P14026: Performance Evaluation Fixture

Detailed Design

Table of Contents

Critical Subsystems

Modeling the Lungs

Functional Decomposition
public/10_17_2013 Lung Functional Decomposition.PNG

Design Proposals
The team deliberated on whether to use a computer-simulated lung or a physical, box-and-bag
simulation, and decided that the mechanical lung was the most feasible choice.

The team was later provided with a Siemens 1L Test Lung as well, but the box and bag simulation
was the standard which prompted the team to make the Test Lung more flexible and incorporate it
into the design.

PUGH Chart

These charts were used to aid in the decision-making process and are instructive with regards to what the
team considered feasible at the time of their creation.

public/10_17_2013 Lung PUGH.jpg public/10_17_2013 Lung PUGH2.jpg October 17, 2013

public/10_24_2013 Lung PUGH.jpg public/10_24_2013 Lung PUGH2.jpg October 24, 2013

Modeling Patient's Spontaneous Breath

Design Proposals
The design must be capable of simulating healthy and diseased respiratory states by the adjustment of
pressure in the system.
In the process of brainstorming, the team came up with the ideas of using either an EMT Hand Pump,
a Syringe Pump, or a Turkey Baster. The first and the last would be cheap and effective, but the Syringe
Pump is graduated and guaranteed accurate by an outside company.
The team settled on the Syringe Pump since, with calibration, it would be capable of measuring the exact
volume being added to or taken from the system.
Proof of Concept
See Detailed Design Review (Slides 13 - 19)

Modeling Patient Receiving Compressions

Another reason the team chose to use the Syringe Pump is that it could be actuated in pulses to simulate
the pressure spike caused by a chest compression.
Originally, Team 13027 was going to simulate compressions by spikes in pressure, but at present Team
13027's MediResp-IV displays a timer and assumes the compressions are occurring.
Due to the scaling back of this functionality, the syringe is no longer being used to test compressions.
The team hopes future work on this project will incorporate the pulse motion for a syringe so that the
MediResp can redevelop this functionality.


The team bookmarked seven different pressure sensors developed for low-flow conditions. The specifications
show that many are out of the team's price range, but there are several suitable models.

The team also looked into other available lung assemblies to possibly expand the range of lung conditions
the team could test and validate the PEV's functionality.


The team went through several iterations of schematics in an attempt to ensure that the design being discussed at
the time was physically feasible. The first design included the LTV 900 (itself a PEV) as a bundle of sensors:
public/System Design v1.jpeg

The issue came up when reviewing its documentation. It was discovered that, although nominally a PEV tester, the LTV
is only capable of testing itself. The team verified that this was the case by examining the unit's catalog as well as its
quick reference tool and specifications

After this realization in Week 6, the design was overhauled to make do without the LTV sensors previously designed
around. The second iteration was greatly simplified and used parts that were already in the team's possession to
minimize cost:

public/System Design v4.jpeg

At this point, after further studies of Team 13026's Medi-Resp IV PEV, it was discovered that the system's flow is
not bidirectional, that is, the Medi-Resp does not accept flow back in after it has inflated the lung. This prompted
the fifth version of the design:

public/System Design v5.jpeg

The next sketched design saw the distinctive inclusion of an orifice plate to simulate airway resistance in place of the
resistance tubing depicted before. The drawing also shows the concept (developed further below) of increasing
the weight on the lung, thereby lowering compliance while relying on the restrictive plastic on the Siemens 1L
test lung to produce a correctly-shaped waveform.

public/2-19-2014 System Detailed Design.PNG

The last sketched design is the team's final system design with all the different components that are needed. This design
reflects Revision 5 of the Bill of Materials.

Lung Compliance


Varying the compliance of the test lung poses several unique challenges. Lung tissue is elastic but nonlinear, making it difficult to replicate its changing resistance at different volumes (i.e. its "compliance"). It is imperative that the team acquires the ability to vary this characteristic if it aims to simulate several states of health. Note how, in the figure below, a single health condition can change the compliance curve:


The team considered using a rubber band or a weight on top of the Siemens Test Lung, leaning toward the weight method since it will likely be more repeatable in an experimental setting.


Team 14026 needed a way to model different lungs' compliances, varying with parameters like volume, elasticity, and resistance, so that the simulated data could be compared to the data from the tester. This would serve as a basic check for quality data.

To fill this need, Team 14026 developed a program in MS Excel that will generate ideal breath curves using Bezier splines. The program will calculate all the points on the curve and give a simulated set of data for that curve. Alternatively the team can plot a curve by defining six measured points that were guessed to be on the curve. This second method generates a simulated data-set that will be useful during testing.


The program was validated by comparison to real time breath data collected by Dr. Risa Robinson. Her data comes from earlier testing using a strain gauge chest band. Below is a graph of the program's generated curve (showing volume in mL and time in seconds) and the real time data from Dr. Robinson with 5% error bars displayed.

public/12_12_2013 SimulatedvsRealBreath.PNG

The generated data falls almost completely within 5% error of the gathered data, meaning that this Excel model will prove to be a valuable tool moving forward with the design.

