Tuesday, January 6: Hydrogen Sensors

With the assembly of fuel cell being the main source of my stress right now, I did what I normally do and focus on another aspect. Good idea? Not really, but things still need to get done so at least I am getting something done.

What I worked on instead was the wiring of the two hydrogen sensors that will be positioned above the system to indicate a possible hydrogen leak when the system it is running. I already had the first one wired and working (see previous posts), but I needed to make the wires long enough to go through the PVC tube(s) that I constructed as holders. I also needed to wire the second sensor.

Hydrogen Sensor 1 before rewiring.

Hydrogen Sensor 1 before rewiring.

Above is the same image from one of the earlier posts about the hydrogen sensor. It shows the hydrogen sensor (now Hydrogen Sensor 1) wired to the Arduino ready to go. I had to take the picture above and basically extend the wires a few feet. The pictures below show the new wires ready to be soldered.

The long wires before they were soldered directly onto the sensor.

The long wires with shrink tubing (the white things) before they were soldered directly onto one of the sensors.

Wires about to be soldered onto one of the sensors.

Wires about to be soldered onto one of the sensors.

After both sensors had new wires soldered onto them, I used a heat gun to shrink the shrink tubing to assure that the connections did not touch.

IMG_1547 IMG_1549

IMG_1551 IMG_1553 IMG_1552


With the wiring done, I need to now figure out how to best wire them to the Arduino board so that they both work and are neatly wired.

– Harrison


Monday Update: January 5, 2015

And we are back!

Over the two weeks of Winter Break, I got a lot less done than I hoped. A lot less… Which means that I have a week and a half to finish EVERYTHING. So what exactly needs to be done?

Basically, almost all pieces of the project are in somewhat of a finished condition, meaning that I now have to put it all together. The fuel cell is in 30 some-odd parts that need to be assembled. The gas processing system is in a handful of separate parts that need to be connected, and then the entire project needs to be positioned and secured onto the base (which I have yet to get).

Once everything is on the base, I need to construct the case and hide the wires and do that sort of presentation/aesthetic stuff.

As simple as I may have made it sound, I have quite a bit of work and surely a few very late nights.

I’ll keep you posted!

– Harrison

Winter Break: Creating the Scrubbers


With all of the PVC and materials I needed, it was time to build the scrubbers.

All PVC for the Scrubbers and for the Hydrogen Sensor Tubes.

All PVC for the Scrubbers and for the Hydrogen Sensor Tubes.

I already had to necessary holes cut into the end caps, so the next step was to clean them so that no dirt mixes into the solution when they are completed, and to sand them to assure for better paint quality when I spray paint them.

Excited to paint, I went ahead and spray painted all of the pieces, which, in retrospect, was an almost totally useless task because, as I found out later, I needed to paint them all—except for the hydrogen sensor tubes—again after they were completed. Below is the set up for spray painting the pieces.

IMG_1358 IMG_1354


While the end caps were drying, I built the tow tubes that stick down almost all the way to the bottom of the scrubber. How these work is the gas enters the first gas port on the top (which will be one of the end caps with two holes as seen in the picture above), and travel down this tube into the solution, which will be a vinegar of sorts. At the bottom end of the tube is a PVC nipple with a metal mesh screen glued to the end so that as the gas exits the tube, it gets broken up into smaller bubbles, allowing for more gas to be cleaned as it travels up to the exit port on the top.

Here are the top caps for both scrubbers. Each has a gas entry port that is cemented to the plastic tube that has the PVC nipple and mesh screen at the other end. They both also have the exit gas port cemented into the exit hole.

Here are the top caps for both scrubbers. Each has a gas entry port that is cemented to the plastic tube that has the PVC nipple and mesh screen at the other end. They both also have the exit gas port cemented into the exit hole.

I have to say, working with PVC cement is a little messy. First, you have to use PVC primer that cleans the area before you cement two pieces of PVC together. Then you have to apply the PVC cement which is this clear, gelatinous, and pungent stuff that gets all over your fingers. Anyway… The next step was to prime and cement in the ports on the two bottom caps that connect to the refill tube. While all that was setting in, I took two extra 4″ end caps and made each scrubber a base to sit on. I had to do this because the bottom cap had the refill port and tube sticking out and could not stand. So I took the two end caps and cut off the end so that they were basically rings that the scrubbers could sit on so that the refill port and tube were suspended off of the ground. I also cut an inch wide section out of each ring for easy access to the refill ports.

Here is one scrubber upside-down showing the refill port sticking out of the bottom cap and the ring base it sits on.

Here is one scrubber upside-down showing the refill port sticking out of the bottom cap and the ring base it sits on.

I then cemented both rings bases to their respective bottom caps, as well as cementing the top caps to their respective bodies (the 4″ PVC tubing like the one in the picture above). After they dried, I cemented the top cap and body to the bottom cap and ring base, completing the scrubbers. Once they were dry, I taped over the gas ports and refill ports and painted them once more to give them a finished look.

Final painting of the two scrubbers as well as the finished tubes that hold the hydrogen sensors (in back).

Final painting of the two scrubbers as well as the finished tubes that hold the hydrogen sensors (in back).


– Harrison

ATo-Do List: Just About Everything That Needs to get Done

Here is my almost completed To-Do list, which is just missing some things that I am not sure if/when/how they will be done.

