Desalinator [moved from MV Catapult build]

The watery and smelly sides of life.
groper
Posts: 239
Joined: Sun Nov 18, 2012 11:23 am
Location: cairns

Desalinator [moved from MV Catapult build]

Post by groper » Fri Jan 15, 2016 4:50 am

Ok time to talk about desalinators.

Me and Dave (aka finally) are building our own. We have started acquiring all the parts and i will begin putting it together once it all arrives.

I have a jabsco deckwash pump for washing down the cockpit, so i am going to feed the high pressure pump from it via 2x prefilters and a low pressure guage.
We are using a CAT pump 2Sf35SEEL SS high pressure pump.
This is direct coupled (flange mounted) to a 1.1kW 6pole 240VAC induction motor which will spin it at a reduced speed of 940rpm giving a total flow rate of 1.88GPM. The pump will absorb approx 0.82kw at this rpm but electric motor sizes step up from 0.75kw to 1.1kw so we had to go for the 1.1kw. Electrical power consumption will be around 1kW as the induction motor has approx 80% electrical efficiency.
The high pressure line feeds a 2540 membrane in a FRP housing, to a high pressure guage and finishes with a needle valve to regulate the pressure.
The system should provide approx 50L of freshwater per hour, although calculations are showing more like 66L per hour if a recovery rate of 15% is assumed... 66L sounds a bit optimistic, from comparison to other similar systems 50L sounds about right.

We have sourced the SS high pressure pump, SS guages and SS needle valves from the USA for better pricing despite the tanking AUD...go figure... We have found a local supplier in australia for the high pressure FRP housing and membrane which were competitive on the pricing compared to importing from the US. All the other little bits will be sourced locally, ENZED and pirtek etc can make up the high pressure hoses and the prefilters can be bought cheap from bunnings ebay etc... Total cost for the system should be under $3500.
Reason i put this up here is if anyone else wants to build one, let me know as i will be ordering the membranes and housings from an Australian wholesaler soon - same goes for the electric motors and adapter flanges etc. If you want to buy the other gear from the USA, i can put you in touch with my supplier over there which saved us quite a bit compared to any quotes i had from buying locally in this banana republic of ours :roll:

44c
Posts: 1148
Joined: Tue Aug 08, 2006 5:08 am
Location: Hervey Bay, Qld

Re: MV catapult

Post by 44c » Fri Jan 15, 2016 10:31 am

I'm surprised your projected fresh water output is only 50 - 60 lph. But I'd guess you could run two membranes from that size pump? And double the output for not too much more money. (And no more power)

I say this because our watermaker makes ~ 40 litres/hour and runs at around 25 amps (12 volts). And it's not an energy recovery unit like a Spectra, just a straightforward 12 volt motor driven piston pump.

Getting ~60 litres from ~83 amps (12 volts) is a bit power hungry.

One suggestion, which won't cost much. Have a simple ball valve bypass around the pressure control needle valve. You don't strictly NEED this, but having one allows you to "set and forget" the needle valve. I've found they can be a bit fiddly to get right, so it's nice to be able to get them set right and leave them there. Just open the bypass valve on startup, and gradually close it to bring pressure up.

groper
Posts: 239
Joined: Sun Nov 18, 2012 11:23 am
Location: cairns

Re: MV catapult

Post by groper » Fri Jan 15, 2016 12:45 pm

Hi Alan,

The projected output is only a calculation based on the manufacturers fairly rudimentary formulas for output and horsepower etc... until its up and running we dont really know what it will produce or how much power it will consume - likely the power consumption will be less as i believe the manufacturer (CAT Pump) is quite conservative when it comes to the formula for recommending the HP required for the pump.
Some increase in efficiency could be realised by going to a more efficient brushless DC motor, or even a brushed DC motor, which would raised the motor efficiency from ~80% upto ~95% if using a really good brushless DC motor. However the inverter drive for such a motor is quite expensive and also prone to failure based on what ive seen in my line of work on inverter air conditioners. An induction motor is pretty bullet proof, the run capacitor being the only failure i typically see. A brushed DC motor is another option, likely what you have, and is an option worth investigating. Do you have the specs on your motor and pump? And yes, with 2 membranes the output would almost double with only a small increase in power consumption... The idea of the whole thing being, total cost is under $3500 - the penalty is less efficiency, but with enough solar panels - who cares?

The power consumption hasnt really been an issue in my consideration as solar power is so cheap nowdays, spending the money on more solar power to run less efficient systems is a more economical and flexible solution - hence all my domestic refrigeration equipment as opposed to running expensive 12V fridge and freezer...

44c
Posts: 1148
Joined: Tue Aug 08, 2006 5:08 am
Location: Hervey Bay, Qld

Re: MV catapult

Post by 44c » Fri Jan 15, 2016 2:30 pm

Yep, fair point about power production, but still it's nice to be able to make water when it's not 100% sunny too.

You should be able to get an accurate idea for water production from the membranes specs. ie a Filmtec 2540 membrane should produce about 115 litres per hour.

http://msdssearch.dow.com/PublishedLite ... age=GetDoc

Which looks a fair bit better from 83 amps. You'd only need to run the thing for about 20-25 minutes every other day.

groper
Posts: 239
Joined: Sun Nov 18, 2012 11:23 am
Location: cairns

Re: MV catapult

Post by groper » Fri Jan 15, 2016 5:25 pm

I dont think its a s simple as that Alan... the permeate flow rates given by DOW on their membrane is onlt a data point at a specific test condition ie water temp, pressure, total flow rate, recovery rate = 8% etc... its not an indication of what you always get.

