Deconstructing Spectra's Clark Pump

[Frode]

Originally Posted by ssmoot

Anyways, I thought it was interesting that once you knew the right name for this sort of pump, it kinda takes the mystery out of the system. I have no idea if that means finding an intensifier with a bypass (to keep the brine separated), that’s rated for potable water and an appropriate duty cycle for a reasonable price is viable for the DIY inclined.

You are correct in that part of the operation of the Clark device can be explained as being a pressure intensifier. You could use such a device, in place of the conventional piston-based high-pressure pump, in its corresponding desalination system, where the membrane pressure is regulated by adjusting the output flow.

The additional feature of the Clark device is its energy recovery ability, which is what (in my mind) makes it so attractive. Instead of wasting the energy contained in the highly pressurized brine from the RO vessel, it recovers most of it, and applies it to the incoming water. This makes the overall system's energy consumption, per unit of desalinized water, significantly lower than when using the powered piston pump.

There are other devices that are very much like the Clark device. Here's a link to a product page with a descriptive video.

[RaymondR]

Prior to the introduction of reverse osmosis water makers on drilling rigs unit called vapor compression water makers were used.

These consisted of a heating tank, with an electric heater, to which a vent was connected to a Roots blower. The output of the roots blower went into a heat exchanger, which was installed in the heating tank and the fresh water output was passed through some more heat exchangers to heat the raw water input to the heating tank.

The flaw in the scheme was the Roots blower which rather than boosting cool, clean air into an internal combustion engine was now boosting hot, wet steam. Later models were fitted with a centrifugal blower which was a monster about 3' in diameter.

Another type, which were fitted on drill ships where there is plenty of room available for the large footprint, were waste heat units.

Waste heat from the engine exhausts heated warm raw water from engine cooling in a large vacuum chamber. The salt free vapor condensed and was extracted from the vacuum tank using a fresh water venturi.

The membranes for the early RO units used on drilling rigs were very expensive, about $30,000 of 1970 dollars for a replacement set.

[Pizzazz]

Originally Posted by RaymondR

Waste heat from the engine exhausts heated warm raw water from engine cooling in a large vacuum chamber. The salt free vapor condensed and was extracted from the vacuum tank using a fresh water venturi.

I have always been curious if you can make this work on a small scale. It looks to me as if it is just the vacuum pump and some vessels. Probably you cannot make it as efficient and flexible as RO but looks interesting.

SV Pizzazz

[RaymondR]

Originally Posted by Pizzazz

I have always been curious if you can make this work on a small scale. It looks to me as if it is just the vacuum pump and some vessels. Probably you cannot make it as efficient and flexible as RO but looks interesting.

SV Pizzazz

Yep, just about every time I looked over the transom on my previous boat I would observe steam ie. condensation, along with warm water coming from the exhaust. This would set me to pondering whether it could be a significant addition to my fresh water supply.

Years ago I was responsible for the operation of an offshore drilling tog in the South Atlantic off the coast of Brazil. On one of my visits I noticed an overboard discharge and on inquiry was informed that it was the condensation drain from the living quarters climate control. I instructed the crew to divert it into a tank and we could use it for wash down instead of sea water. We were amazed to find that in very hot humid weather we could collect 3-4 tons of fresh water per day. When we spent a little time considering the phenomena we realized that with the quantity of fresh, moisture laden air being taken in from the outside and the moisture being transpired inside by 100 or so humans and the vapor produced from showers, cooking, coffee pots etc. the surprisingly high volume was easily explained.

[admiralslater]

Raymondr did you ever test to see if the water would be safe for humans . At a minimum it would have been easy clean up I would have thought

[RaymondR]

Originally Posted by admiralslater

Raymondr did you ever test to see if the water would be safe for humans . At a minimum it would have been easy clean up I would have thought

No, since it had been exposed to exhaust fumes it probably had nasties in it. However what I think it does indicate is that the waste heat from the engine could be sufficient to extract fresh water by distillation with a scheme similar to the old vapour compression or vacuum tank water makers.

[Pizzazz]

I did check the power requirements for these flash evaporation, vacuum evaporation, etc. approached to distilling water. It does not make sense for a small boat application. The large multistage units (multiple pressure vessels, pipes, scaling issues, etc.) achieve efficiency comparable to Spectra’s assuming that the water is preheated (from an engine or chemical plant). You could probably get close to normal RO filtration efficiency but the unit will be complex and heavy. So, we can scratch this idea. It makes sense for a large ship, submarine, oil rig, etc. For us, nothing beats Spectra or a regular RO watermaker without energy recovery.

[RaymondR]

The spectrum of pleasure vessel owners extends from the extreme DIY folks through to the multi billionaire. Included in the spectrum are those who achieve gratification from designing and building their own boats and equipment, lets call them the "doers" and those who just want to have another status symbol, we'll call them the "havers".

The gratification of your extreme doer is greatly enhanced when they are able to maximize the outputs whilst minimizing the inputs. The gratification of the havers is greatly enhanced when they achieve an extravagance of outputs for the most extravagant of inputs.

Be mindful that we are discussing a spectrum with a normalish distribution and that consequently the majority of us lie somewhere near the mid point. That is we want to enjoy a reasonable amount of civilized luxury without exorbitant cost.

