DIY Watermaker

[Yachts66]

Daddle, the short answer is NO. Some do it, that is true, but running your main engine to do a 2HP job does not make a lot of sense to me. I like the idea of running a AC motor to make water and I think the way to power that motor is with a gen set, unless you had a very large battery bank and an inverter large enough for the job. Your main propulsion engine is the most expensive piece of machinery on the boat, so using it for small jobs like making water is nonsensical to me. It's also not a good idea to run a diesel engine without a load, and a 2HP draw on a 50+HP engine is no draw. My $0.02 anyways!

Regards,

Thomas

[hellosailor]

Nick-
" Our membranes are connected in series ,so the flow to the 2nd membrane is full pump output minus the product output of the first membrane. Roughly 66% or more of the pump output reaches the second membrane. This difference is exactly why that 2nd membrane always fails first."
Maybe I also took my "dense" pills tonight. When you say your filter membranes are in sereis, do you mean you supply pressurized raw water:


>=====(1)=======(2)=====(3)=====>

Through membranes in serial sequence??

As opposed to feeding a "Y" that splits to feed two membranes in parallel?

In the former, I would think the second membrane sees less pressure and cleaner water, so it would fail later rather than sooner.

What's the muddy part here?

I would have thought that since water is basically incompressible, once you pressurize one "tank" of water, it wouldn't matter how many membranes (or square feet of membrane) were attached to it in common, as long as that tank pressure stayed high enough (the membranes provided enough resistance) so they all were being "pushed" by the tank pressure.

[daddle]

Yacht66: True. I suppose with a little planning I'd run the water and charging at the same time. The energy for that 2hp (150A @ 12v) comes from somewhere, seems foolish to run it thru the batteries which will then need to be charged by the genset. It all depends on the bigger power picture.

[Yachts66]

Someone please correct me if I'm wrong, but when you have filters connected in series, the brine from the first filter flows to the input port on the second filter. Thus the input to the second filter will have a higher "salt" content since some "fresh" water has been removed. Since you start the filtering process on the second filter with water which has a higher solids (salt) content, the resulting output will also carry a higher salt content (be not as "pure" as the fresh water from the first filter. Still at acceptable levels, just not as "pure." Also, since the product going in has a higher solids content, the efficiency of the second filter is lessened, so you get a reduced quantity of fresh water from the second filter. Did I get that right?

Thanks,

Thomas

[Yachts66]

It's just too darn bad no one makes 2HP 24V motors ya' know? 3HP would be even better! ;-)

Regards,

Thomas

[s/v Thea]

Originally Posted by Yachts66

It's just too darn bad no one makes 2HP 24V motors ya' know? 3HP would be even better! ;-)

Regards,

Thomas

Leeson make a 2hp 24V:

LEESON Electric Corporation - www.leeson.com

[s/v Jedi]

Hmmm okay... let me start by stating that Tellie should answer this because he is the RO expert here and I am just an amateur ;-)

But this is how I see it: first, the output from the pump must go somewhere, i.e. water doesn't disappear. With the membranes connected in series, this means that the reject of the 1st membrane becomes the input of the 2nd membrane. Reject = input - product output so it's easy to calculate the flow for each membrane.
I don't think the product output from the 2nd membrane is saltier than that of the first. The membrane will not allow salt molecules to pass except of not being perfect at this but the position of the membrane isn't a factor in that... at least not initially.
Membranes are cross-flow filters. They depend of the "reject" to clean out the filter. As the 1st membranes gets more flow than the other(s) it is cleaned more thoroughly so will last longer. This means that the 2nd membrane deteriorates faster so it's output can become less and/or saltier for that reason.

My experience is that a 1.5hp motor is enough for a dual (in series) membrane setup for 40-50 gph. I have been running that for 7 years full time now, trouble free. The first 6 years I powered the motor with 120V and a year ago I changed our setup and re-jumpered the motor and power it with 240V. The 240V works better for smaller gensets and I'm almost sure a 3.5 kW genset can handle the surge. You can always replace the start-capacitor with a special soft-start type to improve that.
We have a 6 kW genset and never had trouble. A 16kBTU A/C unit has much more start-up surge than the 1.5hp motor.
I must also state that I always start without pressure, i.e. the regulator is fully open. I let it flow that way until bubbles cease before slowly applying pressure. Our water maker is fully manual control. If you start with 800psi pressure you get a pressure shock and a bigger power surge.

cheers,
Nick.