Work Plans

  • Disassemble all scrubber components
  • Sand, prime, and paint all PVC
    • Two scrubber bodies
    • Four or six end caps
    • Two 2′ long pipes for sensors
    • Two 45° elbows
    • Two 16″ long pipes for sensors
  • Take all six 90° elbows and prime the threads and the edge of the holes
  • Use the PVC cement and totally seal, gas-tight, the elbows in place (from the inside and the outside)
  • Take the two inside pipes and two nipples and prime and cement them in place
  • (Whenever the nickel mesh sheet comes, cut out circles and cement onto nipples)
  • With the nipples in place, prime and cement the other end of the pipes on to their respective ends of the 90° scrubbers
  • When the two other end caps arrive, cut off the tops so they are just rings
  • Cut out doors for the refill tubes to get out
  • Sand, prime, and paint the two caps
  • Prime and cement the “rings” to the bottom of each scrubber
  • (Maybe prime and cement the bottom caps to the scrubber bodies)
  • Attach all necessary tubes to their connectors
  • Attach all necessary valves and fittings
Hydrogen Sensors
  • Upload program to second hydrogen sensor
  • Find 4′ wires to attach the sensors through the tubes
  • Fit together the already painted pipes for the sensors
Display Case
  • Base
    • Take two 12″ x 12″ squares and attach them side-by-side to make a 24″ x 12″ rectangle
    • Attach to that another 12″ x 12″ square to make a 36″ x 12″ rectangle
    • Attach the new 36″ x 12″ piece to the just made 36″ x 12″ piece to make it 72″ x 12″
  • Top
    • Take the two 18″ x 36″ pieces and cut off 1.5″ across the tops, be very precise
    • Take one edge of the long sides of the now 16.5″ x 36″ pieces and shave cut along them to make the edges a 45° angle
    • Do the same with one edge of the two 1.5″ x 36″ pieces
    • Cement the two 16.5″ x 36″ pieces together side-by-side to make it 16.5″ x 72″
    • Take the chamfered edge of the two 1.5″ x 36″ pieces and cement them onto the non-chamfered edge of the 16.5″ x 36″ pieces
      • Also cement the sides to each other
      • If the 16.5″ x 36″ pieces were flat on the ground, the 1.5″ x 36″ pieces would be attached sticking up at a 45° angle inward
  • Back
    • Cement together the two 24″ x 36″ pieces to make a 24″ x 72″ piece
  • Completing the Case
    • Cement the chamfered edge of the top assembled piece to one edge of the assembled back piece
    • Cement the assembled back+top to the base by cementing the edge of the base to the face of the back+top
Fuel Cell
  • CNC drill the holes for the end plates
  • Line fuel cell components up with the nylon alignment rods
  • Cut out six 7″ x 7″ gaskets from the 1/32″ rubber
    • Carefully punch the four holes for the alignment rods in the rubber gaskets
    • The components should be tightly fitted together
      • Trim excess rubber sticking out past the graphite plates with exact-o
      • Find the gas port hole and punch it out
    • Once the excess is trimmed, take off the air plate so that one gasket is sitting on top of a serpentine flow patter, still aligned
    • Draw out the 4″ x 4″ square on to the gasket
    • Remove from the rods
    • Cut out the square with exact-o
      • Make sure the MEA fits nicely
    • Put back the gasket and make sure that it lines up with the plate and the MEA
  • Do that with four of the six gaskets
    • Make sure that each gasket is fitted and cut in its respective place and will be marked where it belongs
      • Two sandwich the first MEA, two sandwich the second MEA, and two go on the outside of the first and last graphite plates
  • Line up MEA inside of the gaskets
  • Punch corner holes for alignment
  • Put the rods through it so that the MEA is secure
    • Punch gas port hole
    • Tightly sandwich and trim off excess with exact-o
  • Do this with both MEA’s
  • With everything from the first to last graphite plate done, align all with the rods
  • Take the two gaskets that were fitted to the outside of the first and last graphite plates
    • Secure in place
  • With the two end gaskets in place, take the nickel sheet and measure about 7″ up and cut into two current collector plates
  • Line one up and each end, over the square rubber gasket
    • Using a pen, trace the nickel plate onto the gasket
    • Remove nickel plate and carefully cut along the lines with an exact-o
    • This should get rid of the middle 4″, top to bottom, of the gasket, leaving only 1″ on each side
  • Do for both sides
  • Place nickel plates where they belong
  • See how to make the excess inch look and function the best
  • Secure the plates in place
  • Take two 1/8″ gas connectors and cut off one side of each
  • Place into end plate
    • Must not go in further than the graphite plate
  • If fits well, cement both in place
  • Take nylon alignment rods and run the four of them through the end plates and use the wing nuts to tighten
  • Take the four support rods and connect the end plates
Positioning Components onto Base
  • Take electrolyzer, scrubbers, 6″ x 18″ plexiglass, and fuel cell, hydrogen sensors and position them all on the base
    • Mark down exactly where all things go
      • Take into account the possible addition of the power source box
      • Take into account the possible addition of and end product, be it a lightbulb or another electronic device
    • Make sure that the scrubbers and the  6″ x 18″ plexiglass are marked down exactly where they will be placed so that the gas processing system can be finished
Gas Processing System
  • Screw the 3/8″-barbed-tube-to-1/4″-threaded-male-pipe adapter into the air filter
  • Screw one end of the 1/4″ threaded male-to-male nipple into the air filter and the other end into the air regulator
  • Screw the 1/4″-threaded-male-pipe-to-1/8″-barbed-tube adapter into the other side of the air regulator
    • Use the pipe dope tape on all connections
  • Take the 6″ x 18″ plexiglass piece and mark exactly where the rest of the gas processing system will be placed
  • Once done…
    • Secure the 3-way valve onto the face of the plexiglass
    • Secure the check valve onto the face of the plexiglass
    • Secure the air filter plus air regulator combo onto the face of the plexiglass
  • Measure distance from the 3-way valve to the check valve, then again from the check valve to the adapter connected to the air filter
    • Cut and place both pieces out of 3/8″ ID tubing
  • Measure length needed to attach the scrubber port to the 3-way valve
    • Cut and place piece out of 3/8″ ID tubing
  • Measure length needed to attach the air regulator to the fuel cell
    • Cut and place piece out of 1/8″ ID tubing
  • Cut piece out of 3/8″ ID tubing and place onto the Oxygen port on the electrolyzer
    • Attach check valve to the end
There will be some more to follow!
– Harrison


Monday Update: December 14

With only a few days left to order parts, I made a final list of all parts I still need (at least that I am aware of). I drew a 1:1 scale diagram of what the final product will look like, base included. IMG_1335


With this diagram drawn I was able to put together parts list. Now I need to put together a to-do list so that I stay on track and finish in time!