The Spectra water maker and it's Clarke pump are a good solution to the problem of supplying appropriately pressurized sea water to the membrane filters of a separation system however this comes at relatively high cost and mechanical complexity.

In the water making sphere pressure recovery is viewed as a very desirable feature as it enhances the potential for operating efficiency, that is measure by the ratio of power input to fresh water output. From an engineering philosophy viewpoint the Spectra appears to do this fairly well which adds to it's appeal to the middle of the spectrum boat owner.

To the doers though, whilst it may appear to be an efficient use on energy it in no way exploits "wasted" energy and from their minimalist viewpoint does so with an excessive input of engineering resources in it's manufacture and hence it's relatively high cost.

Hence the doers interest in systems which may present a solution which maximally exploits the waste energy available and may allow them to do so with a minimal of probably recycled material engineered using cheap and effective manufacturing techniques readily available to them.

[Northbound44]

It took me some time to really understand the principle of the Spectra water maker. In fact, after rebuilding a Clark’s pump, I realized that the concept is simple but completely different from the classic way, meaning using high pressure reciprocating pump combine with a pressure relief valve making constant flow and pressure.

First thing I didn’t understand was the pistons on both sides of the Clark’s pump have the same diameter, so, the pistons area been the same, the pressure should also be the same. What I missed was that the water circulated back under the driven piston were a rod is pushing it. The area is then reduced to the area of the pushing rod, rising the pressure in the same ratio as their areas, in my case 10:1.

Furthermore, the membrane been in the closed loop, the rod takes the place of some water under the driving piston and since the water at this point has no place to go, the pressure instantly rises until there is enough pressure for reverse osmosis to occur in the membrane, allowing the displaced water to be forced out as product.

We can now understand why this technology is so efficient since the energy is used essentially to produce water in comparison with the classic way were only a small percentage of the energy is used to produce water and most of it is wasted and released thru the released valve.

[RaymondR]

It's called an "Intensifier" and the dreaded Halliburton company has some monsters. The feed pumps, about six of them in parallel, each driven by GM diesels, are rated to about 6'000 psi and the outputs from the intensifiers about 20,000 psi.

I have noticed a couple of other interesting developments in RO pumps with pressure recovery, one was a reciprocating plunger pump with pressure recovery built into the fluid end and another a couple of swash plate developments by Danfoss.

[4kMike]

Interesting! How about replacing my old noisy hi pressure pump with clarks pump and using the old membrane/HP vessel?

[Tellie]

Originally Posted by 4kMike

Interesting! How about replacing my old noisy hi pressure pump with clarks pump and using the old membrane/HP vessel?

It can be done. It depends on how big a watermaker you want to build. Clark pumps are not all the same. They are rated in percentages. You need to make sure the size of the Clark pump will fit your desired build.



That being said, it is that time of year again where we clear out our shop of old equipment and parts that we don't use anymore. This year we have about 6 Clark pumps that we have rebuilt that we are going to sell along with a membrane vessel and connections for the Clark pump and membrane. They will all have the new desired SS fittings on the pump and on the membrane vessel. We will cut to length the hydraulic lines that connect the Clark pump to the membrane vessel that fit your installation requirements. We can certainly add a new membrane and we can add the feed pump/s as well. This way you can either set up a 6.5 gph watermaker or a 14gph watermaker, depending on the pump's percentage. You would need to supply the simple things like 5/8" braided hose, hose clamps and nylon hose fittings. At this point I believe we have one 7% pump (6.5gph) one feed pump, and five 10% pumps (14 gph) one/two feed pumps. These will be basic manual systems minus the electronics. I will also send plumbing and wiring diagrams to see how to plumb these systems. Clark pumps are very hard to come by. We occasionally get these pumps when customers upgrade to a newer watermaker for example. Two have come from Hurricane Dorian damaged boats. I have personally been involved with rebuilding each of these pumps. We bench test them for two hours and will give a six month warranty on the pump from the date of purchase. For those that are a bit handy this might be an opportunity to get a watermaker on board for around half the price of new.

[4kMike]

Thank you Tellie!
We are only two onboard and the 6,5 gph unit woould be OK for us. We are now running our WM with HP pump and genset every 3rd day to make 20+ gal water. Noisy, and needs gas. Is the membrane size critical? We have now one 2.5″ diameter by 40″ Dow R O membrane. I installed the set myself, so basic stuff is already there and working.

[Tellie]

A standard SW2540 membrane is perfect. I would suggest the 10% pump because for a few hundred dollars more you can add a second pump to a 40" membrane and do the 14gph with both pumps running or you can run either pump alone and make 7gph. In this mode it is thee most energy efficient watermaker in the industry. But if you only wanted to install one feed pump it will still work fine.

[4kMike]

Originally Posted by Tellie

A standard SW2540 membrane is perfect. I would suggest the 10% pump because for a few hundred dollars more you can add a second pump to a 40" membrane and do the 14gph with both pumps running or you can run either pump alone and make 7gph. In this mode it is thee most energy efficient watermaker in the industry. But if you only wanted to install one feed pump it will still work fine.

Hi Tellie! I sended a pm to you.