Flux Capacitor ( Home Made RO)

[not_for_sale]

I built an evaporator out of a steam table pan and an old barrel stove and was in need of a flux capacitor to multiply my GPH.

Since today was my day with my boys I thought we would try to build one. Thanks to the ideas on this forum I had ordered some stuff to built one.

I have some pictures attached.

Small run of Sap over the past couple of days with ice in it this morning clocked in at 1.9 % and 32-33 degree Fahrenheit.

6780
6781

First pass with three membranes in series concentrated the sap to about 4.2%. Water stayed at about 0%.


6782

6783

Took about an hour to get to 6.3 %.

6784

That is when we started boiling because my water bucket was full and I wanted to use the Water to back flush the membranes.

[not_for_sale]

While my helper was boiling I back washed the membranes until the sugar concentration was back down to almost zilch.

6785
6786
6787

Finished product tastes pretty good and is very light - clocks in at 68%

6788

6789

[not_for_sale]

Here are pictures of the unit outside of the cooler it's going to go in.

6790
6791
6792
6793
6794

Sorry for the mad scientist setup but it seems to work pretty well. Backwash took about 2 gallons of RO water and about 20 minutes of time for the predicted to not supply sugar concentrate to the membranes anymore.

Pump gets the required 88 psi without having to supply anything with pressure with three membranes in series.

I also have a UV light unit that is not in this setup test yet.

[syrup2make]

This is an RO? Can you provide part info so that I could make one?

[not_for_sale]

Making an RO is easy. All you need is a pump, a pre-filter and housing and a membrane and housing as well as a couple of fittings and tubing. You need to flow restrict the concentrate line, you can do that with a needle valve or a flow restrictor available on wateranywhere.com.

I used three membranes in series. I haven't tried if one single membrane would be sufficient.

This is an aquatec 8852 pressure pump (eBay ), three 150 GPD membranes in series and a 1 micron prefilter.

The housing are less than 10 bucks each, the membrane is less than $30 each, and the prefilters are cheap too.


Pretty good price for these components on Amazon. Amazon has been one of my favorite supplies for this stuff.

Pumps:

1. http://www.amazon.com/gp/product/B0099TTV76?ie=UTF8&camp=1789&creativeASIN=B0099TTV76&linkCode=xm2&tag=mapletrader-20

2. http://www.amazon.com/gp/product/B000LDG4L0?ie=UTF8&camp=1789&creativeASIN=B000LDG4L0&linkCode=xm2&tag=mapletrader-20

3. If you get teh aquatic, make sure you get the 8852 - some sellers sell an 8800 pump and then its likely the low flow pump. My Aquatec can handle 3 150 GPD membranes in parallel with cold sap easily. this Amazon price on it is only $79 right now - pretty inexpensive: http://www.amazon.com/gp/product/B00A7ZV2GO?ie=UTF8&camp=1789&creativeASIN=B00A7ZV2GO&linkCode=xm2&tag=mapletrader-20

Membranes:

1. http://www.amazon.com/gp/product/B000WPHUAG?ie=UTF8&camp=1789&creativeASIN=B000WPHUAG&linkCode=xm2&tag=mapletrader-20

2. http://www.amazon.com/gp/product/B004BG5ETG?ie=UTF8&camp=1789&creativeASIN=B004BG5ETG&linkCode=xm2&tag=mapletrader-20


The stainless housing for the XLE 4040 is actually cheapest on ebay. There is a seller selling them for $80. Amazon sometimes has the housings for the residential membranes at $7.00 plus shipping.

[CTsugarMan]

What would you use this for?

[hodorskib]

I built one last year using a needle valve for back pressure and was bringing 1.5% sap to 6-7% with 3 100gpd membranes. I could only process around 70 gallons in 24 hours though. Here is a link to my website that has more information and detailed schematic drawing. Your setup looks good, I mounted mine to a piece of 2 x 4 plywood so it was easier to follow and keep track of things.

https://sites.google.com/site/mattatuckmadnessmaplesyrup/

[not_for_sale]

I like your write up. However there are some things that I need to understand.

Why check valves? If check valves why not only one at the concentrate tank?

Why three membranes? Just for good measure? I plan to test if one membrane will do the job or not? So why not two or four?

Why three membranes in series, and not two parallel and the third one in series to the two ? That way the third membrane would see some flow that is higher?

Why switch the membranes around? Just a guess or some deeper insight?

Why concentrate in one pass? Higher flow will better for the membranes! Less chance of fouling because of bacteria. Just move the concentrate back to your sap tank?

Why does your filter plug up at 150 gallons? It won't filter sugar at 5 microns. Just dirt. Do you have that much dirt in your sap?

[hodorskib]

Each year I found myself turning my hobby into a second full time job. I wanted to find a way to spend more time enjoying making syrup instead of killing myself boiling until all hours. I live in a residential neighborhood and rent my tree from the local state forest. I do not have the space to build a sugar house or for a larger evaporator. What I have is a 2x3 flat pan with 3 partitions. I have it on wheels so it lets me move it into the garage for storage. With that said I spent a considerable amount of time on this site and talking with several people who deal with residential RO systems.

I like your write up. However there are some things that I need to understand.

Why check valves? If check valves why not only one at the concentrate tank? I use check valves after each membrane as protection. It is most likely overkill but I found them for 8 bucks each and they are a one time purchase so I went ahead and put them in. I also have one on the permeate line since I have my setup plumbed directly into my waste line for the house and don't want anything from that line coming back in. One advantage of having city utilities.

Why three membranes? Just for good measure? I plan to test if one membrane will do the job or not? So why not two or four? When looking at the performance of a 100gpd membrane it will produce 1 gallon of pure water for every 2 gallons of waste. So basically it will remove 1/3 of the water. So my thinking was it takes me 60 gallons of sap to make 1 gallon of syrup (my trees run around 1.5%) So 60 gallons goes into the first membrane (20 gallons goes to waste and 40 goes to the next membrane), then 40 goes in and 13 goes to waste and 27 goes to the next one, and then finally 9 goes to waste and 18 goes to the concentrate tank. I added a needle valve at the end and was able to add additional back pressure to actually take it down to around 12 gallons for each gallon produced. Or brought 1.5% sap up to 7% in one pass. I have actually added a 4th this year so I can increase the flow rate while maintaining the 7% or try for a higher concentrate.

Why three membranes in series, and not two parallel and the third one in series to the two ? That way the third membrane would see some flow that is higher? I really don't know but when designing the system this is the way it made more sense to me. However, I am not sure if other ways would work better - I got the results I was hoping for so I never thought about moving things around. The season is too short and I really don't want to take the chance of having something go wrong.

