Learn How A Radial Engine Works or Gawk at Amazing Wood Model

via hardware – Hackaday

[Ian Jimmerson] has constructed a detailed model of a radial engine out of wood and MDF for an undisclosed reason. Rather than just delivering the wooden engine to wherever wood engines go, [Ian] decided to take the time to film himself disassembling and reassembling his engine, explaining in detail how it works as he goes. He starts by teaching about the cylinder numbering and the different possible cylinder configurations. It only gets better after that, and it’s worth watching the full 20 minutes of video. You’ll leave with a definite understanding of how radial engines work, and maybe build something neat with the knowledge.

Our only complaint is the lack of build photos or construction techniques. It’s a real feat to build something with this many moving parts that can run off an electric drill. Was a CNC involved, or was he one of those hardcore guys who manage to get precision parts with manual methods? Part 1 and 2 after the break.

Filed under: hardware

Visualizing “data visualization” with Leds and bubbles

via Arduino Blog


“Data transparency” is a project by Jiayu Liu, a designer and media artist, interested in physical data visualisation and interactive code. The installation runs on Arduino Mega: when the microphone senses a person’s blow, it transforms it in a Led animation and then activates the bubble machine for 8 seconds. The project is not aiming to visualize any specific data but “data visualization” itself:

In my point of view, data is not dissimilar to a conclusion of our past, and we need it for our future. When we see a data from a computer, it is something that has already happened. We use intelligent methods of computing science to analyze the data so that to predict the future. We are living in a world of data, and data is like a language objectively describing our past. In this work, I take more attention on rethinking and recalibrating the role of data in our lives, and the relationship between the virtual world we build as a main method of data storing, analyzing and visualization and ourselves.

Also, I am thinking of that it is better to make sense of the role of data visualization before really visualizing it. Finally, I found a good perspective to see how data connects with our lives, which is Time.
Therefore, the project is not aiming to visualize any specific data but what I am trying to visualize is the “data visualization” itself. I would like to bring a new experience to the viewer in different space. So I want to create a interesting play space and bubble game to the viewer . Let them have a really funny and relaxing experience.

Take a look at the “making of” video below to see it in action:

Enginursday: The cost of cutting edge

via SparkFun Electronics Blog Posts

I’ll admit, it’s a really cool part of my job – new and crazy technologies are brought to me like they would to VC or on a show like Shark Tank. It’s devices and technologies that the consumer market won’t see for years. Things I need to remind myself are amazing simply because in my mind, it’s so far fetched that this is happening before my eyes. Probably the most wild of instances will be when I am browsing a tech blog and see something like this being demonstrated. It will click in my head and I’ll dig through the mess that is my office (picture omitted due to embarrassment) find a box, open it, and find an evaluation sample of this amazing thing I’m seeing in this article.

Now I wish my job was done there - this thing is cool, let’s sell this. But now starts the arduous process of evaluation, when the shiny must be put to the side and a real-world look at the product must be done. This is where the “cost” of cutting edge comes into play. There’s some aspects that are pretty simple to comprehend, and others that are much more complex and, more times that not, require a leap of faith assumption to reason with.


This is probably the most obvious. Cutting edge comes at a monetary cost. These days, this cost can easily be masked when it’s a consumer good. A good example is mobile phones; new features and better hardware are almost an annual event now. Yet the price remains very close to the same. Large orders and large margins allow for the added cost to be hidden well.

I harp on this point a lot in my blog posts, but SparkFun doesn’t have that luxury. Our order quantities are much, much smaller, so prices go way up. So you might ask, “Why not buy in larger quantities?” Sometimes we do, but the amount of items in our “large quantity” and a consumer good large quantity are orders of magnitude in difference. I always giggle a little when a supplier gives me a great price in quantity because my next question is always “define quantity,” to which they give us a number that would represent almost a decade’s worth of stock. So most often, we’re forced to consider whether we can justify the price of this technology to our customers.