Monday Update: Monday, December 8

Last week was something…

It started off as another week just woking on my project, to…

  • Being confused about how the fuel cell I was building would actually work
  • Realizing that my design, in fact, was just not going to workout for me
  • Accepting the fact that I had to redesign my project with only a few weeks left to go
  • Trying to understand how I could design it to make it work without totally starting over
  • Figuring out that I might be able to change it to a hydrogen/air fuel cell
  • Drawing diagram after diagram to fully conceptualize the new design
  • Designing all new templates to upload to the CNC machine
  • Realizing that I was not precise when I cut the graphite plates a few weeks ago
  • Taking that realization into account and continuing on with that now as a factor
  • Constructing brackets for the plates taking into account the previous impreciseness
  • Repurposing the already cut plates into the new configuration
  • Routing the air flow field onto the back of a repurposed plate for the bipolar plate
  • Routing the air flow field and drilling holes into the new plate for the air-only plate
  • Drawing a diagram of the entire system, now including the gas processing system
  • Creating an extensive parts list for the gas processing system and other hardware
  • Ordering that list
  • Writing blog posts
  • Sitting here, ready to keep moving

It definitely hit the lowest-low of the entire project to date, but I would say that right now I am better off because of it.

What I Plan to do This Week

Monday: Hopefully all of the parts for the gas processing system are in early enough Monday so I can work on them. What I first want to do is fire up the electrolyzer and get a feel for the electrolyte and the hydrogen production. With the electrolyzer working, I can start to assemble the gas processing system.

Tuesday: Go to the plumbing store to get the PVC pipe and end caps so that I can start building the scrubbers. While there, figure out the necessary air filter and regulator to implement into the system. Building the scrubbers should take no longer than a day or three to build, but I want to start Tuesday with at least the body (some stuff needs at least 24 hours to cement, so that gives me time to work on the other 1000 things I need to do).

Wednesday and Thursday: Continue on with the scrubbers. If scrubber work is going well, maybe finish them. If not, take a break and jump over to the fuel cell. I should figure out a way to place the garolite endplates on the CNC machine so that the holes line up, with the same imperfections, as the graphite plates. Once I decide on how to do it, I have to do it. The endplates should take about two days to finish.

Friday: Once the endplates are finished, I need to attend to the smaller components of the fuel cell: the gaskets, the current collectors, the MEAs. I basically need to punch holes in the right spot and cut the current collectors to the right size. This should hopefully be done in a day, maybe two. The scrubbers should be finished by Friday, or at least 90% done.

Weekend: Finish scrubber work and begin to put assemble the gas processing system. Connect the electrolyzer to the scrubbers with the appropriate tubing and fittings and see if it all flows well.

…And the Next Week (task are all variable due to extrapolating this far into the future ):

Finish putting together the entire gas processing system so that I can start to assemble the fuel cell. This should not be too difficult because all I really have to do (besides securing the current collectors and cementing in the two gas ports) is put the nylon rods through the four corner holes, then secure it all with the four steel support rods.

With the gas processing system completed and the fuel cell assembled, I should begin testing by the Following Week. I should be able to produce some amount of power before break starts, so that Over Break I can design and build the display.


– Harrison





Done with the Graphite Plates. And this time, I mean it: Friday, December 5

SIDE NOTE: All of the parts that I listed yesterday were ordered today and should arrive monday. I need to still figure out what air filter and air regulator to use, and to get PVC pipe and end caps. 

With the bipolar plate done, all I had to do now was to put in the new sheet of graphite and make the air-only plate. I set in in the brackets, uploaded the file that only had the holes programmed to cut, and hit start. The CNC did its thing and drilled the holes. For some reason, the two holes on the left side (top and bottom) did not go all the way through, so I just had to take a 1/8″ drill bit and finish them. I then flipped it horizontally (if I flipped vertically, the holes would not be in the right oder), loaded the air flow field file to the CNC, and hit start.

First two (out of many) lines of the air flow field. Note that the lines do not make it all the way through the end.

First two (out of many) lines of the air flow field. Note that the lines do not make it all the way through the end.

As it was about to finish the third to last line—after about ten minutes of it cutting and me vacuuming behind it—leaned over the machine to continue vacuuming, and I hit the kill switch. Uh oh. It stopped in its tracks. And that is bad news.

This machine, coupled with the program used to control it, does not make for the most sophisticated technology. Because I hit the kill switch, it lost the memory of what it had just cut, where it was turned off, but more importantly, it lost the information on where it was zeroed. If it knew where it was zeroed when it started the cut, I would have been able to do a number of things to manipulate the machine into finish the plate. But of course, it didn’t. So I had to zero the bit again, and there was no way I would ever be able to zero it exactly in the same spot as before. What I decided to do was to take the file with all of the vertical lines, and remove all but the last three. I set it so instead of starting with the right most line and cutting downwards, I set it to start from the left most line and cut upwards.

I was very tentative to cut this, but I went ahead and zeroed the bit as closely to where I zeroed it before, and hit start. It began cutting the left most line upwards, moved to the second line, cut that, and then moved to the third line. Only about a half of an inch up is where the new line should join with the old line, hopefully as straight as possible. As soon as the new line joined with the old line, I hit the kill switch. Guess what? It actually worked—and better yet, it doesn’t even look bad!