Why switch the membranes around? Just a guess or some deeper insight?
I move the membranes because the last one will be getting beat up more because of the higher concentrate - at least that was my thinking.

Why concentrate in one pass? Higher flow will better for the membranes! Less chance of fouling because of bacteria. Just move the concentrate back to your sap tank? This is something I have been giving some thought too. I normally collect around 5 or 6 at night and then put it in the system and start it up. I work all day and cannot tend to the system so it needs to run by itself. I have been in contact with John who originally posted his system a while ago and he uses a concentrate tank and runs his concentrate line to the bottom of the tank and pulls his feed from the top of the tank. This makes sense and I have been toying around with the idea as well. However, these are 100gpd membranes with 60 degree water. Running them in one pass gives me 72 gallons in 24 hours I'm not sure how much more you will be able to do trying to make multiple passes?

Why does your filter plug up at 150 gallons? It won't filter sugar at 5 microns. Just dirt. Do you have that much dirt in your sap? It doesn't plug up but since the sap is moving so slowly through it I noticed it gets funky collecting any bacteria or other sediments. I make around 15 gallons per year so 7 or 8 filters is well worth it to change out regularly.

Hope this was helpful?

[HyeOnMaple]

Hodorskib,

Nifty, I don't understand it all, but I don't understand lots of stuff that I still benefit from. This will be next years project. This year I'm still working on Franken-Arch.

Thanx for the web-site and schematic. Right now I'm wondering if I have enuf sap to run my four pans as they are. Is there a ruberic as to how many taps one should have per sq ft of evaporator. I have just less than 12 sq ft.

[hodorskib]

Every season is different! I always went by the rule of each tap should yield 1 quart of syrup. But mother nature has never given that exact ratio. My best year was 1 gallon for every 3 taps and my worst was 2 years ago when it was 8 taps per gallon. Basically if it is flowing your boiling and trying to figure out how you can do it more efficiently, if it is not your trying to figure why it isn't. Keep it simple and have fun. Growing up if someone told me I could play with fire till all hours of the night I would have jumped at the chance. Sometimes the syrup is just a bonus - the hot sticky bug infested summer will be here soon enough enjoy this time of year.

Good luck

[briansickler]

What would you use this for?

For concentrating maple sap....What else is there? This is just a small ro, operating in the same manner as our much larger versions. I think this would really benefit the micro producers out there, as the original post demonstrates. I like it. Anyone could use this basic plan to make a much larger ro...

Brian

[not_for_sale]

It's way more cost effective to increase your evaporation rate this way than to buy a bigger pan. Also the evaporation rate is characterized by a lot less work.

For example- if you get 5 GPH with the small RO and have a small setup that gives you 4-5 GPH (Half Pint) by boiling.

Lets say you start with 100 Gallons of SAP that you need to boil. With the half pint you'll be there for 20 hours firing and watching.

With the RO you'd collect - dump it in your storage tank and let the RO reduce the 100 gallons to 25 gallons at 8% - without watching the thing and without having to fire anything. Then you'd fire up your half pint and boil for 5 hours for 3 gallons of syrup.

Lot less work, lot more fun.

[not_for_sale]

I got a private message asking some questions and I thought I would also post the answers here:

The reason I built the RO is because I want it to be able to run unattended. The Evaporator always requires you to sit in the vicinity. You turn it on and walk away and its doing its work.

The pump (I bought the Aquatec 8852 on ebay - higher flow that the 8851) because its a membrane pump that can run dry. Its not going to hurt the pump. However, if you set it up right its never going to run dry.

The way I have set it up is that I have a supply line from the barrel to the pump - and run the concentrate back to the barrel. This way you start with 2 % sap, and the concentrate is lets say 4.5 % and it removes about half the water in the first pass. The efficiency - because its such small system goes down as the sugar content in the barrel increases, so it removes less and less water with every pass but the pump will never run dry. So after the first pass the barrel is only half empty, after the second pass its getting lower but it never runs empty.

Regarding the needle valve: Aquatec sells an "AutoFlush Valve" that runs the concentrate line at full speed for about 30 seconds every hour. That increases the efficiency of the membranes. I did not buy that one - but I should have. if you run it on a continuous basis its the better way to go - that way it really is an unattended set up. If you use a needle valve, you just adjust the flow. The lower the flow is - the bigger pressure you have at the membrane and the higher the rate of water removal. However, the efficiency goes down and you have to manually flush. But - you may get to 6% concentrate in one run through. The sugar content of the concentrate goes up the slower it runs.

As far as backwashing - the filter and the membranes are full of perfectly good highly concentrated maple sap. There is about a gallon of that stuff in the lines. You save the water the RO produces and run that back though it and collect the first gallon of concentrate. After about a gallon of backwash mine drops down to 2% sugar.

If you don't continue backwashing the pre-filter is full of sap level sugar - that could spoil and as such you would need to change it for every run. I run the RO water through it long enough so that the sugar content is basically zero. Yesterday that was 4 gallons. But - you also do that while you are boiling and its just a few lines out of one barrel in the water barrel (or a couple of valves). Don't run city water through them as that has chlorine and membranes do not like chlorine. The membranes if you backwash them can last years. The pre-filter is 3 bucks, the membranes are 26 bucks each on wateranywhere.com so its not that big of an expense if you foul a even a membrane.

Now - if you want to get a faster working system - its not that much more money. My system cost me about $300 (Pump and Power Supply $110 with shipping, Membranes, housings, filters about $200 with shipping) and I also bought a small UV system that goes in before the membranes that is not in the pictures ($100). wateranywhere.com sells a 100 gph 99 PSI procon pump with motor and clamp for $220 - a 200 PSI housing and a 4x40 inch 2400 GPD Low Energy membrane for $300 so you could conceivable built a 40-50 gph RO for under $600.

[Snowy Pass Maple]

I agree with a lot of the points made above - a small RO is a fantastic way to manage a lot more taps and the diaphragm pumps are a great way to go. Always collect the initial sweet rinse. And recirculating in 55 gal drums is great - cool to make syrup in your sleep, or at the day job. I think even 50 tap hobbyists should be seriously considering some sort of RO before any kind of evaporator upgrade... big stainless steel pans and evaporators are $$$!