Going along with price, the other obvious consideration is risk. Beyond the financial aspect of bringing in cutting-edge technology comes a lot of other possible pitfalls. While it’s not as pronounced as the consumer market, our market does show demand for established products. This makes them easier to consider and estimate interest. Cutting edge is a bit different. We’re essentially going off the assumption that a lot of people will think this is cool and buy it. This is where the whole buying large quantities issue comes back into play. How you would usually mitigate risk here is start off with a smaller amount of the product. Evaluate sales and make your next build or purchase suit the sales numbers you’re seeing. This makes getting better pricing through larger order quantities difficult. So a balance must be found.


The initial novelty of the technology or product is tough to get over. You just saw this thing do something amazing. You’re not worried about interfacing with it, practical uses, or anything that you need to be thinking about when looking into a new product. But these are all part of the consideration. Here’s an example.

Let’s say we find or are brought a device that is a low cost Neutrino detector. That’s cool, this is something that could allow anyone to detect Neutrinos. But who is our target demographic for that? How many citizen scientists or even researchers will this benefit? What will they do with it? Here we have a really cool technology that we would love to promote, but what’s the use if its functionality is strictly limited to detecting Neutrinos, a solution that’s useful to very few?

You might have noticed my example uses an established technology at a “low price.” Sometimes the value in a cutting-edge product or technology is a comparative “low cost” of its functionality. It might be the ability to accomplish something for $250, when it used to cost $5000. That’s neat and should put that technology at a point where it’s more accessible to everyone. But what if the value is still lost on the customer? Part of the difficulty of cutting edge is proving the value of the functionality or price point.

Most recently we started stocking the RockBLOCK Satellite Communications module. The cutting edge aspect of this was affordable satellite messaging, and the ability to send messages over the Iridium Satellite network. For $250 and a fee for each message sent, you were able to do something that used to cost thousands. Yet still, the price point was seen as ridiculous to some for what could be accomplished by cellular in a good number of situations for a fraction of the price. Now, most saw this for what it was, but the point remained: If we weren’t able to convey the value of the product or the technology, would our customers be able to find it on their own?

RockBLOCK Iridium SatComm Module


We’ve established that there’s value in the functionality of this cutting edge technology or product. Now comes the final big question: How easy is it to use? Usually products or technologies that would be described as cutting edge are not the easiest to use. Companies want their new part or tech in new products as soon as possible. They want to be seen as a leader in this technology, a pioneer of the market. So the resources around such are still often in their infant stages. The companies assume their customer is an informed engineering team who have one end goal in mind. As for any questions their customer might have, the company’s FAE can field them easily.

But as our customer, often enough you’re not a team of informed engineers. You don’t have a very specific use for this in mind. Yet, as I pointed out in my last post there is value in the company getting their tech in your hands. Furthermore, (and in the simplest terms possible) SparkFun sees something we think our customers will dig and we want to do our best to let them have access to it. So we figure out what value we might provide to the product or technology by making it easy to use for the average hobbyist. But there are limits, and only so much value that can be added before we need to weigh usability against other factors.

Here’s where a multitude of issues (let’s call them Pearce Headaches) show up. The first of which is the projects that end in proof-of-concept level functionality (I know, that was the last subheading. These two mesh together in some places). The company wants their technology out there being used, but they’re not ready to provide full support for customers on our level (and buying power). So what is available is very basic functionality of the technology or product. This raises big questions on how usable our product would be for our customers. What if it were a gas sensor which can only detect gasses in a closed jar? It proves the functionality of the sensor, but who’s going to use something that only works in pre-determined, closed systems? There are situations where we can find value in products at this level, but they’re very rare.

The next big one which more often than not I find to be the deal-breaker is costly or unreasonable toolchains or interfaces. We’re big on toolchains that come at no additional cost to the user. So if they have to pay a big price for a proprietary GUI, it’s more than likely not going to work for us. This is a pretty understood idea in our market, so rarely do we have to confront this problem. On the other end of things is the unreasonable tools or interfaces. These pop up often. I could write all day about these issues, but things like buggy tools, obscure programming languages, or limited technical information can very easily (and very often) make the product difficult for us to justify.

I will talk about that last one though: limited technical information. This comes as a very large point of contention around here. We’re an open source company, and we promote the open sharing of technical information. But, especially in cases of cutting edge technology, products will come to us with limited technical information. Certain components might not have publicly available datasheets or proprietary firmware. But is that reason enough to discredit a cool and useful product or technology? Yet another area of consideration.