[reka30]

Nick,

Your answer was very helpful.

In your setup,

1) Are you pushing 2.3 GPM with your Cat Pump or a different amount?
2) Are you using (2) 2.5 x 40" membranes are (2) 4 x 40" membranes?

Thanks again for your willingness to share!

[s/v Jedi]

Hi Reka,

I have an AquaPro pump (it's titanium!). Without pressure, I get about 3 gpm or just a bit more (my meter maxes out at 3). With pressure on, it's about a quart less so more than 2.5 gallons. The actual readings are 2gpm reject and 40 gph product which translates to 2.67 gpm.

I have 2x 2.5"x40" membranes. This is a standard setup for 40 gph water makers and more and more membranes allow you to take it up to 50 gph product output. You need to check the specs of the membranes for their rated output.

cheers,
Nick.

[foggysail]

WOW!!!

I just finished reading this entire thread!!!! THE INFORMATION HERE IS PRECIOUS. AND IT IS FREE!!!

Last week I ordered a brass Arimitus model 307 pump. I have worried about using brass but I also worried about the higher cost materials. I even thought of building a simple low voltage current limited DC power supply to apply electron flow from the brass head to the body in an attempt to minimize galvanic corrosion.

I don't think it is needed! My next immediate hurdle is to mate a 16mm, 2 HP, 220VAC motor shaft to an internal 28mm pump shaft.

Foggy

[muskoka]

Foggysail,

Can you provide details on where you purchased your pump and the cost?

Thanks

[foggysail]

Let me get back to you in a few days on this. I need to check with the company I purchased it from first. If they don't want it public, I will send you an email. I selected the Arimitsu over the CAT only because of post by car washers that the Arimitsu is less noisey. Now I know CAT makes fine products and I am not putting their stuff down at all. It was just my choice.

Oh while I am at it, the electric motor can be found at Omega.com. Their 2HP motor was much less than anything I could find elsewhere and it is a 1725 RPM, 56C machine. MOtor is about $200 with ball bearings, TEFC features.

Foggy

[goboatingnow]

With non energy recovery based Watermakers the vast majority of the energy is lost at the needle valve. Hence series based membranes make better use of the available energy. But there's lots of ifs and buts and a read of the Dow filmtec website is very useful and you can get their design software too. In a series configuration it's important that the second membrane remains within it's design spec. Mutiple serial ones generally can't be used as the over system pressure becomes very high.

Hence two in series is more efficient then two in parallel. ( without erd).

By the way ERDs for small seawater ROs are becoming available danfoss being the leader here. The Clark pump isn't the only way nor the best.

In my opinion it's not worth bothering with anything other then brass heads because if your treating your watermaker right you're doing plenty of fresh water flushes so the unit will not. Be seeing too much seawater. Electrolysis isn't a problem either ( if your really freaked then put a pencil zinc into the pump body )

The other thing I don't understand is why nit consider the 4" modules they seem to be a sweet spot for price, twice the output for about 25% more

[foggysail]

Originally Posted by goboatingnow

Hence two in series is more efficient then two in parallel. ( without erd).

By the way ERDs for small seawater ROs are becoming available danfoss being the leader here. The Clark pump isn't the only way nor the best.

In my opinion it's not worth bothering with anything other then brass heads because if your treating your watermaker right you're doing plenty of fresh water flushes so the unit will not. Be seeing too much seawater. Electrolysis isn't a problem either ( if your really freaked then put a pencil zinc into the pump body )

The other thing I don't understand is why nit consider the 4" modules they seem to be a sweet spot for price, twice the output for about 25% more

You and I are in solid agreement!!! I doubt that I will have enough water for parallel operation so the plan is to go series. But you mentioned the 4" membrane!!! I was looking at that last night. But I don't know the specifications on a 4 but heck, if one can get twice the product, why bother with two each 2 1/2s?

BRASS!!!!!!!! The more feedback I read on brass convinces me that I made the right choice. The cost of brass is about 1/5 the cost of SS and half that of bronze.

GREAT SITE, GREAT THREAD!!

Foggy

[beau]

I have 2 X 2.5inch X 40 inch RO membranes set up in series.

I intend to buy a brass pump pressure washer with a gasoline motor.
These have a high pressure rating of 3,000 psi, they also quote the flow rate but you are never told at what pressure rating that flow is.
With these plunger type pumps does the flow rate vary with the pressure say like on a centrifugal pump or is it fixed.

What flow rate will i need at 800 psi to run my system.