The main advice I'd offer is as follows... I think it's far better in the long run to run faster flowrates and potentially recirculate concentrate back to the input tank vs. try to achieve a really high sugar in a single pass at higher pressure/lower flow. As one of the posts says, you can definitely impress yourself with high sugar numbers, but that is asking for trouble. Just let it work for a few hours on a 55 gallon drum ahead of time vs. feel that you have to do it all in a steady state flow with the evaporator. And even getting to 5% is an amazing gain compared to 2% - you can more than double your potential production!

I would be concerned about planning a system to run at a rate of 1 part permeate for 2 parts concentrate on each membrane as discussed earlier - the standard flow restrictors are designed to maintain a 4:1 concentrate:permeate ratio. Many of the membrane spec sheets even recommend 5:1. This means you have to plan a bigger pump providing higher flowrates.

I also found that the residential membranes didn't hold up very well - despite regular permeate washes, they lost way too much of their performance after a few hundred gallons of sap being processed from ~2 to 5% in ~30-50 gallon batches, typically running 6-10 hours in any given run. The 88xx pump IMHO is very marginal for this application - particularly by the time you start putting a lot of membranes in series, pulling more and more permeate flow.

Since then, I have been testing a higher flowrate pump (about 2X the 8852 flowrate) with a single 21" nanofiltration membrane, regulating the pressure to achieve a 4:1 concentrate:permeate ratio. That has held up much better so far. It will get much more thoroughly tested this season as we have moved up to well over 100 maple taps, with a lot of birch to follow. I don't want to declare success with the design until it's processed a few thousand gallons of sap, but I can already say it's significantly better than the residential membranes. Also, the reason for 21" and not larger in this case was I want to be able to run in a sugarhouse without power... this system uses about 50W of power when running, so it can go for quite a while on deep cycle batteries.

Also, regarding the following:



wateranywhere.com sells a 100 gph 99 PSI procon pump with motor and clamp for $220 - a 200 PSI housing and a 4x40 inch 2400 GPD Low Energy membrane for $300 so you could conceivable built a 40-50 gph RO for under $600.

I think you would need a much higher flow pump to support this membrane. The 100 GPH pump = 2400 GPD - but that total pump output would only cover the rated permeate alone on this membrane, and not the additional 10,000+ GPD of concentrate you would ideally run across it to get that rated permeate. I think when you add in valves, gauges, flowmeters, etc. you'd be close to $1000 in parts - but still well below any commercial unit.

And I agree with the point that there is an economy of scale to go to higher throughputs/larger membranes... but I also think there is a lot of value to a system in the $500-$1000 range that can double the steady state production of a 2x4. (and even do it on batteries!) That would be assuming all new parts including flowmeters, SS vessels, filters, ample valving for flexible routings, etc. And when you consider getting even more output from unattended recirculation time (taking advantage of time that you can't boil), you are potentially tripling the output of that same 2x4 hobby setup if you can manage more sap, and boil 5% concentrate.

[saekeaton64]

Snowy, could you post a few pictures of the ro system that you have built?

[Snowy Pass Maple]

Snowy, could you post a few pictures of the ro system that you have built?

I will in a few days I hope - it's in a Frankenstein state at the moment as I'm in the middle of modifying it to add a second pump/chamber to give me more capacity and flexibility in how I process with it - but once I get it all re-mounted, I will post pictures.

[not_for_sale]

I think you would need a much higher flow pump to support this membrane. The 100 GPH pump = 2400 GPD - but that total pump output would only cover the rated permeate alone on this membrane, and not the additional 10,000+ GPD of concentrate you would ideally run across it to get that rated permeate. I think when you add in valves, gauges, flowmeters, etc. you'd be close to $1000 in parts - but still well below any commercial unit.



The pump is slightly undersized for that membrane - but not by as much as your statement may make it seem. The 2400 GPD membrane is rated at 77F - running it at Sap temperature of 40 degrees will reduce the permeate flow to 35 gallons per hour (840 GPD) at 100 PSI. The specs also allow up to 300 PSI on this 100 PSI membrane as long as you don't exceed 100 GPH permeate flow. The concentrate flow rate can be achieved by using a recirculation pump - which is what I planed on doing with my Aquatec 8852 once I upgrade the membranes.

[Snowy Pass Maple]

The pump is slightly undersized for that membrane - but not by as much as your statement may make it seem. The 2400 GPD membrane is rated at 77F - running it at Sap temperature of 40 degrees will reduce the permeate flow to 35 gallons per hour (840 GPD) at 100 PSI. The specs also allow up to 300 PSI on this 100 PSI membrane as long as you don't exceed 100 GPH permeate flow. The concentrate flow rate can be achieved by using a recirculation pump - which is what I planed on doing with my Aquatec 8852 once I upgrade the membranes.

You could probably also make the argument that the sugar will also slow the permeate even more as long as you're on maple sap and not doing birch... I would just be hesitant to design without a bit more flowrate capacity particularly for maintaining good high flow rinsing cycles. One size down on the membrane may be a much more robust system with that pump. As you say, recirculation can help - but then you're over $600 on just two pumps, membrane and housing - then you have filters, valves, flowmeters, tubing, etc... and all that's assuming you're staying at relatively low pressure. Over 150 psi, then you're into much more expensive plumbing. I would guess $1000 is a probably a minimum budget to get into something that size just in parts.

I also happen to have a spare 8800 series pump that was planned for a kitchen RO - but I've toyed with the idea of using it as a recirculation pump. My concern was that even at zero net pressure open flow, I don't think the 8800 series adds a whole lot of flowrate - I think it's around 700 GPD? On a smaller membrane, that would start to become significant, but a 4"x40 is a lot of membrane. You could also look at running a bigger Aquatec - they have some that go up to 5GPM - although the ones that large may not be rated for continuous duty.

Overall, I can't recall seeing any < 150 psi diaphragm pump systems with recirculation done on here yet... could be a fun test. The nice thing staying in that pressure zone is that you can use relatively inexpensive valves and fittings so you can then afford to setup a lot of operating flexibility to route things in and out of tanks, bypass or select subset of vessels, etc. most any way you please!

[not_for_sale]

There is a Flowjet Pump (12 V for RV ) that does 5 GPM at 60 PSI and there is a pro con pump that can do about 350 GPH at 250 PSI. I just know myself. I will tinker until I know what works.

I'd rather upgrade the pump than downgrade the membrane because lower GPH membranes cost essentially the same.

Never the less, if you are building a home made RO with residential membranes there isn't much you can do wrong at $ 26 a membrane. Removing even half the water while you sleep is a huge improvement.

