Get our free book (in Spanish or English) on rainwater now - To Catch the Rain.

Difference between revisions of "Dialysis water cooler"

From Appropedia
Jump to: navigation, search
(Second prototype)
(2 intermediate revisions by the same user not shown)
Line 63: Line 63:
=== Second prototype ===
=== Second prototype ===
A second prototype is in construction.
The second protoype has been built and is awaiting testing.  We have purchased a new heat exchanger and added high-pressure fluid couplings so that the unit can tolerate mains pressure water.  It is now ready to be tested.
The fluid couplings are dimensioned to take 1.25 inch fluid fittings. 
=== Next steps ===
=== Next steps ===

Revision as of 20:56, 10 November 2019

Dialysis is delivered in remote communities in small clinics, usually for one or two patients at a time, with each treatment lasting 5-6 hours. Each treatment uses RO water at about 500 ml/min. To generate RO water at about 50% efficiency, up to 1.0 L/min of town water is needed. Heat needs to be dumped at about 1.2 kW for each treatment chair. This should bring the supply water temperature down from up to 46 C to about 28C. A suitable operating supply temperature is around 30 C or below.


Dialysis in the desert

In remote Australian communities, Aboriginal Australians suffer kidney failure at a rate that approaches 10 times the national average. Transplants are rare and frequent dialysis is usually prescribed. Many dialysis patients are elderly and senior members of these remote communities. In an effort to improve the social health of both communities and patients, there has been a recent push to increase dialysis services within remote communities and clinics (Devitt and McMasters, 1998, pg.67). Unfortunately, many of these communities are in arid regions, which means town water tends to be warm where it comes out of the tap.

Remote water temperature

In remote communities, town water is typically raised from the aquifer, then treated and stored in large tanks, often on the top of a hill in the sun. Tank water is piped over-land to various points including each remote clinic, where it may be stored in a smaller clinic’s tank. In summer, air temperatures may approach 50C. By the time the water gets to a clinic it is usually quite warm. Dialysis water pre-treatment depends on small RO systems, which tend to lose effectiveness as the water temperature rises, eventually posting errors and shutting down with water temps around 35C.

The RO systems are designed to fail safely when the incoming water temperature is too high. However, any equipment failure may cause dialysis treatments to be cancelled. Since patients need dialysis treatments on alternating days, a delay of more than one day means transport of the patient to a regional centre (usually Alice Springs) for treatment. This is both expensive and detrimental to the health and wellbeing of the patients. In short, the consequences of missed dialysis are significant from a patient wellbeing and public-health perspective.

Project details

Technical requirements

To provide a source of cool water (below 35 C) to remote clinics at a rate of 1L/min per dialysis machine.



A remote community water supply may reach temperatures greater than 45 C. An RO machine will alarm when its supply temperature is greater than 37 C. An effective pre-cooling device would be one that could reliably take 10-15 degrees C off the supply water temperature at a rate of about 1 L/min. This is equivalent to about 1.2 kW of heat that needs to be extracted from the incoming water.

One of the main challenges of this project involves finding a suitable place and means to sink the heat that is extracted from the water. It is difficult to sink the heat into the outside air, as it can be much hotter than the supply water. Such techniques have been attempted by using chiller packages that operate outside a dialysis clinic and attempt to pump heat from the water directly into the even hotter air. This has been shown to be unreliable due to solar loading and other challenges of heat transfer.

Air conditioning cooling capacity

One approach is to sink the heat into the room's inside air and allow the air-conditioning systems to pump the heat outside. The main advantage of this approach is that the air represents a thermal store that can quickly absorb the heat from the water before it is slowly pumped out by its air-conditioners. The second advantage is that there are usually more than one split-system air-conditioning unit installed in each clinic, providing ample capacity to pump heat from the inside air to the outside environment. A typical split-system air-conditioner has a cooling capacity of about 8 kW. This is more than six times what is required to pump out the heat required to run a single dialysis chair on a hot day.

Existing solutions

Various hack-jobs have been attempted, such as running hose through a chest-freezer, draping tanks with wet cloth, storing water in a tank inside the clinic. None of these solutions have been engineered - they are mostly guesswork. At present, sites have purchased chiller packages - these are worth $8-10K each, are heavy, are not that effective at higher temperatures and are prone to failure.

