Waste Plastic Pump

This instructable is focused on techniques for reusing / welding waste HDPE plastic. If you want to rework old bottles / containers and make things out of them, please keep reading.

The techniques are demonstrated by showing / describing how I built a 1.2 V NiCd AA battery powered pump which is chemically tolerant for use in circulating an acidic bath. I considered using ABS and PLA for their 3D Printed appeal, but the material would not have survived. Using a windshield washer motor was also considered but rejected since I have no way of knowing what material was used in its manufacture.

Although HDPE is notorious for its shrinkage when cooling, this instructable shows how it can be used to make something useful.

Most household cleaning products and non-transparent pill bottles are made of HDPE. It can be identified by its resin identification symbol, a triangle with a two inside. Once I found a decent container to use, I did the following:

1. Removed stickers with Goo Gone
2. Washed with Soap
3. Dried Overnight
4. Cut Continuous Square Sheets with Scissors
5. Saved Handle with Utility Knife
6. Cut Scraps into 1/4" (inch) Chips

This instructable only deals sheets from step #4
If you want to reuse the leftover chips from step #6, try:

• molded mouthpiece for blowgun by lennyb
• Making Blocks out of HDPE milk jugs by mfoster
• How to recycle HDPE (milk bottles and caps) into usable sheet material by Atomic Shrimp
• Waste Plastic Injector by vkoudymov

Tip: Don't use bottles used for smelly compounds. I used a laundry bottle and noticed that when heated, it emitted the smell of laundry liquid, despite having cleaned it. If you can obtain, hydrogen peroxide bottles are a good source of HDPE, they're also infrared blocking which is great if you want to use IR LEDs, most other HDPE is transparent to infrared.

In upcoming steps, I needed to cut out square strips.

To do so, I used a paper cutter, with the safety bar removed. To get a uniform strip width, I used a template to check whether the strip extended equally over either end of the cutting blade. Because of its thickness, HDPE tends to slide out when cut. To prevent this, the following steps were taken:

• Masking tape on three sides to secure the position
• Downward pressure on the HDPE
• Quick drop of the cutting blade

Lastly, to get a square edge at the end of a long strip, I taped on a right angle ruler to allow alignment.

In this step, I created a round pole. To do so, I used a wood burning iron which I temperature controlled using a dimmer. I set the temperature so that HDPE would melt without smoke. If you need to purchase such an iron, make sure that it includes a small tear-drop shaped tip since it would allow the following:

• Tack Welding: Press pointy tip between two pieces of plastic which fixes the position of two pieces
• Drag Welding: Slide tip side over/between two pieces of plastic which leaves a joined channel
• Fill Welding: Drag tip side over a 3^rd piece of thin plastic to join all three together
• Flat Welding: Press tip bottom onto a piece to secure it to the piece below

Because HDPE does not have the best heat transfer characteristics, it took about 40 minutes to complete this step for a piece 1.5" long. Using the above terminology, I took a square strip and tack-welded it onto base piece. I bent it over itself repeatedly and flat welded it into a cylinder.

In this step, I smoothed the cylinder. To do so, the rough cylinder was inserted into a drill press. A cutting edge, a piece of sharpened steel bar stock, was secured in a vice and aligned using a square block. The workpiece was placed into the drill chuck. The cutting edge was heated with a propane flame to make machining easier. With the drill press on, the cutting edge smoothed half the cylinder. The workpiece was flipped and the process repeated. Lastly, imperfections were sanded away.

I flattened the ends of the cylinder by dragging a utility knife blade, secured in a vice, across the bottom of the cylinder as it rotated. To mark the center point of the rod, I placed the point of a utility knife blade against the cylinder end as it rotated, which tends to push the point towards the true axis of rotation. Once marked, I held a drill, slightly smaller than the motor's shaft, underneath the cylinder, forcing it to drill a center hole.

I used a 1/2" copper coupling since it would handle heat well and was round. I wrapped a strip of HDPE around it and taped it down, exposing the edges. I fill-welded each of the seam edges and then took the tape off, fill-welding the middle. I flat welded the seam, in halves, applying pressure on half the seam while flat-welding the other half. When complete, this formed a ring.

Because the blades will be rotating quickly, I decided to repeat the previous step and added another layer. To prevent warping while heating, I wrapped the assembly in aluminum foil, making sure to outline the edges with my finger-nail. To merge the layers and keep the strips from returning to their original shape, I baked the assembly in a toaster oven for 20 minutes at 350°F. Once ready, I dropped the assembly into cold water and removed the foil.

I wasn't happy with the outside of the ring, it was bumpy so I wrapped the biggest drill bit that my drill press would accept in masking tape so that I could slide the copper coupling onto it, allowing me to sand the outside. In retrospect, had I been more careful with the aluminum foil, I may not have needed to do this.

Note: Because HDPE shrinks as it cools, getting the ring off was difficult. I used pliers.

I wanted to make equally sided blades to keep the motor balanced. I measured the outer diameter of the copper pipe to get an idea of the maximum size of the HDPE ring, allowing me to make 12 mm template strips using calipers, masking tape, and a utility knife.

