Monthly Archives: July 2010

Thank you! 202%

Thank you so much for your overwhelming support for the OpenPCR project. We raised 202% of our goal, thanks to the massive support of 158 people that contributed financially and the hundreds of people that helped get the word out on Twitter, Facebook, blogs, and the nightly news!

One point I want to highlight — there are other biotech innovators like Josh and I and they need your support! Josh and I are part of a group called BioCurious, a community biotech lab starting up in San Francisco. BioCurious is trying to raise $30,000 in order to open their doors to dozens of innovators. As a supporter for BioCurious, you will be helping to start a major innovation engine, spurring education, public understanding, and entrepreneurship in biotech. Check out BioCurious at:

And, if you’re in the Bay Area, come by for a meetup:


Tito and Josh

I hate glue (feat. Nodes)

New cases arrived this week, acrylic and wood!

Nodes rock!

That is, when you design them correctly. Working with birch wood, I added 1 simple node on each of the notches for the lid parts. The parts “snap” together, and glue doesn’t even seem necessary.

With 1/4″ acrylic, I used the exact same sized node as with the birch which didn’t work out so well. Acrylic is a lot less flexible than wood. I ended up just sanding off the nodes on the acrylic parts to get them to fit together correctly. (I’ll read up on what size nodes to use with acrylic, or comments are appreciated!)

This was my first time designing a case and immediately having it made with 2 different materials. The design was for birch, and I attempted to change the design for acrylic which is a good bit thicker. I made the holes for the notches wider which was a good first step. What I skipped was making the parts themselves a bit shorter to accommodate for the thicker material.

One other revision to make is 4 sets of holes/T slots don’t line up perfectly. I’m not sure why they got a bit off (Update: the tooth on the bottom edge was a few mm off, which pushed the holes off), but it’s definitely something to fix. I hacked the birch case to get everything to fit.

Down to business, the T-slot design is sturdy, and assembly is straightforward. With the T-slow and notched teeth, glue will no longer be necessary! I also found that rounded corners on laser cut wood make a big difference in durability. In my experience, sharp corners easily chip, wear down, and start to look less than best. Rounded corners are more durable and can take a bit more wear and tear.

Next up is the important step — when the flat aluminum parts arrive from Ponoko NZ we’ll test out the heated lid.

Both the acrylic and birch look great! Thanks Pololu and Ponoko :)

Inspired by:

Treat your machine shop professionally

While searching for machine shops — quoted from:

If you want the best out of your machine shop, it’s important to treat them well. People like to do business with people they like, and it stands to reason that favorite customers get special attention. Here are a few do’s and don’ts regarding machine shop proprieties:

DO communicate regularly with your shop, call them if you perceive a problem. Don’t let problems simmer. Do work with them to resolve issues fairly. Cutting metal to high tolerances is an extremely difficult art, and some mistakes are bound to happen.

DO consult with your shop on design issues. Unless you’ve spent years of your life working as a machinist, don’t assume you know more about manufacturing issues than they do. Virtually all professional machinists have many years of training. Many have advanced degrees. Treat them as peers, not as subordinates.

DO pay your bills on time. Nearly all shops pay close attention to this. There is no question that fast paying customers get treated better.

DON’T use your shop as a free quoting service to scope out the cost of proposed projects. Quotes are expensive for a shop. Only request quotes for jobs the shop has a fair chance to getting. Don’t quote out jobs to more than 2 or 3 shops. Don’t expect a shop to continue to quote multiple jobs without winning some work.

DON’T try to beat a shop down on part costs. Do ask design advice on how part costs can be reduced. Treat your shop as part of your manufacturing team, and foster communication between design and manufacturing.

– PS, to the machine shops we’ve worked with already, thank you!

Update: 9/25/2010

A clarification. I want to point out that this element is important and it can be challenging to do! If you can’t meet in person, it’s tougher to build repertoire over email or telephone no matter how charismatic you are. And, though I’m usually close to my email, people in shops are on tools and machines most of the time. A phone call is a lot quicker and more personal, but conversations on the phone often lead to miscommunications and frustration. What’s worked for me is talking on the phone, and then summarizing in a follow-up email immediately afterwards. Other points of friction I’ve found are in file formats (I was in Bangalore, India, bouncing between coffee shops and I just couldn’t get the right program in order) and computer use  (for instance a shop that insists on 2D drawings of a nice CAD file you struggled to put together…).

Experience says: lavishly praise shops you come across when everything goes smoothly.

Right now: I’m working with an shop called OharaRP (out of Dayton, Ohio) on a PCB board and they kick ass!