Now that I have all of the graphite cut and all of the materials on the way, the next step is to cut the holes into the garolite end plates. It is going to be Very tricky lining them up as perfectly imperfect with the graphite. That is my task for Monday!

– Harrison

Getting Back on Track: Friday, December 5

With all of the ideas conceptually set in stone, I needed to figure out exactly how I was going to do what I planned. I thought about the nature of the CNC machine and how I used it. The machine put those four holes in the graphite so that if I were to draw a line from the top left hole to the top right hole, it would be perfectly parallel to the horizontal axis of the machine. The new plates had to be cut so that they are all imperfect in the exact same way—when I put them all together, even though the holes are not squared with the plate itself, they will still line up so perfectly so that from the outside, it should look like there was never any issue.

Urban and I came up with a way to make this happen. I uploaded a file to the CNC machine that just had the four corner holes from the graphite plates, and told the machine to drill those holes directly into the wood base. These holes matched up with the holes on the graphite plate. I then took one of the plates I had already cut, and placed it on top of the base and lined up the holes. I put a nail in each corner, through the plate and into the wood, and hammered it down so that the graphite plate was sturdy in its place.

With the plate sturdy in place, I cut wood for the bracket. I put two pieces of wood on the base and pressed them up to the right and left sides, respectively. I then drilled them in place so that the graphite plate could not move side to side (if the nails were not there). I then took four small pieces of wood and placed them in the corners, two against the top edge and two against the bottom edge. I could not just put a piece of wood all the across the top and bottom because the air flow fields need to be cut the whole way down the plate and having wood on either end will interfere with the path.

New Brackets with open top and bottom. The plate in the brackets is the bipolar plate with the hydrogen serpentine flow field facing down, ready for the air flow filed to be milled

New Brackets with open top and bottom. The plate in the brackets is the bipolar plate with the hydrogen serpentine flow field facing down, ready for the air flow filed to be milled

Once the brackets were made, I removed the nails from the plate (which was much easier to take out with the new bracket design) and flipped it over so that I could cut the air flow field and make the first bipolar plate. I zeroed the axis, uploaded the file to the machine, and cut. It worked beautifully. Well, as beautifully as I could have made it. The flow fields are supposed to go all the way through the top and bottom so that air can get in and out.

Due to the fact that the left and right edges of the graphite plate itself were not perfectly parallel to the flow fields that the machine was milling, about half of the lines did not make it all the way through the bottom (all were open at the top). This just means I have to take a dremmel tool and finish them off. After I dremmel it and drill hole 6, this plate is ready to go!

Feeling a little better now! I’ll cut the other air plate later this afternoon.

– Harrison

Don’t Forget About Everything Else: Thursday, December 4

Now that I was ready to continue work on the fuel cell in the morning, I paused to think about the rest of my project. I needed to think through not just the components of the fuel cell, but to go through, in detail, everything that needs to be done for the entire system. How much needs to be done? Answer: A LOT. I went home and drew a diagram of the entire system, from the initial power source that supplies electricity to the electrolyzer, to all of the tubes and fittings that bring the usable hydrogen to the fuel cell, to the final mechanism that uses the electricity produced from the fuel cell. Below is the diagram. Put digram below

I needed to figure out exactly what I already have, what I don’t have and where to get it, and how to put it all together. I started off the drawing with the electrolyzer. I have the power source, the hydrogen sensor, and the electrolyzer, but after the electrolyzer it gets cloudy. Using another one of Phillip Hurley’s books, Build a Solar Powered Hydrogen Fuel Cell System, I began to piece together the gas processing system and how the hydrogen gets safely and efficiently from the electrolyzer to the fuel cell.

Screen Shot 2014-12-04 at 4.18.56 PM

Based off of his design, I drew up a system of my own. Hurley used parts from his local hardware store, but also gave an item for most of the parts on McMaster-Carr. I spent a solid number of hours looking at his design, reading his descriptions and parts list, matching items from the book to the McMaster-Carr website, and reading in both the book and on the website about each part and how it works, what it’s properties are, and its compatibility with my system and the chemicals involved. I also took this time to find exactly what parts I needed to complete the fuel cell (the gas entry ports, nylon rods and nuts for the four inside alignment holes, and steel rods and nuts for the outer support holes).

Side Note: When I ordered the Hydroxenator, I was under the impression that I would receive an electrolyzer identical to the one advertised in the videos. The Hydroxenator that was pictured had the electrolyzer unit, along with two scrubbers, presented in a handsome white casing. What I received, however, was just the electrolyzer unit and a small black metal stand. Now, I do not want to sound ungrateful at all, because the man who builds the Hydroxenators sent me one for only the price of the materials and machining, it is just that it was unclear exactly what I would be receiving. What this means now, is that I have to build my own scrubbers to clean the hydrogen gas, something that I was not planning on doing. So I had to make a parts list for the scrubbers as well, based, of course, on Hurley’s design.