Compliance Program

Browse the developed program

Airway Resistance

The resistance to flow in the airway of the human respiratory system is defined as (Equation 1):


There are many ways to create the ranges of pressure drops needed for the variation in flow produced by team 13027's PEV. Three Solutions were investigated. The first being series of lengths of tubing this proved to be impractical since the length of these tubes were very large and would be significantly impacted by the variation in flow. The second being laminar flow elements (LFEs). This proved to be impractical as well due to the fact that these are typically used for turbulent flows, and the flow that we will encounter is laminar with a very low Reynolds number of 20.83 (this is the maximum Reynolds number taken at the maximum flow rate of 32 L/min, pipe inner diameter of 1/2", and density and viscosity were taken for air at STP). The third solution investigated was the utilization of an orifice plate. This solution proved to be practical, easily manipulated, and easily verifiable.

For orifice plates, flow is directly proportional to the square root of the pressure difference across the plate as follows (Equation 2):


Rearranging and substituting volumetric flow rate for mass flow rate we get (Equation 3):


Orifice plate diameters were found by averaging upper and lower bound flow rates where the pressure difference was found from equation 1 for various target airway resistances. These will have to be validated to ensure the proper pressure drop is applied for each resistance that will be simulated. The results are in the table below:


A CAD representation of how this simulated variable airway resistance will be integrated into the validation of the team 13027's MedirespIV can be found below:


Bill of Materials (BOM)

public/10_23_2013 BOM.PNG

Detailed BOM:

1st Draft

11_21_2013 BOM.xlsx

Revision 1

12_4_2013 BOM.xlsx

Revision 2

12_5_2013 BOM.xlsx

Revision 3

2_4_2014 BOM.xlsx

Revision 4

2_12_2014 BOM.xlsx

Revision 5

2_19_2014 14026 BOM.xlsx

Revision 6

2_25_2014 14026 BOM.xlsx

Test Plans


Will be split into three phases of Calibration, Evaluation, and Integration, which prioritize the customer's needs and ensure that each process is individually validated before being used to validate other processes.

I. Calibration

A. Test Sensors
1. Pressure
Use manometer to validate pressure sensors
If within 5% of the sensor, then proceed
2. Flow
Use validated pressure sensors in series with flow sensor
With known pipe roughness, calculate flow analytically
If flow sensor is within 2%, proceed

B. Test Sensor Connections to PEV

1. Pressure
Run several tests with Direct Connection to determine actual output
Determine discrepancy between intended input and actual output
2. Flow
Test sequentially over build of assembly using flow meter
If lose 1% of flow, then look into forming better seals

C. Demarcate Values on Dials

1. Pressure Dial
2. Volume Dial
3. Time Dial

II. Evaluation

A. Check Basic PEV Capabilities
1. Pressure
Peak Inspiratory Pressure (PIP)
Peak End-Expiratory Pressure [Extrinsic] (PEEP)
Overpressure Alarm
2. Flow
Tidal Volume
Inspiration Time
3. Time
Breath Rate
Rise Time

B. Check Mode Functionality

1. Automatic
Validated by consistent Input and Output
2. Manual
Validated by response to operator direction
3. CPR
Validated by timing check
4. Assist
Use stopwatch or switch to time and validate actuation is working as expected (See 12_4_13 Mitigation Plan.png)
Validated by application of Added Flow/Pressure in equal measure as a response to a simulated inhalation attempt of known volume
Test for sensitivity of reaction

III. Integration

A. Check Complex PEV Capabilities
1. Pressure
Mean Airway Pressure (correlates with quality of oxygenation)
Peak End-Expiratory Pressure [Intrinsic] (PEEP) (management is necessary for patient safety)

2. Flow

3. Time

Accounting for Error

The sensor uncertainties were noted so that the team could evaluate whether they were suitable for the piece of instrumentation to be designed.

A summary study of error propagation showed that the sensors the teams intends to buy should be accurate enough to provide excellent data.

Design Reviews

Week 9/10: Subsystem Design Review

The following document shows the team's presentation for the week 9 review.

10_29_2013 Subsystem Design Review.pptx

10_29_2013 Subsystem Design Review.pdf

An overview of the feedback received from Week 10's presentation on Subsystem Design can be found using the link 10_29_2013 Week 10 Feedback.docx.

Week 9/10 Summary: Week 9 System Design Summary.docx

Week 15: Detailed Design Review

The following document shows the team's presentation for the week 15 review.

12_5_2013 Detailed Design Review.pptx

12_5_2013 Detailed Design Review.pdf

An overview of the feedback received from Week 15's presentation on the Detailed Design can be found using the link 12_5_2013 Week 15 Feedback.docx

Open Action Items: 12_9_2013 Action Items.docx

Home | Planning & Execution | Problem Definition | Systems Design | Detailed Design | Build, Test, Document | Project Review | Photo Gallery