The Catch

It’s always there, always. If one of the topics above aren’t a problem, something else is. There’s always the catch with a cutting edge technology or product. The question then becomes, “How big a problem is it?”

Take our Neutrino sensor example from before. Let’s say we feel there’s a viable demand for low-cost Neutrino networks. Its toolchain is free, open and useful, and its technical documentation is plentiful. Our Junior Engineer will be running through vetting the product, figuring out how to get it running, then they hit the snag, and it’s a big one. For the Neutrino sensor, it’d be background suppression. In order to detect the faint Neutrinos, a sensor network needs to be shielded from most cosmic rays. So while still accessible by anyone, in order to work well, it needs to be deep underwater, underground, or under the ice of one of the poles. We can’t ask that of our customers for something we call “low cost.”

Neutrino Sensor Network

Here’s what a Neutrino Sensor Network actually looks like. This one in particular is part of the MiniBooNE experiment.

Supporting cutting edge technology is a tough one. I wish SparkFun had the ability to provide all cutting edge tech in ways that work for us, but in reality, very few of these cool technologies or products make it into our catalog, as we balance the cost of cutting edge, both from SparkFun’s standpoint and our customer’s standpoint. We’re always looking out for some that will work for everyone in 2016!

comments | comment feed

Submersible Raspberry Pi drone

via Raspberry Pi

We found something rather fabulous on YouTube. Niels Affourtit has taken his home-made underwater drone (or ROV – Remotely Operated Underwater Vehicle) from bathtub tests to real-world deployment in a great big lake somewhere in the Netherlands.


It’s a sophisticated build – the video below says that it uses an HD camera, the Raspberry Pi, some 3d-printed parts (propellers and a tilt system for the camera), a BOSCH BMP180 Atmospheric Pressure Sensor, and a ADXL345 Digital 3-Axis Gravity Acceleration Sensor. The whole assembly took around 250 hours and cost Niels €350. It’s been tested down to 12m, but is designed to go deeper. The unit updates a webpage with live data: tilt, roll, heading, temperature and internal pressure (leaks) sensor data are all displayed in real time.

Raspberry Pi Submarine ROV (underwater drone)

On a documentary of National Geographic about the salvage of the Costa Concordia I saw a VideoRay for the first time. I was impressed and started searching online to see if I could build one myself. I ended up at openrov.com and decided to build mine from scratch using their knowledge.

The drone is currently controlled via an Ethernet cable that keeps it tethered, so at the moment it has a limited range. Fish-spotters can watch a feed from the drone’s camera here:

ROV in hoofdvaart

My underwater ROV in Dutch channel (Hoofdvaart Nieuw Vennep). 50cm visibility (Gopro4)

There are more examples of the drone being tested on Niels’ YouTube channel. Last May, he wrote this on YouTube:

On a documentary of National Geographic about the salvage of the Costa Concordia I saw a VideoRay for the first time. I was impressed and started searching online to see if I could build one myself. I ended up at openrov.com and decided to build mine from scratch using their knowledge. I started Christmas 2014 and here you see the maiden trip in a sweet water lake in the Netherlands last week.

It’s a lovely build – we hope Niels documents the process somewhere. Thanks very much, Niels!


The post Submersible Raspberry Pi drone appeared first on Raspberry Pi.

HOW TO: China import/export permit and company bank

via Dangerous Prototypes


Last week we described the painful process of opening a Chinese company. That was actually the fast and easy part. There’s still a pile of paperwork and months of waiting ahead. This week we look at the proper, and improper, ways to export from China.

A Chinese import export permit is permission to exchange foreign money to Chinese RMB, and refund sales / VAT tax on exported products. Imports, exports, and foreign currency exchange are attached to your permit number. Any trader operating in China without one is illegally exporting and violating the currency control laws, and is not a reliable supplier in our view.

Our license was handled by an agent, we literally did nothing but hand over the company documents. It took about a month and cost around 5000RMB ($900USD). It isn’t hard to get, but is difficult to use for small scale stuff. Our giant CPA firm even botched our first attempt to get a VAT tax refund. Continue below for more on the right and wrong way to export from China, and a summary of the crisis that almost ended our Chinese company this week.

How to export from China. The legal way. 

