top of page
Experiment [Reinforced Casting Process]

Stage 1 : Fixed volume fraction while varying fiber lengths

Objective: 
To decide where the research is leading to because we had a general idea of what we wanted but we weren't sure where to focus on. We decided to make a 4-stage experiment that coincides with the checkpoints.

 

Procedure:

  1. Lasercut the mold design out of 1/4" thick acrylic for the top and bottom, and 1/8" thick acrylic for the middle piece

  2. Epoxy them together to create a seal while using bolts to keep them under pressure during the curing and tests

  3. Cut the carbon fiber into several lengths in order to see whwat range of carbon fiber would be able to be inserted into the mold. (Using 5 mm, 10 mm, 20 mm and 30 mm).

  4. Mix them into syrup (for future tests, we will be using epoxy/resin) at approximately 40% carbon to syrup mass ratio.

  5. Inject the solutions into the individual mold chambers using a syringe

  6. Take pictures and videos of the chambers during and after injection to record fiber orientations/flow

  7. If the fibers are not visible with a macro lens, put the solutions under a microscope.

  8. Determine the range of fiber lengths to test and problems to fix for future tests/next stages...

​

Materials Needed:

  • Sheet of 1/4" and 1/8" from Jacobs

  • Carbon fiber strands

  • Syrup (Sugar Syrup)

  • Sheet of rubber

  • 1/4-20 bolts

  • Epoxy/Resin (next stages)

  • Syringe

  • Cups

 

Equipment:

  • Laser-cutter

  • Camera

  • Microscope [Not used in the end]

  • Instron Machine [Not used in the end]

17820362_1372278622819936_1847605024_o
17836940_1372278569486608_720618522_o

- (Left) Diagram showing the carbon/resin solution being injected into an acrylic mold cavity using a syringe.
- (Right) Exploded view of the acrylic mold, showing it is made of three acrylic layers sealed and bolted together.

Stage 2 : Varying Viscosites While Improving Experiment Techniques

​

Objective: 
To decide which viscosity provide the range of epoxy viscosity and testing new mold and design techniques.

​

Procedure:

  1. Repeat Steps 1 - 2 from Stage 1 (Updated schematics of mold are uploaded at Checkpoint #2)

    1. The sheets are bolted together after adding adhesive (acrylic glue and 30 minute epoxy).

    2. Left to cure for 24 hours

    3. Note: The molds are more complex this time to account for some of the clogging issues from the last test. We wish to see how the different channels and chambers will affect the clogging problems

  2. Cut carbon fiber lengths â€‹into 10 mm (and 5 mm and 20 mm, which we we had leftovers of)

  3. Prepare syrup of varying viscosities

    1. Heat up syrup for 1 min, 3 min and 5 min over a stove. (anything higher will just solidify it)​

    2. As we hit each time checkpoint, scoop out some syrup to be poured into a container for the test.

    3. Mix well to get rid of air bubbles

  4. Mix the syrup and carbon until the carbon is as uniform in the syrup as possible by hand

    1. Amount of carbon in the syrup was done approximately by how viscous the new mixture became. This was not quantified because syrup and carbon mixtures are not homogeneous, which makes regular viscosity measurement tests inaccurate. We also had issues with shipment, where our steel bearing ball intended for viscosity testing was shipped to the wrong address. We did not have the ball available to test the viscosity of the syrup before the carbon was added.​

    2. Note: The final test will have measured mass or volume percent ratios. We were unable to apply them to these initial tests. These tests are intended to observe their behaviors and check what will be needed for the final test. They are not intended to quantify the orientation results yet.

  5. Fill squeeze bottles with the mixture. Make sure all the mixture is near the nozzle, rather than at the bottom of the bottle.

    1. Spun the squeeze bottles around in a shirt to move the mixture near the nozzle​

  6. Put the squeeze bottle nozzle down into the large hole of the designated chamber.

  7. Connect the hand vacuum pump to the small hole at the other end of the chamber.

  8. Squeeze the bottle and pump the vacuum pump as needed to draw the mixture through the chamber.

  9. Remove the bottle and pump to observe the results and record data.

  10. Observe fiber orientation and alignment and note any clumping

​​

Materials Needed:

  • Sheet of 1/4" and 1/8" acrylic from Jacobs

  • Carbon fiber strands

  • 1/4-20 bolts

  • Squeezebottle

  • Syrup

  • Pot/kitchen

  • Containers

 

Equipment:

  • Laser-cutter

  • Pump

  • Camera

Stage 3 : Data Collection of Mixtures with Varying Viscosity and Measuring Alignment 

​

Objective: 
Before we start using epoxy, we would like to collect data and generate a model for viscosity (using water, syrup and eventually epoxy) against the alignment of carbon fiber by comparing many fiber strands to a axis and measuring the angle.