Here is the list of parts:

See Plumbing Store
Gas Processing System Parts
———McMaster Carr———
  • 2ft. length of 1/2″ ID extreme-purity clear PFA tubing (5773K17)
  • 5ft. length of 3/8″ ID extreme-purity clear PFA tubing (5773K16)
  • x1 (pack of 5) 1/2″ x 3/8″ brass barbed reducing couplings (91355K33)
  • x2 3/8″ barb PVC three-way-ball-valve (4757K58)
  • x2 3/8” barb PVC single-ball-valve (4757K18)
  • x2 3/8″ barb PVC check valve, Buna-N Seal (7933K33)
  • x1 (pack of 5) 3/8″ brass barbed hose fitting (91355K84)
  • x1 3/8″ pipe size Clear air filter, manual drain, max 74 scfm @100 PSI (4959K52)
  • *Need Air Regulator*
  • x1 roll, 42 feet length x 3/8″ width x 0.0032″ thick, pipe thread sealant tape (6802K19)
Fuel Cell Hardware
  • x1 440 (.112″) 4ft long threaded Nylon Rod (91839)
  • x1 threaded 440 (.112″) 100 pack nylon wing nut (91811)
———McMaster Carr———
  • x1 (pack of 10) 1/4″ D x 6″ L threaded zinc plated steel rod (95475A556)
  • x1 (pack of 100) 1/4″ grade 8 steel hex nuts (90499A029)
  • x1 12 pack 1/8″ nylon hose splicer
    • ORDERED Thursday, 12/4/
Scrubber Parts
———McMaster Carr———
  • x6 1/2″ pipe x 3/8″ MPT single-barb tube fittings 90° elbow (53055K191)
  • 1ft 1″ ID extreme-purity clear PFA tubing (5773K2)
  • x2 1ft length 3″ ID x 4″ OD Type 1 thick-wall gray PVC pipe (8749K25)
  • x4 Pipe size 3 thick-wall gray PVC unthreaded end caps (4881K57)
  • x2 1/2 Pipe size x 3″ length thick-wall gray PVC threaded-end pip nipple (4882K43)
  • x1 Nickel Alloy (400) 10 x 10 mesh screen, .025″ wire, 6″ x 6″ sheet (9225T354)
  • x1 16oz can PVC pipe cement (74605A14)

Work Plans For Friday: Thursday, December 4

Work Plans

Friday Morning:
  • Go right down to the ShopBot
    • Drill the four alignment holes into the wood base
    • Nail in a graphite plate
    • Take wood and make the brackets around the nailed in plate
    • Make the bracket not go all the way across the top and the bottom of the plate so that the machine can mill the air flow field all the way through
    • Take out plate
  • Open all the tool paths for the new cuts and match the plunge and feed rates to the values used for the already cut plates
  • Put the plate back in, but facing down as to mill the air flow field on the back
    • This is the bipolar plate.
    • Drill by hand hole 6
  • If the air flow field cut works perfectly the first time, swap in the new plate and cut the last air plate flow field
  • If the previous steps go smoothly, I should end up with the three graphite plates ready to go

Let’s hope it all works out in the morning!

– Harrison

Let the Redesigning Begin: Thursday, December 4

I needed to slow down, pause for a moment, and make sure that the next moves I make are thought through and certain. No more set backs. I drew a diagram for a two-slice convection fuel cell stack that I now plan to build. I tried to make very clear to myself exactly how the components will work together for this fuel cell to produce electricity. No more basing my fuel cell exactly on Hurley’s design without checking everything.

To try to save money and time, I looked at the two graphite plates that were originally cut to be the first hydrogen plate and the first oxygen plate. These were identical and had the four corner holes for alignment, a serpentine flow field with a hole at the bottom tale and a hole going floating off to the side. I though about that possibility of repurposing these plates, and I realized that it was very possible to do so. One could still be used as the first hydrogen plate, the only difference is that hole floating off to the side was useless and would have to be sealed up. The second plate could be used as the hydrogen side bipolar plate, with the floating hole also sealed. That meant that because I had one untouched plate left I did not have to order more material!


The diagram above shows how each plate will be cut, making note of the fact that the holes are imperfect. After realizing yesterday that the way the two plates that were cut made for the holes in the plate to not be level with the square they were cut into, I had to think very hard about how I could continue building the fuel cell with this as a factor.

Screen Shot 2014-12-07 at 1.06.16 PM


  • The first hydrogen graphite plate (second square from the left) and the hydrogen side of the bipolar plate (fourth from the left) were the two plates I already had cut. The numbers in red correspond to the holes in the plate. Looking at the first hydrogen graphite plate, you can see that the holes are numbered 1, 2, 3, 4, and 5, 1-4 being the alignment holes and 5 being the hole for the hydrogen. This was cut facing up. Now look at the first blue square, the hydrogen end plate (the endplates are not drawn to scale, as they are 8” x 8” squares, not 6” x 6” squares like the others). Imagine placing the first hydrogen graphite plate directly on top of the end plate. See how the numbers line up? 1 is on top of 1, 2 on top of 2, and so on.

Screen Shot 2014-12-07 at 1.06.24 PM

  • Now look at the bipolar plate (the third and fourth blue squares from the left). It is hard to conceptualize, but those two squares are either side of the same graphite plate. I already had the serpentine flow field, as well as holes 1-5, cut into one side of the plate. Hole 6 I will drill by hand after both sides are finished. Because the air flow field (the vertical lines) are on the other side, I have to flip the hydrogen side face down to cut the air side. That means that holes will be in a different arrangement. You can see that on the air side in the diagram, as I labeled the holes differently (note that holes 3 and 4 on the air side should be switched). The holes will already be there when I cut the air side, so all I have to do is cut the flow field.

Screen Shot 2014-12-07 at 1.06.29 PM

  • Cutting the last air plate (the furthest blue square from the left) out of the brand new graphite plate will be a two step process. To keep the pattern of the holes correct, I have to cut the holes out from the reverse side, flip it over, and then cut the flow field.

The big question now is how do I actually make it so that all of the holes line up when I cut it with the CNC machine.

I thought about the nature of the CNC machine and how I used it. The machine put those four holes in the graphite so that if I were to draw a line from the top left hole to the top right hole, it would be perfectly parallel to the horizontal axis of the machine. I need to figure out a way to cut the new plates so that they are all imperfect in the exact same way—when I put them all together, even though the holes are not squared with the plate itself, they will still line up so perfectly so that from the outside, it should look like there was never any issue.