[Snowy Pass Maple]

The pump is slightly undersized for that membrane - but not by as much as your statement may make it seem. The 2400 GPD membrane is rated at 77F - running it at Sap temperature of 40 degrees will reduce the permeate flow to 35 gallons per hour (840 GPD) at 100 PSI. The specs also allow up to 300 PSI on this 100 PSI membrane as long as you don't exceed 100 GPH permeate flow. The concentrate flow rate can be achieved by using a recirculation pump - which is what I planed on doing with my Aquatec 8852 once I upgrade the membranes.

One other follow-up to this discussion - while most membrane spec sheets just lay out a permeate flow and minimum concentrate multiplier, there are minimum flowrate specs that must also be adhered to.

The argument above that permeate is derated in a maple sap application, taken to an extreme, can easily illustrate this - if you have a huge membrane yielding very little permeate with ice cold highly concentrated sap, simply running 5x that greatly reduced rate will not maintain good flow velocity on the membrane, thus fouling results

More specifically, tech sheets can help bound this -

http://www.dow.com/PublishedLiterature/dh_0036/0901b803800362e3.pdf

For a 4040, looks like you'd want to be running at least 3 GPM no matter what the permeate flow is, or 4320 GPD. And at that condition, you could at most support taking 1/5 of this as permeate, or 864 GPD. In total, you're looking at 5184 GPD of flow requirement - thus the proposed 2400 GPD pump is pretty grossly undersized. Additionally, an aquatec 8800 series pump in recirculation won't even come close to meeting this defecit in flow requirements.

Not sure how the residential units are rated for minimum flowrate, but would be good to know.

Point of my feedback on these discussions is to not underestimate the need to keep things moving really fast across these membranes. Yes you can go to even bigger procons and high pressure systems - and then you'll be building a normal commercial RO for a lot more money. And there is an economy of scale, but I think many people on here are aiming to stay in the sub-$1000 range. This is much easier with a design that operates under 150 psi, with lower cost diaphragm pumps, plastic line/fittings and a much smaller membrane than a 4040.

As you say later, residential membranes are cheap and I'm totally with you that even getting from 2-4% in recirculation is awesome. No need to shoot for the moon to make a huge difference and keep it running well. But it's easy to burn them up if run too slowly and $26 x multiple membranes can add up. My personal experience is the slightly larger 2.5" single membranes seem to be more robust with the incremental cost being the commercial housings so that's the path I've taken. And yes, they are not as cost effective per gallon as a big commercial unit with huge pumps/membranes - but they are still very cost effective for small producers and can keep you under $1000.

[saekeaton64]

I am interested in building an RO system that will easily handle the increased number of taps that I plan to add in future years, while keeping the cost as low as possible. (Asking a lot I know).

I would like to bring my sap to the 6-8% range, re-circulating through a bulk tank if necessary.

I have read through the various posts on the trader about homemade RO systems, and found the information and vast knowledge of the members to be very impressive.

My proposed parts list:
2.5” x 40” Extra Low Energy Membrane 750gpd.$152.00
2.5” x 40” Membrane Housing (PVC).................$98.00
240GPH w/ 75psi relief Procon Pump.................$162.00
Pump adaptor frame.......................................$36.00
10” 5 micron filter..........................................$6 .00
10” Filter Housing........................................... $8.00
Wateranywhere.com................................. ......Sub-total $462.00

Love-joy coupler........................................... .$19.99
2- Pressure Gauges 0-400PSI..........................$25.98
Northerntool.com.................................. .........Sub-total $45.97

1/2HP 1725rmp C56 Motor..............................$127.95
Surpluscenter.com................................. .........Sub-total $127.95

Total cost minus tubing and fittings..................$635.92

I assume that the total cost would be around $700.

Are there any changes that you think I should do differently? Based on the equipment (theoretically) what would be the GPH rate for a system like this?

[not_for_sale]

This sounds like a great system.

He membrane is rated at 77 F, specs point to the fact that at 41 F you have an adjustment factor of 2.6. The XLE commercial membranes are rated at 100 PSI and can take somewhat more than that. If you run them between 100-150 you will have a bit more permeate I don't know how much though.

So with the above factor in mind the membrane could remove 12 gallons per hour of water from the sap in one pass. You'd have to run 72 GPH through it to stay within a 5:1 ratio. Of course if you preheat the sap before you run it though it you'll get the rated flow with the suggested pump.

[Snowy Pass Maple]

I am interested in building an RO system that will easily handle the increased number of taps that I plan to add in future years, while keeping the cost as low as possible. (Asking a lot I know).

I would like to bring my sap to the 6-8% range, re-circulating through a bulk tank if necessary.

I have read through the various posts on the trader about homemade RO systems, and found the information and vast knowledge of the members to be very impressive.

My proposed parts list:
2.5” x 40” Extra Low Energy Membrane 750gpd.$152.00
2.5” x 40” Membrane Housing (PVC).................$98.00
240GPH w/ 75psi relief Procon Pump.................$162.00
Pump adaptor frame.......................................$36.00
10” 5 micron filter..........................................$6 .00
10” Filter Housing........................................... $8.00
Wateranywhere.com................................. ......Sub-total $462.00

Love-joy coupler........................................... .$19.99
2- Pressure Gauges 0-400PSI..........................$25.98
Northerntool.com.................................. .........Sub-total $45.97

1/2HP 1725rmp C56 Motor..............................$127.95
Surpluscenter.com................................. .........Sub-total $127.95

Total cost minus tubing and fittings..................$635.92

I assume that the total cost would be around $700.

Are there any changes that you think I should do differently? Based on the equipment (theoretically) what would be the GPH rate for a system like this?

While I haven't worked with the Procons, a few quick thoughts...

How does their 75 psi relief work - does it max out at that pressure? I think you'd want to be able to go to at least 100. And of course you don't want to let it get too far beyond if using plastic tubes/fittings.

Also, this looks like a brass pump? Keep in mind that you may want to run NaOH washes - Brass should be OK to reasonably high concentrations, but if it was me, I'd go stainless. I'd do the same on the vessel for the small cost increase just so it looks cool :-)

Also, these cannot run dry so you probably want some sort of shutdown protection to guard against that assuming you'll run it unattended.

Once putting this kind of money in, I would definitely opt for flowmeters on conc and permeate lines - this will really help you monitor the ongoing performance of the system, logging the performance regularly - if you read some commercial maple RO manuals, they talk about this process. If you stay under 150 psi, the meters shouldn't exceed $100. Note that many are only rated to 100 psi, and most with integrated valves are only rated to 100; some allow 150 as long as temp isn't too high (but keep wash temps in mind). I would add a good quality needle valve as well to regulate the concentrate flowrate/pressure - leave permeate to flow openly and freely.