I wanted to keep everything perpendicular so that my blades would be even. I used a square block to keep the ring perpendicular with respect to the blade. I measured each cut using the template strips from the previous step. Using the square block is a neat trick if you want to trim the blades later, in case they are too large or too slow.

I used a flat, scrap piece of 1/4" thick wood as a template/support for completing the blade assembly. I only wanted to have four blades for easier welding, so I drew two perpendicular, intersecting lines. I drilled a pilot hole on their intersection, and enlarged it so that it was slightly smaller than the width of the previously made cylinder. I needed to sand the cylinder some more before I was able to push it through the hole with only moderate force. Lastly, I popped the cylinder out high enough so that I could attach the blades.

I drew some arrows on the template to remember which way the motor rotates so that I could slope the blades away from the direction of rotation. I attached each blade to a point where a line came out using the following technique:

1. Tack-weld blade in three spots, top and bottom of both sides.
2. Position the blade edge with respect to its line.
3. Drag-weld the blade on both sides so fix position.
4. Fill-weld the drag-welded lines to increase durability.

Lastly, I popped the blade assembly out from the bottom side.

Which way should the motor rotate?: With a cheap motor like the one I have it generally doesn't matter, the brushes are low quality and the lifespan is short. Some motors have a preferred direction of rotation for best performance/life. To figure this out, connect a DC voltmeter to the motor terminals and spin the shaft of the motor in the direction that you think the motor should turn. If you are correct, you should get a positive voltage reading, otherwise reverse the probes. This is the correct polarity.

Why does that work?: Motors and generators are closely related. Motors can be used as generators and vice versa. The key difference is efficiency. Motors really suck at being generators and vice versa. When a motor is powered, the spinning generates a voltage called back-EMF. This has the same polarity as the voltage being applied. If you were to disconnect power momentarily, you would be able to measure this back-EMF. By spinning the shaft with your finger, you are creating back-EMF and using its polarity to determine how to apply power to the motor or the direction of spin. For my motor, spinning it generated 30 mV of back-EMF.

Interesting Fact: Back-EMF is directly proportional to speed. If you can measure it, you can use it in a closed loop system to control a motor's speed.

The motor that I used was from an inexpensive personal fan, powered by 2 x AA batteries. The case was made of zinc plated iron, so I needed to protect it. I created the outer case by wrapping a strip around a tube of my wife's favorite lipstick, which was the only item slightly smaller than the motor's diameter. I tack-welded the strip to keep it from unrolling. I drag-welded it first and finished with a fill-weld to keep it sealed from its environment. To make it less ugly, I flat-welded it.

To make the spiral casing fit the motor, I marked the inside with a thin permanent marker to know where to cut. I repeated this twice so that the spiral casing became a tube. Because the terminals were too big, I had to cut in notches to allow the casing to slide over them.

I drilled a hole to align with the motor's body. I put it against the motor, traced with permanent marker, cut, and welded it on using fill-welding. For the casing, fill-welding is necessary because it's easy to make tiny holes if only drag-welding is used.

For the rear cover, I drilled two holes, slightly smaller than the motor wires. I pulled the wires through, traced the plate, cut, and fill-welded. To complete closure, I fill-welded rectangular strips onto the terminals.

Note: It's hard to see, but the motor's positive terminal has a copper wire, whereas the negative terminal is tin-plated.

Using the same method as for blades, I created another ring. The ring was winder than the blades and 1.5 x the height. It ended up as a different color because I ran out of long green strips. I traced the top/bottom plates by securing the ring with mounting putty. I marked the center using a compass to create cross-hairs. The plate nearest the motor casing faceplate had a center-hole slightly larger than the blade shaft and an exhaust hole. The farther plate had a slightly larger hole to intake  fluid.

I secured the exhaust plate using fill welding and I added three legs, using only a mild tack-weld to keep the plate exhaust plate from warping. If it was to warp, it would interfere with the motor's rotation. I aligned the plate using masking tape. To align the legs, it was necessary to put the blade assembly onto the motor after attaching the legs and sliding the blade casing assembly over the motor casing assembly. Alignment of the motor was done by flat-welding one of the legs and powering the motor with a single 1.2 V battery. The other legs were flat-welded on only when they were positioned as to not interfere with motor rotation.

I fill-welded the blade casing shut.

I used the same technique as with the motor casing for the exhaust tube. I used a thin screw as the form. No marker was necessary for cutting because the tube was so small. I secured it to the motor using a light tack-weld followed by drag-welding to reduce the chances of warping the blade casing.

Note: Small leaks in the exhaust tube are okay because the pressure is not large and because it does not come into contact with the motor's iron-zinc body.

The cover shows how well the pump operates from 1.2 V using a NiCd AA battery. In this step, the supply voltage is nearly 6.0 V, nearly twice the rated limit, and water comes out much more quickly.

Warning: Do not use with flammable liquids, may present a fire danger.
Warning: Do not use in hot water environments, plastic becomes rubber-like
Warning: Do not exceed 1/2 way up pump's legs or fluid will destroy motor.

I look forward to seeing you use some of these plastic forming techniques in your own instructables.