Urban and I came up with a way to make this happen. I will upload a file to the CNC machine that just has the four corner holes from the graphite plates, and tell the machine to drill those holes directly into the wood base. These holes will match up with the holes on the graphite plate. I will then take one of the plates I have already cut and place it on top of the base so that the four holes match up. Then, I’ll put a nail in each corner—through the hole in the plate and into the hole in the wood—and hammer them down so that the graphite plate is sturdy in its place.

– Harrison

New Ideas — Hydrogen/Air: Wednesday, December 3

I needed to not freak out. I need to come up with a way to salvage my project and keep my hopes up that I will have a working finished product by January 14. So I went back to my research, and what I quickly decided was to change my design from the more complicated hydrogen/oxygen version to the hydrogen/air version. Why? Because it is much more simple of a design. Instead of feeding the cell directly with pure oxygen, which means gas lines, serpentine flow fields, and an unknown exhaust system, the hydrogen/air fuel cell draws ambient oxygen from the air by having a number of milled lines that go from the bottom to the top ADD MEDIA. Another name for this cell is a “convection” cell. This is because as oxygen is drawn up into the cell, it “works” and heats up. As it heats up, the air in the cell rises out and draws more in. This convection process keeps the cell fed with ambient oxygen. Because the oxygen is from the air, there is no need for an oxygen gas port and the corresponding holes throughout the cell.

It is SO MUCH easier to build. I still didn’t understand exactly how the gas flowed through, so I drew a gas flow diagram to understand it—just like I tried to do with the hydrogen/oxygen one—and thank my lucky stars, I actually understood it! The arrows follow the flow of hydrogen gas through the cell.

IMG_1287So now that I ACTUALLY UNDERSTOOD HOW THIS DESIGN WORKED (unlike the hydrogen/oxygen one), I could plan building it.

– Harrison


Uh Oh… HELP! My project has MAJOR issues: Tuesday, December 2

My design for the fuel cell is based on Phillip Hurley’s book, Build Your Own Fuel Cells, that I have been mentioning throughout my process. His book is not really meant for a DIY project, but more as a very detailed description of how he was able to do it, not how you are supposed to do it. But of course with research and heavy caution and supervision, I used Hurley’s guidelines and materials to plan my own fuel cell.

In his book, there are a few different types of fuel cells that he builds (and I have talked about this before). I decided to build a hard graphite hydrogen/oxygen PEM fuel cell. The other option that I could have gone with was a hard graphite hydrogen/air PEM fuel cell. Now the only difference between these two fuel cells is that in the hydrogen/oxygen fuel cell, the oxygen is pure oxygen supplied through gas lines directly from the electrolyzer to the fuel cell. This requires the graphite flow fields be milled in serpentine pattern just like the hydrogen side. The reason that I chose this design was because a) it is 30% more efficient than the hydrogen/air fuel cell, which means b) I could output more power with the less materials a.k.a c) less money, helping prove my point that d) fuel cells could be relevant in the near future. But today, I hit a MAJOR roadblock…

I had already cut the two endplates and was about to get ready to mill out the bipolar plate (the graphite plate in the middle that has a hydrogen flow field on one side and an oxygen flow field on the other), but I decided to take a second and actually think about how the two gasses would flow through the cell. And what I realized is that I had no idea how it happens, and there was absolutely no description of the gas flow in the book. My thought was that, given that there are two gas ports on each of the end plates, that the hydrogen went in one side and out the other, and the oxygen went in the opposite side and out of the other. So I drew a detailed diagram of the gas flow, and guess what… it made NO sense. I even asked my mentor(s) Dr. Morgan and Urban, both of whom are knowledgable scientists in their own right, and they were just as confused as I was. Did I really just spend months designing, planning, and building a fuel cell that simply made no sense? Well, I had to find out.

What I did was print out on paper, the mini templates that were given in the book (which had no instructions attached to them, really) and printed them out. I cut out cardboard and glued the templates onto the cardboard, cut the gas port holes, and put it together to make a mini cardboard fuel cell.  I took wire, one color for the hydrogen gas, another for the oxygen gas, and ran them through their respective holes and along their alleged paths to see if I could makes sense of it. And, well, it just didn’t work.


So I rightfully freaked out.

And the best part was that not only did I have to go back to the drawing board to rethink my entire fuel cell, I realized that the way that they plates were cut with the CNC machine (or should I say the way cut them on the CNC machine) was totally not though through. Major miscalculation on my part. If you remember back to when I was cutting the graphite, I made jerry-rigged wooden brackets to hold the plates in place while I was cutting them. This made total sense at the time, so I cut the two end plates just like that, in the same spot with the same bracket. But what I totally did not take into account was that the actual plate needed to be perfectly lined up with the machine. Sure, I eyeballed it and set the square plate pretty well lined up with the x and y axes of the machine, but of course, it was not perfect.

What this means to my project is that I was cutting perfectly lined up holes out of imperfectly lined up plates. Its like drawing a perfectly level square inside an imperfectly level square.

Screen Shot 2014-12-07 at 11.28.16 AM

So I have something that looks a little like that. The outside square is the graphite plate, the inside square represents the way the CNC cut it; the machine is damn near perfect, so the corners of the inside square are where the holes are. This means that if I had two plates, one flipped around, the holes will line up perfectly, but the edge of the plate won’t. I really should have thought that one through.

I was now left with a flawed design and two already cut plates that, because I did not line them up, were maybe useless. What do I do?!

– Harrison

Monday Update: December 1st

Only 6 weeks left until final approval!