Cutting flat aluminum parts

If you think I’m making a lot of posts over the past week, it’s true! I took 2 weeks off of my day job and have been able to get a lot done on OpenPCR. Back to my day job on Monday :\

We are prototyping 3 1/8″ flat aluminum parts for OpenPCR:

1. ~ 2″ square plate, with holes for 2 screws for the heated lid. Our prototype heated lid involves a peltier, aluminum plate, and thermal pad

2. ~ 4.25″ square plate for mounting the heat sink and aluminum heat block to one surface. This needs holes for the aluminum block and screw holes for mounting the aluminum block and heat sink.

3. 4 cm square square plate to serve as an “adapter” between smaller heat sinks and the peltier. This might not be necessary if we go with a larger heatsink, though the trade off would be the entire case is about 1″ bigger to accommodate the heatsink.

Designing the part and drawing squares in Illustrator? Easy!

Unlike software, when you want to test a design you get out your wallet. I’m addicted to “Just hit ‘Preview’ or ‘Compile’ and I’ll be done!”. I’m so thankful for the support that we’re receiving from everyone on Kickstarter!

I looked around locally and online  for a shop to make the test pieces. Local prototyping didn’t move as quickly as I had hoped, I still haven’t heard back from the guy I started talking to a week ago. Prototyping shops are pretty pricey, I found rates of about $100 per part. With flat pieces cut from a sheet, Ponoko was the best deal because I could fit a whole slew of parts on one sheet. In about 15 days I expect to receive the flat aluminum parts from Ponoko in New Zealand, hopefully right around the time I receive the new laser cut wooden case. (Ponoko says they’ll have aluminum cutting live in the USA in a month).

Got a lot of things moving over the past 2 weeks! Over and out :)

Laser cutting – Acrylic and Wood

This week I designed and sent off requests for new laser cut cases — one wood sheet from Ponoko, and an acrylic sheet from Pololu.

Material: One surprise for me — I made a slightly bigger case, and adding on 1″ to each dimension roughly doubles the material. For instance, take a 4″ square, that’s 16 square inches of material. A 5″ square is 25 inches of material. This can get out of hand pretty quickly!

Speed: On the recommendation of a friend, I browsed onto, which offers custom laser cutting services. Pololu doesn’t have plywood or bamboo in stock, but they do have 1/4″ acrylic and promise a 3 day turnaround time vs Ponoko’s 11-15 day turn time. Of course, I pay extra for the speed but for prototyping it’s well worth it. I placed my Pololu order on Thursday afternoon and it shipped Friday!

Prototyping: Designing laser cut cases sure can be a pain in the butt. I spent all day on July 4th simply designing a bigger OpenPCR case with bolted edges rather than teethed edges. My workflow goes from the overall design in Google Sketchup, adding features like teeth, bolt holes, nodes, and radii. Then importing that into Illustrator, adding additional teeth, and the submitting it to be cut.

This is fine if you’re putting together a one-off design, but for prototyping, changing, and updating designs I certainly could use some better tools. A good friend of mine has the same issues. If you have any suggestions or experience in converting Sketchup designs to Illustrator for Ponoko/laser cutting, let us know!

Thermal pads

Just received the 2 test thermal pads in the mail from Digikey.

BER165-ND is 3 mm thick and pliable. It runs about about $3.50 per OpenPCR (28 parts fit in a $94 sheet). 3M9601-ND is 2.5 mm thick and very soft, and about $1.20 per OpenPCR (25 parts fit on a $30 sheet)

We’ll want to test these out with our current heated lid to see that they work great and make it easy to maintain the temperature of samples. Once we receive the first set of aluminum and a new wooden case from Ponoko, we’ll be able to test these out even better.

Heat sinks arrived

All right, went and picked up the heatsinks at the UPS office last night. Broken them open while sitting in my car and made a couple quick judgments. Remember, my concerns are:

1. big enough surface area for the peltier
2. cost – $20 to $30
3. overall size and weight

The MassCool, while bigger in all dimensions, just isn’t big enough for the 4 cm square peltier. We would still need an aluminum plate adapter between the heat sink and the peltier.

The Titan on the other hand has the perfect sized surface for the 4 cm square peltier. It’s MASSIVE though.

I’m designing a quick prototype box today to send of to Ponoko. My focus is getting the case made to test out the heated lid. I’ve got a design using a spring hinge + thermal pads that I think will be easy to use and prevent condensation.

I’m also making the overall dimensions a lot bigger so they can fit any components we want to test out. Overall, it’s going to be 7″ tall, 7.25″ long, and 4.7″ wide. Our original case was and 5.7″ tall, 6.7″ long, and 5.4″ wide so this is test is quite a bit bigger. I’m going to test this out with both wood and acrylic, using a bolt design (the same one as the Makerbot) instead of glue. Glue is best for permanent slick looks, but since we’re  this is a hackable kit, bolts are easier to take on and off.

I’m also adding:

  1. Enlarging the hole for the aluminum block (4 cm square) to 4.4 cm square, to accommodate a layer of insulation around the block.
  2. Power port for the MicroATX ATX
  3. Front vents for the MicroATX so that we can test it internally