Plan on a LOT of filters - change them frequently. May be more economical to go with 5" filters - they tend to get funky before you ever reach a point of pressure drop.

The valves and fittings can add up really fast if you want to give yourself flexibility to not be shuffling inlet tubes all the time. I have mine setup so I can use valves to select drawing in sap, or RO water for rinse, or extra free tube to bucket with hot water or wash solution. I can also route concentrate line to either source feed tank, or outside waste when rinsing. I'm going to also add a setup with a float on my preheater so I can flow concentrate output directly to preheater until full, and then have a bypass back to the feed tank. I'm sure I spent over $100 on valves/fittings/tubing/mounting brackets/etc... that stuff adds up. Still, seems like you can get to a really nice system for $1000.

From what I can tell, it seems like you may have lots of flowrate margin with room to expand should you want to... in fact, you're pushing more towards max flow rate spec which gives you a lot of room to work.

I have a question for all - I know very little about the procon motors/pumps and am wondering for a motor like this, is the power draw continuously ~1000W or is that only peak under max surge/startup? How many watts would this draw in normal operation at target pump pressure? I don't have electrical service so I have been aiming to run on more like 50-100W of power so I can run off deep cycles when away, and charge them and run off efficient inverter generator when working in the shack.

[saekeaton64]

I checked the spec. sheet on the Procon pump. The relief valve is adjustable from 60-99 psi, the factory
preset is at 75 psi. So it could easily be adjusted to a higher setting.
https://store-c59cb.mybigcommerce.com/content/Series%204%20Pump_spec.pdf

The spec. sheet also stated that the brake hp for the 240 gph pump at 100 psi was .37, a 1/2hp would be the min. motor to run this size pump without overloading it. The NEMA faceplate rating on the motor is 115v at 8.8 amps, so the power draw would be 1012 watts at the rated 1725 rpms under a continous load. (You would need a lot of amphours to run a pump like this off grid. =$$$$)

After doing a little searching on ebay I found rotameters(flow meters) rated for 0-5 gpm listed for around $20.00. I done a search for sediment filters an found a listing for 50- 10" 5 micron filters for around $65.00 + free shipping. I also found 1/2" full port stainless ball valves for really cheap.

The applied membranes website has an overstocked items section with a really good list of membranes and procon pumps well under the wateranywhere.com price.

If someone was to purchase all of their item for a homemade RO through a retail website, I could see how the cost would be easy to hit the $1000 + mark, but if being thrifty and checking through various sites, overstock sales, and internet auctions one could keep the cost around $700 for a complete system with room for expansion.

[not_for_sale]

Snowy,

I checked the Dow documents. I think the procon pump is plenty powerful to run that 40 inch 2400 gpd membrane at 100 psi if I use the Aquatec as a recirc pump. The aquatec does 2-3 GPM at a 15 psi pressure drop. The membrane only needs that concentrate flow according to that PDF. Since I will also have feed flow at the rated 1.6 GPM from the procon pump I'll be above the minimum concentrate flow.

Permeate flow is not only rated 2400 GPD at 77 F and less that 1000 at 40 F and even less at 33 F, but that rating pertains to a permeate feed. Anything other than a permeate feed reduces the rated GPD significantly. Lets say its 480 at 33 and 960 at 50F.

That will give me plenty of flow at any time it makes sense to recirc through the sap tank. Even if the temperature goes up to rated I still have room for addtl. concentrate flow.

Another issue is concentrate Brix. It's going to depend on pressure across the membrane. 100 psi is likely not enough to go beyond 8 %. The 80 psi I got with the aquatec got me to just under 7 %. Even with running it for 24 hours on the tank. The osmotic pressure created at the membrane surface needs to be overcome. You can do that with feed pressure or by playing with vacuum on the permeate line. The difference between a residential and commercial membrane is simple: squarefootage of membrane surface. Otherwise no significant difference. So recirculating is going to work on either to reduce fouling.

My first step is going to be to set this up with my three residential 150 GPD membranes in parallel, the procon pump and the aquatec recirculating. I will then put a small releaser and high vaccum on the permeate side. Lets see what happens. I can tell you in a couple of days what the procon uses regarding power. I got a meter on the pumps.

[not_for_sale]

The procon pump, with the 1/3 hp motor uses 169-175 watts when idling.

[Snowy Pass Maple]

Snowy,

I checked the Dow documents. I think the procon pump is plenty powerful to run that 40 inch 2400 gpd membrane at 100 psi if I use the Aquatec as a recirc pump. The aquatec does 2-3 GPM at a 15 psi pressure drop. The membrane only needs that concentrate flow according to that PDF. Since I will also have feed flow at the rated 1.6 GPM from the procon pump I'll be above the minimum concentrate flow.

Permeate flow is not only rated 2400 GPD at 77 F and less that 1000 at 40 F and even less at 33 F, but that rating pertains to a permeate feed. Anything other than a permeate feed reduces the rated GPD significantly. Lets say its 480 at 33 and 960 at 50F.

That will give me plenty of flow at any time it makes sense to recirc through the sap tank. Even if the temperature goes up to rated I still have room for addtl. concentrate flow.

Another issue is concentrate Brix. It's going to depend on pressure across the membrane. 100 psi is likely not enough to go beyond 8 %. The 80 psi I got with the aquatec got me to just under 7 %. Even with running it for 24 hours on the tank. The osmotic pressure created at the membrane surface needs to be overcome. You can do that with feed pressure or by playing with vacuum on the permeate line. The difference between a residential and commercial membrane is simple: squarefootage of membrane surface. Otherwise no significant difference. So recirculating is going to work on either to reduce fouling.

My first step is going to be to set this up with my three residential 150 GPD membranes in parallel, the procon pump and the aquatec recirculating. I will then put a small releaser and high vaccum on the permeate side. Lets see what happens. I can tell you in a couple of days what the procon uses regarding power. I got a meter on the pumps.

I thought you were using a Aquatec 8852? I ask because they are rated in 2-3 liters per minute in that range - not gallons...

http://www.aquatec.com/sg_userfiles/Booster_Pumps_Data.pdf

All the aquatec pumps I've seen also have a maximum inlet pressure of 60 psi which would seem to be an issue. I'm still very interested in the idea of an aquatec recirculator but haven't dug into it much to see if they have a way to work around this.