What I need to do in the next week or two:

For the Fuel Cell:

  • Drill the holes into the end plates
  • Cut two electrodes out of the nickel sheet (each 4″ by about 7″)
  • Re-cut the 4 gaskets that line the electrode (two per electrode)
  • Re-cut the 2 gaskets that sandwich the MEA (6″ square with a 4″ square cut out of the middle)
    • Why? Because I want to make them in respect to the endplates, now that I have them, and make sure everything lines up perfectly
  • Find/buy the right fasteners and support rods: the nuts and bolts of the Fuel Cell

For the Electrolyzer:

  • Once the fuel cell is completed, all I have to do for the electrolyzer is
  • screw the gas ports in their respective holes,
  • fill it up with the electrolyte solution,
  • attach tubing from the gas port connections to the fuel cell
  • and hook it up to a power source

For the final presentation of the entire system

  • This happens when the fuel cell is done and has been tested with the Hydroxenator and produces usable electricity
  • I need to build a display case to show all of the components, including the power source, the electrolyzer, the fuel cell, the hydrogen sensor, and whatever it is I decide to use to show that the system works
  • Also need to write a very brief description of what each component is (what, why, how)
  • As of now, I see it as a plexiglass almost diorama casing with the system inside and the sensor mounted to the top

Monday Update: November 17

What I’ve Done Since Last Week:

Monday, Tuesday, Wednesday: I spent the beginning of the week preparing the 8″ endplate file as well as the 5″ endplate file for practice. I then research the garolite—referred to as FR-4 G10, or just G10—and figuring out how to go about fabricating my endplates. After a few hours, I found a company that I could buy a 12″ wide by 24″ long by 1/4″ thick sheet of black coated G10 (which was not easy to find) that they would cut the two 8″ squares for me. It was a great find, and it cost just as much as an uncut 12″ square sheet from the other stores.

Thursday: I used the 5″ endplate design to practice cutting two endplates for the 4″ graphite sheets. This was just to test how the bit cut the material. It cut very well, and I can use these endplates and the 4″ graphite as an exploded view of what a fuel cell is. Now I just have to wait for the 8″ sheets!

What I Plan to Do This Week:

The week: I am waiting for the garolite to arrive so I can drill the holes and finish the endplates. Until then and after, I will start to gather all of the components and start putting on the final touches—cutting the nickel alloy foil, making sure the gaskets fit and cutting the two squares out of the big gaskets for the MEA, cutting the excess plastic around the MEA and punching holes in it, and finally organizing the fasteners, support rods, and gas tubes.

Hope all goes according to plan!

– Harrison

Cutting the Endplates: Thursday, November 13

With the pre-cut 8″ endplates on their way, I practiced cutting the endplates with the garolite that I already had. I quickly made a 5″ endplate for the 4″ graphite plates (just like I  made an 8″ end plate for the 6″ graphite). I possibly could have made the 5″ endplates 5.75″ max, which would have avoided the fact that the very edges of the graphite plate cover part of the rod holes, which would be an issue for the big ones, bit since I am not going to build two cells, it’s no biggy. I was only cutting these plates to test how the drill bit did with the material, and it was totally fine. So great! Now when I get the 8″ squares, all I have to do is drill the holes. Easy. Here are some pics!

End plate for the 4" graphite. Furthest holes are 1/4" for the support rods. Inner holes are 1/8" for the fasteners and the gas ports.

End plate for the 4″ graphite. Furthest holes are 1/4″ for the support rods. Inner holes are 1/8″ for the fasteners and the gas ports.

Endplate with the graphite plate positioned where it would be placed (there would be a thin piece of metal with gaskets in between the two plates if it were finished). Notice that the inner holes all line up. Also notice the calculation error: corners of the graphite overlap the rod holes.

Endplate with the graphite plate positioned where it would be placed (there would be a thin piece of metal with gaskets in between the two plates if it were finished). Notice that the inner holes all line up. Also notice the calculation error: corners of the graphite overlap the rod holes.


Redesigning Endplates: Wednesday, November 12

Redesigning the endplates took about 3 seconds. All I had to do was make a new file on Illustrator with 8″ x 8″ dimensions, copy and past the the drill holes from the 6″ x 6″ into the middle of the new file, and add four 1/4″ holes in the corner for the support rods. Super easy.

Now that I had that ready to go, I figured that I might need the material that it was going to be cut from. Duh. I looked up the 12″ square sheet of garolite on Fuel Cell Store to order another. I was totally shocked to find that the the piece I ordered (12″ x 12″ x .250″) was $93!!! I didn’t want to order another because it was crazy expensive and I only needed an 8″ square from it. I searched the internet for garolite, which is much more commonly called G10 FR4, and found out that every other site that sold this material had it priced much more reasonably.

Hold on… what is this stuff?  As Atlas Fibre Company describes it, this material is a “continuous woven glass fabric laminated with an epoxy resin. This grade [G-10] is extremely high in mechanical strength, has low water absorption and dissipation factors and has superior electrical characteristics, which are exhibited over a wide range of temperatures and humidities. Grade G-10 is used for terminal boards, washers, sleeves, structural components, and parts where the strength to size ratio is critical.”

g10 black phenolic sheet half inch g10 natural phenolic sheet half inch









It typically comes in its natural color (on the right), which is greenish, and black or blue. I ordered the black because I think it’ll look better! I went online and placed an order for a 12″ x 24″ x .250″ and the company, Ridout Plastics, said that they would cut two 8″ squares from the sheet for me! Yay! And they only charge $18 for the custom cut plus the original sheet, which was $53.58. Total cost was almost exactly $100 with shipping. Compared to ordered two 12″ square sheets from Fuel Cell Store which I would then have to cut myself, I save $100 plus time. SWEET!

I am going to practice cutting on the sheet I have so that I get a feel for the bit and the material!


