. It seems NF has a bit more output than full RO and I would think fouling could be less of a problem because of a weebit larger holes in the membrane. The manifacturer still specified to me sugar molecules would be too large for the NF pores.

Any thaughts on this , anybody?

Oh and I you state that recirc to the pump intake would help against fouling but in the same sentence you state that membrane foul easier has concentration increase? My reasonning behind recirc to the thank is that I'd like to get the most water out has quickly has possible.

[bowhunter]

If the evaporator is sized for 300-450 taps, that most likely means on buckets so the bush should produce more sap on vacuum. How much more probably depends on how tight the system is and how much vacuum it can hold. Based on research I've seen even 15-16 inches of vacuum will probably double the sap production. This means you may need to process 900 gallons or more of sap on good flow days. Based on the size of the evaporator you will want to remove about 50% of the water so you can handle the high flow days.

The size of the RO then depends on two things. 1) How much sap you have to process and 2) how many hours a day you want to run the evaporator. I would not recommend operating a home made RO 24 hours unattended for part of the time without installing some safety interlocks for low pressure, high pressure, etc to protect the pump. If you decide to run the RO 8 hours a day then the RO must be capable of processing 110 gallons per hour while removing 50% of the water.

Nano filtration does remove water more easily than RO membranes and under mild operation the sugar molecule will not pass through the membrane. The down sides for nano filtration are 1)the membrane does pass more of the mineral content into the permeate, 2) some of the nano filtration membranes will pass sugar at 13-14% sugar content and 3) some of the nano filtration membranes are very sensitive to pH during caustic cleaning and can be more easily damaged.

As you recycle back to the tank the sugar concentration is increasing assuming you are not adding fresh sap at the same time. My point about sugar concentration is that once the membrane starts to see higher concentration, say above 10% sugar, fouling of the membrane increases more rapidly. I don't like to recirculate back to the sap tank because I don't like to pump concentrate through the filter but many times producers will have to recirculate back to the tank. You just have to change the filters more often and be careful that you don't plug the filter enough to damage your pump.

[wiam]

When I was short on tanks I used to recirculate. Back to feed tank. I drew from half way up the tank and put concentrate back in bottom. This seamed to help with fouling problems. It does work better with 2 tanks.

[sugarphil]

If the evaporator is sized for 300-450 taps, that most likely means on buckets so the bush should produce more sap on vacuum. How much more probably depends on how tight the system is and how much vacuum it can hold. Based on research I've seen even 15-16 inches of vacuum will probably double the sap production. This means you may need to process 900 gallons or more of sap on good flow days. Based on the size of the evaporator you will want to remove about 50% of the water so you can handle the high flow days.

My God, where in-over our heads then, because I don't think we expected to get double amount of water that the evap was rated for. We've never use vaccum before but from what you say, I don't think we have the processing power to process it all. For starter, the receiving tank is only 250 Gal :o



The size of the RO then depends on two things. 1) How much sap you have to process and 2) how many hours a day you want to run the evaporator. I would not recommend operating a home made RO 24 hours unattended for part of the time without installing some safety interlocks for low pressure, high pressure, etc to protect the pump. If you decide to run the RO 8 hours a day then the RO must be capable of processing 110 gallons per hour while removing 50% of the water.

Nano filtration does remove water more easily than RO membranes and under mild operation the sugar molecule will not pass through the membrane. The down sides for nano filtration are 1)the membrane does pass more of the mineral content into the permeate, 2) some of the nano filtration membranes will pass sugar at 13-14% sugar content and 3) some of the nano filtration membranes are very sensitive to pH during caustic cleaning and can be more easily damaged.


We won't run the RO full time because he seem a bit scared about the whole idea of RO (too complex for him) and we don't even run the evap at night. I've decided to come up with somekind of low cost setup so I can prove the concept to him while still not putting to much in the project.

I like the small size housings/membranes because of their small upfront cost and I think they could be reused in a larger setup if he's interrested in scalling up. Basically, I'd like to get the most out of about 300$.

[bowhunter]

Didn't mean to put a fright into you. Too much sap isn't always a bad thing. Your plan to start with a small set up to prove it out is fine. There are one or two really good designs for small systems posted here. If you use the small diaphragm pumps they're pretty fool proof. You can let them run overnight recirculating back to the sap tank with no one watching. A lot of guys use these systems while they're asleep or at work to reduce the sap volume to cook. Good luck.