Custom Bleed Valve and Flow Splitter

Last crawled date: 1 year, 11 months ago

Hi All,

Thank you for taking the time to review my entry into the Oxygen Valve Splitter GrabCAD Challenge.

With the contest rules stating that the object must be entirely printed out of ABS, it makes it impossible to make a working flowmeter/regulator setup that is most commonly used to accurately measure the flow rate of oxygen to patients. These devices use springs, metal threads, and gaskets to achieve their desired function. The restriction to ABS prevents us from using any materials that may be used as a gasket to prevent oxygen from escaping. In the future, it might be worthwhile to open the challenge to use Ninjaflex or other flexible filaments that can be easily printed on a hobbyist grade 3D printer.

Using a common FDM printer also eliminates many design options when creating the flow splitting device. Despite these challenges, I worked to find a viable solution that is easy to print and accomplishes the goal of controlling the flow rate. I can change the diameters of the internal geometry using conservation of mass to create known flow rates out of different orifices. For the flow rates to divide properly, each exit must be tuned based on the cross-sectional area of the opening. For the first part named the 5050 Splitter, it takes the entrance flow off of the tank regulator and divides the flow in half. I utilized McMaster Carr Part 2429K56 as an example for sizing the ID of the part at 5.6mm. I then calculated the diameters of the new splitter sections by dividing the area of the 5.6mm diameter entrance and recalculating the new cross-sectional areas. The spreadsheet linked below has a nice outline of the calculations:

https://www.swiftcalcs.com/worksheets/37bdda44c30ca482cf2e931933a51d5c/Splitter_Area_Calculation

Now that I had a splitter dividing the flow, I needed a way to choose a precise flow rate. Using the splitter, the user of the system will provide 20 L/min of oxygen to the splitter. The splitter then splits the flow into 10 L/s for each branch. Without using any moving parts we can simply “bleed” off flow to reach the desired amounts. The second part is used for this purpose. Using the spreadsheet linked, 30% of the flow is bled off from that branch, which would allow us to provide 7 L/s to the patient at the end of the branch.

Using Solidworks’ flow simulation package I was able to replicate my results for different diameters. It’s certainly not perfect, but the above solution does seem to provide reasonable results. The parts can be tuned and validated by the community here before the STLs are sent to the field. Using a simple oxygen tank and flow meter downstream, it would be very simple to print and validate bleed splitters for each desired flow rate. The bleed splitter shown in my photo only took approximately 20 minutes to print.

I know it is not an ideal solution to have to bleed off oxygen in a low resource setting, but it seemed to provide the best compromise in making a splitter system with real flow adjusting capabilities. Some of the waste can be avoided if the technician is able to group patients with similar oxygen needs. They can do the math quickly in their heads based on the bleed splitter’s percentage of bleed off to select the proper splitter. It also requires extra sets of tubing, but I think it works nicely to be able to hot swap splitters and not have to print new sets entirely. Ideally, one would just have to print new bleed splitters. Here is an excellent blog post by Mike Yeats about repairing flow meters out in the field which helped push me to a simple design without gaskets or springs.

https://theelectricsquirrel.wordpress.com/2013/11/18/oxygen-pressure-regulator-maintenance-in-pictures/

Feel free to ask any questions for clarification on how the system works. I can provide more information about some of the flow analysis upon request. Thanks again for taking the time to read my entry!

Best,
Sean