Design preferences

  • A good solution to this problem would be inexpensive, lightweight, effective, reliable and robust for us in remote sites.
  • As maintenance is difficult at a remote site, a good design would be reliable and low-maintenance and/or usermaintainable
  • Another approach to maintenance is to make the device lightweight and portable, such that it could easily be loaded by a single person into the cabin or tray of a 4WD and installed with a minimum of technical expertise.
  • A good design would be able to be used together with any other cooling system without having to reconfigure the plumbing of the equipment or building.

Online cooling

First prototype


We have worked on the simplest possible concepts for cooling water for dialysis. At present, we have shown that water can be cooled sufficiently by running it through a typical air-conditioner condensor coil, while using a ~100 W household fan to blow air through its fins.

Measured Steady State Temperature Changes

  • Example: Input water temp 46 C, Output temp 28 C, 1L/min = 1.2 kW
  • Example: Input water temp 36 C, Output temp 28 C, 1L/min = 0.7 kW

By approximating the behaviour of this cooling system, we can estimate cooling capacity at 0.07 kW per degree C removed at 1.0 L/min flow-rate.

Physical dimensions

  • Total size including extended surface matrix = 0.7 x 0.46 x 0.60 m
  • Approximately 50 m of 7.7 mm OD, 6.7 mm ID pipe. 1.76 L total fluid capacity.

Material costs

  • Heat exchanger (new) about $300-$400
  • Case and hardware ($100-$200)

Second prototype


The second protoype has been built and is awaiting testing. We have purchased a new heat exchanger and added high-pressure fluid couplings so that the unit can tolerate mains pressure water. It is now ready to be tested.


The fluid couplings are dimensioned to take 1.25 inch fluid fittings.

Next steps

  • The current prototype is made from junk parts and seems very effective, but is bulky. It would be nice to calculate exactly the efficiencies of various off-the-shelf fluid-air heat exchangers and optimise for smaller/lighter/cheaper etc.
  • The design is based around a single-treatment at 1.0 L/min. In most clinics, more than one treatment may be operating at any time - it may be that a single design can cool enough water for more than one machine, or it may be simpler to develop one cooler per dialysis chair. This may be limited by plumbing and space arrangements in each clinic.

Other potential solutions

Offline cooling/storage


Some stakeholders have suggested pre-cooling an indoor tank full of water for dialysis, to save cooling it on-demand. This may be done to take advantage of the cooler night-time weather for cooling water, then storing it until the morning for treatments. By using a heat-exchanger similar to that described for the online approach, water could be circulated through a storage tank and a heat exchanger device over a 12 hour period and the heat could be dumped into the building's inside air and allowing the air-conditioner to pump it outside.

What shape and size of water storage vessel in an airconditioned room can chill 300 L of water by 10 C in 12 hours overnight?

The main design challenges with this approach are as follows:

  • A large tank needs to be constructed and installed within the building.
  • The entire volume of the tank needs to be cooled for any of it to be usable.

Water used at 1.0 L/min for five hours is 300 L. Rather than have one tank per chair, it makes sense to cool all the required water for all dialysis chairs. To run two chairs requires 600 L of water for treatment, plus extra water for cleaning cycles before and after each treatment. If we aim for a minimum of 1000 L tank capacity for two chairs (2000 L capacity for a four-chair clinic), then we must sink a large amount of heat overnight.

With a heat exchanger operating at twice the capacity of our tested online prototype, it is possible to cool sufficient water and store it in a tank overnight. Generally speaking, it would appear that the optimal combination is a small tank and a large, efficient heat exchanger.

"What is the optimal capacity of offline storage volume and heat transfer capacity (kW) to provide enough cooled water in 12 hours overnight for two dialysis treatment chairs?"

Active cooling


Heat collection


Counter-current heat exchange




Future work

Where else in the world would this be useful, and who else has solved this problem and in what ways? Is this technology useful in other parts of the world? This problem exists in about 10-20 remote communities in Central Australia, and there are rumours of a similar problem in South Australia. There may be similar problems in arid/hot regions of other countries that suffer chronic kidney failure (Sahara desert, perhaps?). There is more to be said about this later.


Contact details