Graphite Plate 2/2: Monday, November 10

You have all heard about these graphite plates way too much and I am sure you are all sick of them, but I just have to mention that I cut the second plate today. This means that the graphite plates are done and ready to go, so I can focus on other components. Next is the endplates! And yes, you will soon get sick of hearing about the endplates just like the graphite plates!

– Harrison

Monday Update: November, 10

Happy Monday!

This week has produced probably the most physical progress to date.

What I’ve Done Since Last Week:

Monday: With one 4″ x 4″ plate tested and cut, I started off in the morning by cutting the other 4″ x 4″ plate just to build up my confidence before I cut the 6″ x 6″ plate that afternoon. I built the wooden bracket to hold the plate in place, zeroed the axes (x and y are always done manually while the z—up/down—axes can be zeroed more precisely with the metal zeroing plate), and I started to cut the piece. What happened was the bit went way too deep and cut all the way through, ruining the plate. I figured out that it went too deep because for some reason the zeroing plate zeroed the z-axis totally wrong.

Tuesday: Knowing that the zeroing plate was the reason that it went too deep, I decided to zero the z-axis manually. I put a new plate in the bracket, and very carefully and as exact as I could get, zeroed all of the axes. This worked perfectly, and I had my first (of two) 6″ x 6″ graphite plates.

Wednesday: I only ordered two 6″ x 6″ plates, and since one was ruined, I had to order more. So in the mean time I had to figure out something to do. I secured the endplate material on the base of the CNC, but I did not have the proper bit to cut hard plastics, which I ordered as well. I decided to pull out the rubber for the gaskets and cut them to size. I measure and cut four .9″ x 6″ pieces that go on either side of the 4″ x 6″ piece of metal (I’ve yet to discuss this, so don’t worry if you have no clue what I’m talking about) that sits between the endplate and the graphite plate, and two 6″ x 6″ pieces that I will cut a 4″ square from the middle that sandwiches the MEA. I needed to make 1/8″ holes in the rubber, so I ordered a leather puncher that can do the job.

Thursday: I was actually home sick today, but neither the graphite plates or the hard plastic drill bit had come yet, so I spent my time catching up on blogging.

Friday: In the morning, I quickly punched out the holes in the gaskets with the leather puncher. Though the both the drill bit and the graphite plates had arrived, I thought it would be better to start with those on Monday. So I began the side task of coding and wiring a hydrogen sensor that will come in handy while testing to make sure there are no leaks and for my final product as a safety feature. I coded it, set warning limits that beep obnoxiously if the hydrogen concentration in the air (measured in parts per million) gets too high. That was started and completed in the afternoon and evening.


What I Plan to Do This Week:

Monday: Now that I have the new graphite plates, I can definitely cut the other 6″ x 6″ relatively quickly, and if for some reason it goes wrong, I have a backup. So the second plate should be done by Monday.

Tuesday: What I have yet to blog about is that I plan to change the design of the endplates. Right now, they are 6″ x 6″ so they are exactly the same size as all of the other components. What I want to do is actually make the endplates bigger, say 8″ by 8″ inches for three reasons. One, it gives an inch on all sides so that when it is set on a table, or god forbid it was dropped, all of the inner, more fragile components will be safe. Secondly, the metal plates that collect the current has to have a little piece sticking out so it can be connected to a circuit. It would be exposed if the endplates were 6″ x 6″, but if they are 8″ x 8″ it would provide protection for this piece. And thirdly, with the original 6″ x 6″ endplates, there is a hole in each corner that goes through the entire cell that will be used to fasten the cell together. This is okay, but If the endplates were 8″ x 8″, I would keep the original fastener holes that are for the 6″ x 6″ plate, as well as add four support rods in the corner of the 8″ x 8″ that would just go from endplate to end plate, making the entire cell more secure while also adding more protection for the inside components. And fourthly… I think it’ll look way cooler. So I will start redesigning the endplates Tuesday and hopefully finish and have the new version ready to go by Thursday.

Wednesday: Continue redesigning the endplates as well as figure out if/what materials I need to order and order them—I am pretty sure that I will have to order another sheet of garolite because I only have 12″ square which means I can only get one endplate out of it.

Thursday: Hopefully by Thursday, I will be done redesigning and have ordered all of the parts. If I am done, the I might be able to cut out one endplate! If not, Friday.

Friday: By the end of Friday, I should have one plate done and all of the necessary materials en route so come Monday, I can quickly do the other one.

Hope all goes according to plan!

– Harrison

Hydrogen Sensor Part 2: Friday, November 7

Quickly, here is the updated code. It starts beeping when the value reaches 4,000ppm. The frequency of the beep is at 262 Hz and the beep duration is 150ms with a 200ms delay. It is a less high pitched, slower beep (though it is very high pitched and fast) and gives the message “WARNING! HIGH HYDROGEN LEVELS!” This beep happens from 4,000ppm to 8,000ppm. When the value hits 8,000ppm—and beyond—it beeps at a frequency of 500 Hz with a 150ms duration and a 100ms delay, with the message “DANGER! SHUT OFF IMMEDIATELY!” This is a much higher pitched and faster beep. It’s pretty annoying! Here is the updated code on Arduino.

Screen Shot 2014-11-07 at 5.11.55 PM

Here is the code to copy, if you want! Just copy and past into Arduino!

– Harrison

int sensorValue;

void setup()
Serial.begin(9600); // sets the serial port to 9600

void loop()
sensorValue = analogRead(0); // read analog input pin 0
Serial.println(sensorValue, DEC); // prints the value read

if((sensorValue>4000)&&(sensorValue<8000)) {
tone(9,262,150); delay(200);
tone(9,500,150); delay(100);
delay(100); // wait 100ms for next reading