Oxygen Installation

Introduction:
During a hospital stay for a broken patella, it was determined that Ramona required oxygen 24/7, especially at night and when at rest. Following release from the hospital, we obtained oxygen equipment that included an oxygen concentrator (a device that "generates" oxygen content of 50% to 95% from the local atmosphere), oxygen bottles, regulators, hoses, nasal cannula, humidifiers, etc. Upon arrival home, I installed the equipment in a 'convenient' location so that Ramona could walk around the house with extension hoses. We found that the concentrator noise was quite high for the location in the bedroom, but another location with extended tubing made things very difficult when nocturnal visits to another room were necessary. Because the remodeling project of our 'new' home was still underway and most of the basement was still accessible, I decided to try and setup things differently.

Research:
I researched to find local and other codes for oxygen supplies for medical purposes, equipment for flow and pressure regulation, humidifiers, and any other information available. My research turned up some interesting sidelights. The first being that general or US building codes are not available unless you want to pay large amounts for subscriptions to the standards. I was, however, able to find some 'standards' to build my system toward by inspecting drawings and plans provided by contractors that were building medical facilities. Here are a few of the 'codes' that I thought pertained:

Design Phase
So, I have the concentrator with a constant flow indicator, oxygen bottles with a flow regulator, small oxygen hoses, cannula, humidifier 'bottles', etc. At first I tried to find flow devices like those located in hospitals to install at each location. Again, difficult on a consumer level to locate, besides both oxygen supplies (concentrator and bottled O²) already have flow 'regulators'. So the design needs to be with flow control at the source along with piping, valves, nipples, etc. sized and able to handle the flow.

Rather than install a single humidifier at the source, I thought it better to install a small 'bottle humidifier' at each station. This would allow for easy cleaning and monitoring of the water necessary to keep them running correctly, and would not provide moist air in the copper piping. It would be possible that, as the piping runs from a warm room in the basement to a colder room before rising to the main floor, there could be some condensation accumulation that may cause some problems. Besides, the small 'bottle humidifiers' could be obtained from our oxygen vendor.

I felt that I also needed a pressure relief valve so that the piping does not become 'pressurized' when all the valves are closed. Why would all the valves be closed? The design of the system is setup so that Ramona can turn off one station, walk to the next station, turn on the next station and install the cannula. Remember, we are working with a 'flow' system at extremely low pressures. If we close the system and let the oxygen source continue at a constant flow until it reaches its upper pressure limit, it would most certainly add a lot of oxygen to the volume of the pipe installed and necessarily be released when the next valve is opened. I also would not want to 'blow off' the plastic tubing supplying O² in the system because of over pressurization. (During the testing phase, I found that the concentrator provides pressure up to about 6 psi. The oxygen bottles with a flow regulator provided much higher pressure, maybe on the order of 20 psi (My gauge only goes to 10 psi, so this is an estimate. Suffice it to say, either increase is too much.).) Besides, can you imagine the amount of water that might be ejected from the humidifier moving 10^s if not 100^s of liters of O² in a very short time by opening the valve? (The system has over 75 liters of pipe volume, so when it's pressurized to 20 psi, there's LOTs of O² that will be required to exit the system.)

Plumbing:
I have some 1/2 inch copper pipe schedule M. Now schedule M is thinner walled, but rather than spend the extra money for schedule K or L, I decided to use the schedule M. The reason for the K or L in medical facilities is that they use a highly pressurized system (I've seen 50 psi), but I am designing a low pressure 'flow' system, so it shouldn't be a problem. I obtained standard schedule M copper pipe, elbows, tees, adapters, nipples, etc. I also obtained 1/2" gas valves. Had to repaint the handles green rather than the red for natural gas. I also extensively cleaned these valves to remove any oil or grease contained in them and used Teflon tape rather than pipe dope to make the joints. 1/2" pipe seemed like it would be OK. During the testing, the fact that with a short cannula attached there was no measurable back pressure in the system. The only back pressure present was when a 25' extension hose was used, and 1/2 psi pressure was of no issue. Also the entire flow was delivered to the cannula as there were no leaks and the flow meter at the supply either showed or was adjusted to the correct value.

The copper pipe, fittings, etc. were installed through out the house with 7 outlets located strategically for ease of operation and access. I was careful to route the piping below the floor joists as I have installed floor heating and did not want to heat the oxygen supply to 130° or more. A solution of hot water and TSP (trisodium phosphate) was used to clean the piping. Clear water used to rinse it. Then I used oxygen from my supply to dry the system. I also installed an adjustable pressure relief valve and a 0 to 10 psi pressure gauge for adjustment and monitoring. Most of the parts for this system are available at the local hardware store, Home Depot, or Lowes. The pressure relief valve and pressure gauge were obtained from W. W. Grainger, although other sources are certainly available. The relief valve I chose is a VR series valve. That means that it is not a 'pop' type safety valve, which opens to full capacity very quickly, but opens enough to keep a specific pressure and then closes if below the set pressure.

Adjusting the system:
After installing all the above I tested things out and found that the pressure relief valve seems to work best when it is set to about 2 psi or so. This allows the valve to open if all the stations are closed and also will exhaust the entire flow (currently set at 2 liters per minute (lpm), the valve I have will exhaust up to 2 cfm or 56.6 lpm) and only pressures the pipes to 2 psi. This seemed to be low enough to not create any noticeable difference or problem with the plastic humidifier bottles located at each station, no mater what source was used for the O². It also was high enough to stay closed when the oxygen was in normal use. Short lengths of tubing to the cannula caused the pressure gauge to register 0 psi. A 25 foot length of tubing connected to the cannula caused a reading of about 1/2 psi. So the two pound relief should be just about right.

Pictures:
Here are some pictures of the finished product (click on any picture for a larger version):

Oxygen Concentrator	Concentrator connected to system	Oxygen bottle connected to system
Pressure gauge & relief valve with 25' hose connected	Pressure gauge & relief valve with supply valve off	Typical main floor oxygen station