I agree that you'll never be limited by the concentrate vs. permeate ratio - only concern would be keeping the concentrate flow up.

Also, if you're putting that procon on the small membranes, keep in mind that the membranes also have an upper flowrate limit - for residential, I believe it's around 2 GPM, so you're close, but OK.

Very interested in how the procon wattage looks vs. flow/psi... long-term, I could envision doing something like using an aquatec to run at lower pressure for recirculation when away or asleep (using the batteries) while once I crank up the inverter generator running while working in the sugarhouse, which also charges the batteries, I could consider using a small procon and maybe additional membrane in series to aim for higher brix in a continuous mode feeding the evaporator. Or in a real crunch, run the procon off the generator for an extended run. But I'm really interested in what the power requirements would be to see if it makes sense.

Another member is using a copper preheater to go up ~25 F to double permeate throughput on their RO (only when feeding direct to evaporator) - something I have thought about and not tried, but will probably do now that I'm setup in a sugarhouse. They have not had microbe issues doing this, using just long water flushes between runs. When you put all this together, you could potentially quadruple output when actively running generator and evaporator, which could justify a decent amount of increased generator load when it's probably running anyway, while being able to isolate a subset of the system on the aquatec would still let me churn through a couple 55 gal drums each day when not at the sugarhouse. Much beyond this, I'm sure it starts to make more sense to just go to a gas powered RO - I don't know what kind of fuel they go through vs. an inverter generator, but I'd guess they are sized in a way that requires a lot more sap to make it worthwhile than one can do on a couple hundred taps.

[not_for_sale]

All the aquatec pumps I've seen also have a maximum inlet pressure of 60 psi which would seem to be an issue. I'm still very interested in the idea of an aquatec recirculator but haven't dug into it much to see if they have a way to work around this.

I am using the Aquatec as a recirculation pump right now - I realize that if I go to the pressures that I want to go to, its not going to work.

The pro con, as well as the Aquatec are differential pumps, meaning if you supply an Aquatec 8852 with an Aquatec 8852 you are going to 160 PSI at 1.4 LPM. I am just going to see what happens. You have a 10-15 PSI pressure drop across the membranes, so if the procon hits the membranes at 99 PSI, the Aquatech 8852 will recirculate at essentially 10 PSI differential pressure, but with a 90 PSI inlet pressure.

I have tried the temperature rise with the residential membranes - and it works - but only for raw sap. Anything that is concentrated in the 4% or above range the permeate flow essentially stops or gets sloooow no matter the temperature, unless you pressurize .

Here is what I have learned while playing around with the Residential Membranes and the Aquatecs:

1. Temperature of SAP is important for water removal speed only below 4% sugar content, above 4% sugar content water removal performance depends largely on pressure.

2. Running Membranes in parallel as long as you have a pump that can do the pressure will multiply your water removal flow because you have more surface area to work with.

3. Running two membranes in series only makes sense if you are processing raw sap, and adding a third membrane yields no improvement in either flow increase or sugar concentration.

4. An Auto Flush is essential because the membranes foul quickly (less than 2 hours with raw sap) . Fouling is not bad - its just what happens. Flushing the membranes removes the "stuck" molecules from the surface of the membrane. Aquatecs Autoflush flush for 20 seconds every hour and are absolutely necessary for decent performance over time with a residential membrane.

5. Aquatec supplies 80 PSI - Good enough for 6.7 to somewhere below 8% of brix. In an experiment I mixed a 10% sugar solution, and flooded the membranes with permeate on the permeate side (bucket and made sure the permeate line stayed full) After an hour of running my 10% sugar solution through the system it was below 9% - so I actually added water to the sugar and did not remove it. The supplied pressure of 80 PSI is not enough to prevent the water in a natural osmosis to get back into the sugar solution.

6. The higher the pressure the better the one-pass brix in the concentrate as long as you keep the membranes from fouling (recirculation or timed AutoFlush)

7. A XLE 4040 membrane has 85+ sqft of area, while a residential 150 GPD has less than 5 soft, meaning it would take 18 residential membranes in parallel to do the work of one XLE 4040 (pressure being equal).

[bowhunter]

As has been stated above you have to be able to overcome the osmotic pressure of sucrose in water with pump pressure to get reverse osmosis. The osmotic pressure is dependent upon the sugar concentration and the temperature of the solution. Your experiment falls pretty much in line with the calculations for osmotic pressure. I calculated the osmotic pressure at several concentrations so you can see where you hit the wall. I did all of the calculations at 40 degree F. It isn't too temperature dependent.

wt% sucrose Osmotic Pressure Delta Pressure @ 90PSI % permeate flow

1.5 ----------------14.7 PSI---------------------75.3-----------------------100%
3.0 ----------------29.4--------------------------60.6------------------------90%
5.0 ----------------49.0--------------------------41.0------------------------74%
6.0 ----------------58.8--------------------------31.2 -----------------------64%
7.0 ----------------68.6--------------------------21.4 -----------------------53%
8.0 ----------------78.4--------------------------11.6 -----------------------39%
9.0 ----------------88.2---------------------------1.8 -----------------------15%

This tells you 8-9% is the practical limit for a unit capable of 90 PSI. The driving force is the difference between the pressure on the membrane and the osmotic pressure, so putting elements in series vs. recirculation back to the sap storage tank is probably not the best configuration. In the fourth column I try to estimate the flow of permeate relative to the maximum you would achieve at 1.5% sugar. I believe the flow increases as the square root of the ratio of differential pressure through the membrane, but I'm not absolutely certain so take that column of numbers with a grain of salt.

[hodorskib]

Thank you for this information. I have a small RO that I built last year. I have 4 - 100 gpd membranes in series. I have a needle valve to control the back pressure and the concentrate levels. The best I have managed was 9% but that reduced the flow down to less than 2 gallons per hour. After playing around with the valve the best concentrate vs flow has been between 6-6.5% while processing around 4 gallons per hour with 40 degree sap. Not too bad considering I'm starting with 1.75% sap. So after a season and a half my experimentations fit with your calculations.

[bowhunter]

I'm amazed at how close the calculations came to your experience and Colin was able to identify the upper limit exactly with his 10% sugar experiment. After seeing your system I'm planning to build a small unit next year. But I'm still trying to decide on the type membrane. If I remember you're using the regular home (Dow TF?) membrane, right? How are they doing with regards to fouling? I've gotten recommendations from a membrane supplier to go with an Axeon NF4 series which is 2 1/2 times more expensive than the ones you're using, and Colin (snowy pass maple) is using an Axeon NF3 with good results so far. The NF4 gets 80-90% mineral removal vs. about 50% for the NF3. I think using a membrane that leaves everything but the water with the sugar is best for maintaining syrup properties consistent with non-ro syrup, but that 's just my theory.