Looking at the specs of the efficient marine watermakers, without energy recovery devices, they seem to share one thing in common. They reduce the speed of the pump so the total high pressure flow rate is small, whilst keeping the membrane suface area the same (2540 size). This results in a higher recovery rate of somewhere near 20% - not the 8% data point quoted by dow in their specifications. The problem with doing this in a simple DIY setup is that its very difficult to direct couple the pump to a motor which only turns at circa 500rpm. all of the single phase induction motors and most of the DC motors ive looked at spin at a minimum of 1400rpm and thus move alot more water than is required in order to get the 20% recovery rate. Echotec uses a gear reduction pulley system on theirs, but i dont want that extra complexity. I think where its heading for me, will be a 940rpm motor which is driven via a variable frequency drive so i can slow it down to circa 500rpm. The drive adds an extra $300 to the cost, but it also allows flexibility in terms of flow rates and power consumption... An 240VAC induction motor plus variable speed drive ($600) still costs less than a 375w 12vdc motor on its own...

Once i get this all up and running, the efficiency should not be too far different to the commercial units, the only difference being a slightly lower motor efficiency of im guessing around 10% compared to a brushed DC motor...

groper
Posts: 239
Joined: Sun Nov 18, 2012 11:23 am
Location: cairns

Re: MV catapult

Post by groper » Sun Jan 17, 2016 12:03 am

Ok ive got this sorted now - weve changed to smaller displacement CAT pumps - 2SF15SEEL - and gone to a higher speed 4 pole 750w motor. The rated load of this pump at 1460rpm is only ~500W and the available motor has an improved efficiency at this reduced load of around 83%. after converting the load current to 12V i get 41.6 amps and need to allow an extra 10% for the inverter efficiency of 90%. So all up about 45amps @ 12v now. Product water flow we dont know - i cant calculate it without knowing the recovery percentage rate, but its likely around 50L per hour...

Motors are available in 3 phase with a variable speed drive, or single phase fixed speed, both options are low cost with high quality WEG motors under $300...

44c
Posts: 1148
Joined: Tue Aug 08, 2006 5:08 am
Location: Hervey Bay, Qld

Re: MV catapult

Post by 44c » Sun Jan 17, 2016 7:01 am

How much does the pump flow?

Anyway, I think you'll be pleasantly surprised by how much freshwater you'll make. I say this because our unit has a very small pump, and a 2538 canister size, which limits membrane options and we still get a genuine 40 litres per hour.

Our pump is the 708 - 1 : http://www.villagemarine.com/images/788 ... _pumps.pdf

Maximum flow of about 130 litres/hour.

BTW I don't know if you've priced the Racor titan series pumps. No idea how expensive they might be, and the price might be prohibitive, but I can strongly recommend them from a reliability standpoint. Ours has been in constant use for nearly six years, has made thousands of litres of water, and has simply been completely reliable, and has required no maintenance other than routine crankcase oil changes.

44c
Posts: 1148
Joined: Tue Aug 08, 2006 5:08 am
Location: Hervey Bay, Qld

Re: MV catapult

Post by 44c » Sun Jan 17, 2016 7:48 am

groper wrote:Hi Alan,

Do you have the specs on your motor and pump? ...
The pump specs are in the data sheet previously linked to.

The motor is 12 vdc, 170 watt. Pump is belt driven.

groper
Posts: 239
Joined: Sun Nov 18, 2012 11:23 am
Location: cairns

Re: MV catapult

Post by groper » Sun Jan 17, 2016 8:39 am

Thanks alan,

As you can see from those specs, the max flow that pump will pass is between 0.25-0.48 gallons per minute - its very small.

This is where they are saving energy, by moving a very small amount of seawater... The problem is, this flow rate is below the minimum recommended flow of 0.7 GPM which gives a recovery rate of ~20% @ 25deg C as per the DOW sw30-2540 membrane specs. They suggest a recovery rate this high is the maximum permissible for "intermittent use" and expect the membrane to last less than 3 years. If the recovery rate is lower, such as 8% in commercial continuous use applications, then they are expected to last over 7 years. Going for ultra high recovery rates and minimum energy consumption pays a penalty in terms of membrane life as the membrane doesnt have sufficient flow to carry away contaminants in the waste water...

With your setup i have to ask, how many times have you replaced your membrane since youve had this watermaker? Do you notice reduced product water flow after a while when the membrane is getting some age on it?

On the system im putting together - the new motor and pump combination will produce a flow rate of 1.25GPM - which is above the 0.7 recommended minimum. It will use more energy per litre of produced water than yours does, but the membrane should last a bit longer in return...

44c
Posts: 1148
Joined: Tue Aug 08, 2006 5:08 am
Location: Hervey Bay, Qld

Re: MV catapult

Post by 44c » Sun Jan 17, 2016 9:39 am

Well, yes, I already stated, ours has a very small pump. But the titan range includes much larger pumps. Anyway it was only a suggestion to look at the range and see if any would suit, as I know our pump has been very reliable. I expect Cat pumps would be reliable too, but can't comment based on experience.

We had to replace a membrane a few years ago, but that was due to chlorine contamination from flushing with tap water. It was still producing the same amount of water, but salinity was higher. (After the contamination.)

Our current membrane has a much higher capacity than the original. But because of our small flow pump, we don't get anywhere near the maximum the membrane is capable of. But still very good return in terms of litres/watts. Better than the original membrane, in fact. Maybe it's life will be shorter than "normal", I don't know. It's been in constant use for about 4 years now, still seems to perform like new.

Post Reply