​

Procedure:

  1. Repeat Steps 1 - 2 from Stage 1 (Updated schematics are at Checkpoint #3)

    1. The molds will be in the shape of a dog bone for instron testing. The dog-bone
      shape is so that when the two ends are clamped on, the failure point is for sure
      in middle of the sample and not where the clamps are holding on.

    2. The molds will have small holes around the dog-bone, which we will squeeze
      superglue in so that it will spread out and seal the dog-bone with a clean, clear
      bond.

    3. The sheets are bolted together and then superglue will be injected to holes to
      bond the bottom two layers together.

    4. Left to cure for 2 hours.

    5. Drill holes into side of acrylic to act as channels into the dog-bone chamber.
      (We decided to do horizontal flow instead of squeezing the mixture down

    6. These molds might be reused for Stage 4 (Actual epoxy tests)

  2. Cut carbon fiber into lengths of 20, 10 and 5 mm.

  3. Prepare syrup of varying viscosity:

    1. Heat up syrup for 1 min and 3 min over a stove. (anything higher will just solidify it)​

    2. As we hit each time checkpoint, scoop out some syrup to be poured into a container for the test.

    3. Mix well to get rid of air bubbles

  4. Perform viscosity test using steel ball bearings, and dropping it into the mixture. We would then measure the time it takes and distance to find average velocity and plug that into the viscosity equation to find the viscosity of each mixture.

  5. Create a standard of measurement possibly using a four quadrant spectrum that will help us create a correlation of fiber alignment by comparing each sample to one another.

    1. We thought of using some image analysis on MatLAB or etc., but it proved too complex and 3-D.​

  6. Mix the syrup and carbon until the carbon is as uniform in the syrup as possible by hand

    1. Amount of carbon in the syrup was done approximately by how viscous the new mixture became. 

    2. Looking around for a precision scale for us to do mass fractions mixtures, but will resort to sub-step 1 if there is none available at the time.

  7. Fill squeeze bottles with the mixture. Make sure all the mixture is near the nozzle, rather than at the bottom of the bottle.​

  8. Lock up the mold with the bolts.

  9. Put the squeeze bottle nozzle into the hole of the designated chamber.

  10. Remove the bottle and observe the results and record data.

  11. Observe fiber orientation and alignment and note any clumping

  12. Wash out the mold and prep for next stage (which will involve actual epoxy samples!

​​

Materials Needed:

  • Sheet of 1/4" and 1/8" acrylic from Jacobs

  • Carbon fiber strands

  • 1/4-20 bolts

  • Squeezebottle

  • Syrup

  • Pot/kitchen

  • Containers

  • Steel ball bearings

 

Equipment:

  • Laser-cutter

  • Pump

  • Camera

20170407_130549.jpg

Stage 4 : Testing Actual Carbon Fiber Dog-Bones

​

Objective:

Having analyzed the affects of carbon fiber lengths and viscosity in the previous stages, we now focused on quantifying the strengths of various carbon and resin samples. We were initially going to try to determine fiber alignment from images during out injection testing in the molds with syrup, but lack of quality images and constant variables has forced us to focus on fiber length versus strength. Although the correlation between carbon fiber lengths and strengths is well studied, injection molded samples may not follow the trend completely, due to several factors:

​

  • Fiber orientation in the part, changing the directional strength of the part

  • Low carbon to resin ratio compared to typical composite parts

  • Carbon's tendency to clump together, creating a significantly less uniform part

  • Part imperfections associated with the manufacturing method (in this case, injection molding)
     

For a less uniform part with clumped fibers, the areas with higher resin ratios would be much weaker. Because of these factors, a part with longer fibers may not result in a stronger part.