[hodorskib]

I am using the cheapest 100gpd membranes I found on Amazon.com for 27 dollars each. This is the second season with them. I flushed the whole system with a 5% solution of peroxide and water and then rinsed with 10 gallons of water. I wrapped the membranes in plastic wrap and placed them in the bottom of the fridge. I did the same thing at the start of this season and everything is good. I can run about 40-50 gallons at 5.5-6.5% concentrate before I have to flush with luke warm water (about 5-10 gallons) then the membranes perform back to normal. Anything beyond that and they foul and both gallons per hour and concentrate go down. I also flush the system with the peroxide solution and change the filter after about every 150-200 gallons processed. Yesterday I processed 85 gallons to 5.5% in 24 hours. This is perfect for me with 75 taps. Thanks again for the numbers they were right on.

[bowhunter]

Awesome work! I was wondering about peroxide because it such a good oxidizer, but I don't remember anyone mentioning it in the literature. I think your work definitely indicates that the fouling is probably just microbial growth similar to the stuff that plugs off the taps eventually. I have the about the same set up as you with 70-80 taps and a 2 x 3 pan so I have a similar need. I was going to take your advice and use the Aquatech Auto Flush valve to try and slow down the rate of fouling.

[not_for_sale]

Fouling can be prevented or significantly slowed down by recirculation. What you do is take a second pump that takes the concentrate from the concentrate outlet of the membranes and circulate it back to the inlet. This is behind the pressure pump. So the casing of the recirculation pump needs to be able to take the pressure. The aquatech takes the 145 psi. Or you can take the chugger stainless pump for this.

Fouling is the accumulation of organic molecules on the surface of the membrane. All you need to do is regularly flush them away. If you don't do that, the surface is just not effective anymore. The speedier the flow, the less chance the molecules have to settle on the surface.

As far as membranes go: residential vs other membranes is just a function of surface area. The larger the membrane the larger the area is where your pressure pushes water through. The second type of membrane is a Nanofiltration membrane. They are not true RO membranes but they do filtration on the molecular level and are thus faster and more efficient at the same pressure. The nano membranes do pass some minerals though, so it will change the taste somewhat.

The number or type of membranes you use depends largely on the GPH your pump can do at the specified pressure. As I said before: 18 150 gpd residential membranes are needed to do the work of one XLE 4040. The residential membranes cost 26 bucks, the XLE 4040 cost 250 bucks. So the question here becomes simply - what do you need in flow from your RO.

I personally think that if you have electricity in your shack, the best way to go is size your pumps so they can take the larger membrane and start with the residential membranes. Get a feed pump - i.e. a pump for an RV that can do a decent amount of GPM or a stainless sprinkler pump. Then get two larger procon that have lift on the suction side. The biggest pump - 300 gph- does not have any lift so you need to make darn sure that the pump does not run dry. Use the procon's as pressure and recirc pumps. Putting procon pumps in series increases your pressure. So if you then upsize your membranes you can reuse the pumps and just get a third one for recirc.

For example 3 residential membranes and housings on wateranywhere cost 84 dollars. If your pumps can take it - doubling your capacity only takes another 84 bucks.

The stainless 4040 housings are $200 so for 450 you go to a XLE 440. If you plumb with Pex Al Pex from be start you have 300 psi capability in your system.

[bowhunter]

Thanks for all the info. You helped confirm my thoughts about nano filtration. I was leaning that way and did get a recommendation from a membrane supplier to use nano filtration, but I was concerned about potentially stripping out enough minerals to significantly change the syrup. I hadn't thought about using a recirculation pump to keep the velocity up, but that is a good idea that I may do as a refinement later.

[not_for_sale]

If you don't have the required flow on the concentrate side (look at the spec sheets of the membrane) you need recirculation for any decent performance.

There is actually research out there that says that some Nanofiltration membranes remove a significant amount of potassium but that it only changes taste later in the season. Search for "Adulterated Syrup with NF270 Membranes". I posted the PDF there.

[not_for_sale]

I always say that I am a physicist and know a little about everything, but nothing right. ;-))
In addition, I've been in the real world for over 20 years so I am a little rusty.

But - regarding the osmotic pressure calculations above: there is an engineering consideration you forgot. First, those pressures are the theoretical limit in a static system. Second, the membrane material changes with temperature - that is why temperature has a higher effect. Third, if you start with two percent solution, push it through, osmotic pressure changes but you also move fresh, and lower concentration solution through and the higher concentration away.

So the practical limit may very well be different than the theoretical limit. That is what makes experiments interesting.

[not_for_sale]

Flux capacitor got my 70 gallons of sap to 9.8% overnight. Ran with recirculation and ran the residential membranes at 150 psi. 7299

[bowhunter]

You're probably right. I was just trying to see where the limit might be. What you demonstrated overnight is probably very close to the expected concentration starting at 2% and running the membrane at 150 p.s.i.. I spent over 40 years in the oil/chemical processing industry and recognize that theory is theory and the real world is always different, but I am constantly amazed to see how well the laws of physics work in the real world.

[not_for_sale]

I just ran raw sap at 150 psi. Don't know what the input was but the concentrate came out at 7.3 %. I collected this sap this morning. So it is a little warmer than the one I processed yesterday in multiple passes.

I also did a permeate was before I ran the sap. Did the permeate wash with 100 degree permeate that I had heated on my induction plate.

This thing never lost much efficiency in 10 hours and I am pretty sure it's because the concentrate side flows so fast with the recirc pump and the procon driving 3 membranes in parallel.

[saekeaton64]

not_for_sale, could you post a few more pics of your setup now that you have made your new modifications?

[Snowy Pass Maple]

Just a word of caution here - earlier discussion was referencing using an Aquatec as the recirculator... not sure if that is still the case...

Aquatec is quite emphatic that their pumps are not designed for anything about 60 psi at the inlet side. When I inquired about how serious this spec was, I was told some people have done this as high as 100psi - but this was not advised. This is also clearly stated on the spec sheets. With this style pump, if you're going to have two, you're better to run in parallel to your target pressure to get more flow, even though this does result in less than what you'd get putting it at "open flow" in the recirc. loop. The failure would be seen as leaking out the side of the pump as the inlet seal is overwhelmed. The parallel flow has been working pretty nicely for me on some recent tests.

Note that this rating is of course different from the pump's internal/exit pressure capability.

[not_for_sale]

Yes, it's still the Aquatec.

I will have power in my shack so I am not limited to 12 v pumps. The way this is working right now I think I am going to use 5 elements in parallel and use two shallow well stainless booster pumps. One as the feed pump and one as the recirculation pump.

I tested the practical limit of this setup as far as Brix go. I took a small amount of 2 % am and circulated it until I hit maximum pressure. The gauge was a 175 - 30 psi above what everything was rated at and the max I got with the permeate line bare dripping was 10.8 %.

I want to get into the 9.5- 10.5 % with overnight recirculation. So I need the 145. I can achieve that with a residential setup without going too much outside of specs.

[Snowy Pass Maple]

Why not aim for something closer to 8% and stay within the manufacturer specs? You're talking about the difference of removing 29 vs. 31 gallons of water before going to boil. I can't really see the point to run things above pressure limits to achieve that incremental benefit - will only take one ruptured line, failed fitting, or damaged piece of equipment dumping a load of concentrate at night to negate any value of that. The energy cost of each gallon of removal will also escalate approaching the brix limit as permeate rates slow, pushing to the asymptotic limit.

[bowhunter]

I agree completely with Snowy Pass Maple. Never exceed the manufacturers recommended pressures. Obviously everything is built with some safety factor in case the maximum pressure limits are exceeded for a short period of time, but not for a sustained operation. Once pressures exceed 100 psi it becomes more risky. If you ignore the manufacturer's spec's, you will never get the performance, reliability or safety the system could achieve.

There is a lot of really good information on this thread for safely designing a small unit operating at or below 100 p.s.i. following the manufacturer's spec and data sheets. If you really need more performance, get someone to design the system who's qualified or buy a system that was engineered for the service.

[not_for_sale]

I generally agree with the safety concerns. But go tell that to the bus that use nitro to boost their engine.

100 psi is an arbitrary number. Take PexAl for example. Why is it rated at 200 psi when minimum burst specs are 900 psi and long term fatigue stability is at 300 psi?

I think some engineer said one thing and the lawyer made the decision.

Never the less - for me this is experimental. The thread is about a flux capacitor. Reason: I am experimenting. And the German word for capacitor is condensator. ;-)

[not_for_sale]

I hate autocorrect. One of the things I found hilarious over here that engineers have to design for maximum user stupidity. While its legal and sensible to assume basic intelligence of the user in Germany, it is frowned upon by the legal department in America.

[not_for_sale]

Now my third post about this and then I shut up. I guess I should've though before I tapped.

Experimenters can only develop something great by bending the rules. When you commercialize everything before you build it it won't become great. See what is possible first, then find the suitable material. So what if something breaks while you try. If you are afraid of something breaking you won't try. That also assumes basic intelligence of the experimenter.

[Diesel Pro]

A lot of us beginners would really appreciate a step by step guide to building a home RO. If there is something out there online please point us to it.

[jrgagne99]

Pretty sure these threads on Maple Trader are the best information you'll find anywhere the web.

[not_for_sale]

I agree. I started with jrgagne's excellent information and just built it.

One thing everyone has to realize is that there is value in manufacturers making this stuff. Too often we'll fall prey to thinking like Buzz Cohn in the sugar poem thinks.

Yes, the materials can be had for cheap, yes, it's simple, but having the experience what works and what doesn't can't be had without actually experiencing it. Some errors can be avoided by reading others trials and tribulations, but by far not all errors can be avoided.

That is why I started this thread. Just to help someone learn with me.

Of course, I haven't made any mistakes or wasted any money, but that is just because I am perfect. Most people aren't as perfect as myself......

Oh, am I full of myself now ;-))))

[ihuntbear]

still no pics yet ???still waiting hahah

[not_for_sale]

I have lots of pics ... the box is just under lots of pressure right now since we have great flow. My residential system has seen over 500 Gallons in the past week, and still running strong. Not too bad for $300 in parts.

I have posted this in other RO threads, but wanted to give a short update here:


Pretty good price for these components on Amazon. Amazon has been one of my favorite supplier for this stuff.

Pumps:

1. http://www.amazon.com/gp/product/B0099TTV76?ie=UTF8&camp=1789&creativeASIN=B0099TTV76&linkCode=xm2&tag=mapletrader-20

2. http://www.amazon.com/gp/product/B000LDG4L0?ie=UTF8&camp=1789&creativeASIN=B000LDG4L0&linkCode=xm2&tag=mapletrader-20

3. If you get the Aquatec, make sure you get the 8852 - some sellers sell an 8800 pump and then its likely the low flow pump. My Aquatec can handle three 150 GPD membranes in parallel with cold sap easily. This Amazon price on it is only $79 right now - pretty inexpensive: http://www.amazon.com/gp/product/B00A7ZV2GO?ie=UTF8&camp=1789&creativeASIN=B00A7ZV2GO&linkCode=xm2&tag=mapletrader-20

Membranes:

1. http://www.amazon.com/gp/product/B000WPHUAG?ie=UTF8&camp=1789&creativeASIN=B000WPHUAG&linkCode=xm2&tag=mapletrader-20

2. http://www.amazon.com/gp/product/B004BG5ETG?ie=UTF8&camp=1789&creativeASIN=B004BG5ETG&linkCode=xm2&tag=mapletrader-20


The stainless housing for the XLE 4040 is actually cheapest on ebay. There is a seller selling them for $80. Amazon sometimes has the housings for the residential membranes at $7.00 plus shipping.

[not_for_sale]

Here is what I use to backwash. RO soap is actually lye (NaOH). This is pure lye (food grade even though it says drain opener :) ).

http://www.amazon.com/gp/product/B002BW4MV8?ie=UTF8&camp=1789&creativeASIN=B002BW4MV8&linkCode=xm2&tag=mapletrader-20

In order to backwash with this - you need to make sure you know the PH of your bachwash water. You can either get a PH meter, or, your could just get PH strips. Effective backwash is at a PH of 12. Some recommend a PH of 11. 12 is more effective though.

The review on the meter are kinda bad, so I just used strips.

Strips: http://www.amazon.com/gp/product/B001T77WW2?ie=UTF8&camp=1789&creativeASIN=B001T77WW2&linkCode=xm2&tag=mapletrader-20

Meter: http://www.amazon.com/gp/product/B004G8PWAU?ie=UTF8&camp=1789&creativeASIN=B004G8PWAU&linkCode=xm2&tag=mapletrader-20