​

From the previous results, we decided to focus on 5 mm, 10 mm, and 15 mm carbon fiber samples, in addition to two reference cases: pure resin and full carbon fibers. The carbon fiber fiber parts would need to have the same carbon fiber to resin ratios. We decided to make 3 dogbone samples per case, due to limited resources. Dogbones are the standard sample shape used for Instron tensile tests.

​

To create the samples, the following procedures were followed:

  • Materials:

    • Carbon fibers from 0-90 twill weave​

    • West Systems 105 Resin and 205 Fast Hardener

    • Acrylic Molds from Checkpoint #3

    • Instron Model 5593

    • 5 Syringes, 60 mL

    • PAM cooking spray

    • Calipers

​

  1. Carbon fibers taken from a woven fabric were cut to the same length, 30.5 cm. Ten of these 30.5 cm fibers were cut into each sample length: 5 mm, 10 mm, 15 mm, and 10.16 cm (4 in, full mold length).

  2. Spray PAM over the molds and their covers.

  3. Two pumps of epoxy resin, approximately 30 mL, with no hardener is mixed with the 5 mm, 10 mm, and 15 mm samples as well as possible.

  4. Cap the larger mold mold and bolt it down tightly.

  5. Two pumps of epoxy hardener is added to the 5 mm solution, mixing for 1 minute as instructed on the epoxy containers.

  6. The mixture is placed in a syringe, and then injected evenly into 3 molds (10 mL capacity each)

  7. Repeat steps 5 and 6 for the 10 mm and 15 mm solution.

  8. The 10.16 cm fibers are laid in a separate, smaller mold.

  9. Four pumps of epoxy and epoxy hardener each are mixed together, and then used to fill the molds with the 10.16 cm fibers, and three empty molds (can be injected into the larger mold, or poured into the smaller mold because there are no fibers).

  10. Cap the smaller mold and bolt it down tightly.

  11. Wait 24 hours for the resin to cure.

  12. Remove the bolts and mold cover, and release the parts. The molds may be broken if necessary.

  13. Measure the dimensions of the samples with calipers in at least 3 different locations on the smaller portion of the dogbones.

  14. Place the sample in the Instron to test one by one.

  15. Record the data provided by the Instron.

 

ALTERNATIVE METHOD:

When doing the viscosity test for the epoxy resin, we realized the epoxy was much less viscous then our syrup samples. Because of this, we suspected there may be issues with injection such as clogging, clumped carbon, air pockets, etc. In the case that this happened, we would use this alternative method.

​

  1. Follow steps 1 to 3 from the previous method.

  2. Follow step 5 from the previous method.

  3. Pour the mixture into 3 different molds.

  4. Repeat steps 2 and 3 for the 10 mm and 15 mm solution.

  5. Follow steps 8 to 9, and 11 from the previous method.

  6. Release the parts. The molds may be broken if necessary.

  7. Post process the parts: sand and/or file the rough side until it has a uniform surface.

  8. Follow steps 13 to 16 from the previous method.

​

This method is not optimal because it no longer resembles an injection case, where there would be an orientation influenced by the injection method chosen. The data acquired would be for a randomized orientation. It also requires a large amount of post processing due to the open top face. However, this would enable us to acquire some data in the case that injection fails.​

 

MEASUREMENT UNCERTAINTY:

​

Once the data is acquired, the percent confidence interval needs to be calculated. This is because every form of measuring equipment has a certain amount of 'uncertainty', creating a potential range of actual values surrounding the measured valued. Generally, the more measurements are taken, the smaller the uncertainty is.

​

Because we are taking less than 30 measurements in this experiment, the uncertainty can be calculated using a Student's t-distribution, with the degrees of freedom determined by the number of measurements taken. We will use a 95% confidence interval, as in, there is a 95% change that the actual value is within the calculated range.

​

For caliper measurements:​

  • Standard error (SE) = Standard Deviation/√(#samples)

  • Uncertainty with 95% confidence = SE x t

    • t is taken by the Student's t-distribution, with degrees of freedom = number of samples - 1​

For the Instron measurements:

  • Error range is provided by the manufacturer in the specifications sheet: +/- 5% of the measured load

​

Generally, uncertainties should not have more than two significant figures unless given a specific reason to.​

Berkeley, CA, USA

©2017 by Carbon

bottom of page