This is the biggie. It took lots of trial and error, but to the best of our knowledge this is the proper way to export products from China and receive payment in foreign currency.

1. Receive foreign currency (USD,EUR, etc) payment by wire to the company bank account

2. Go to the bank, show an invoice for products and the import export permit

3. Money is converted to RMB. The amount and purpose is reported to government

4. Purchase products and get the official ‘fapiao’ VAT tax receipt

5. When the products leave China get a stamped export declaration form from the shipper

6. Take the VAT tax receipt and export declaration form to the tax office to rebate the 17% VAT tax. Many people skip this step. It is quite a hassle with very specific timing requirements. Unless you ship a container load of the same thing, or something very expensive, it is generally not worth the effort to get a VAT refund

7. Give the export declaration form to the accountant. The government will audit to see if the amount of currency converted to RMB matches the products exported

The tricky thing is the export declaration form. It is not a simple commercial invoice or proof of shipping. It has to come from an authorized authority with the correct stamp. It costs 300RMB ($50USD), and you need one for each package exported!


For example a simple 5x5cm $14 Dirty PCBs order. To convert that payment to RMB to pay the factory and refund the 17% VAT we need an export declaration form that costs $50 for this order alone. Unpossible.

Why are we still doing this?

So how do you deal with this? Hong Kong company to the rescue! Instead of shipping individual orders directly to hackers from China, we export the PCBs in bulk to Dangerous Prototypes Limited Hong Kong. For a while it looked like we had to rent a warehouse in Hong Kong and hire people, but eventually we found a service to do it for us.

Xiao Tang packages PCBs into boxes and puts on all the postage labels in China. A logistics company picks up the boxes, imports them to Hong Kong as a single shipment, and drops them at the post. They charge 300RMB to handle the customs export inspection and prepare an export declaration form. Now we can exchange money to pay suppliers and refund VAT.

It seems like Seeed Studio has had to make big changes to comply with this in preparation to go IPO. Small(er) Chinese companies have a different standard of accounting and compliance requirements, but when you take a bunch of government money and prepare for public listing many more rules apply.

We’ve noticed our stuff go out of stock at Seeed a few times recently, evidently while they move between Hong Kong warehouses. Our guess is they had to ship all or most of their stock to Hong Kong to refund the VAT and get the export declaration form required to exchange currency.

We’re a WFOE, which means we are held to the highest standard from day one.

How to export from China. The wrong way.

There’s a number of loopholes and unsavory practices foreign and Chinese agents use to circumvent the currency control system.

The process we describe above is only to pay for products. Payments for services can be converted with a simple invoice. So a small Chinese company might do a production run of 100 PCBs, but bill the client for consulting services. This seems so widespread for small stuff that bankers and accountants openly encourage it.

Each individual can freely convert $50,000USD to RMB each year. According to our CPA, around 70% of foreign business with Huaqiangbei market traders is paid to the boss’ personal account. This way the boss avoids paying VAT, and they don’t need an export declaration form to convert foreign currency to RMB. We won’t pay suppliers this way until a lawyer says it is actually legal to convert funds for business. Even if it is legal, $50,000 doesn’t go far for any sizable production.

A variant on the above is to recruit Chinese people to “rent” you their yearly allotment. You wire $50,000 to their account and they keep a percentage. This is so fraught with risk and uncertainty it hardly seems like a way to run a company, but it does seem like a good way to have your money stolen…


No experience here yet, but some general observations.

The import export license can be used to import and pay taxes on stuff coming into China. For example microcontrollers. Tax is generally 17%, and can be refunded when the chips are exported in a finished product. It is a bit of a hassle, especially for a small production run, but it is very doable.

In practice, almost everyone doing production in China has some variant of a story where they smuggle chips into the country in a backpack, pants pocket, etc. Foreign engineers becoming smugglers and tax cheats, over a 17% tax that’s refundable.

Supply chains are delicate enough already, you want to throw SMUGGLING into the mix!?!? Do you want your production held up for a month while you re-source chips because you got busted smuggling them into China to cheat a 17% tax? Then DON’T SMUGGLE!

Always ask your Chinese supplier for a copy of their import export license! It is at least moderate assurance your money won’t be stolen on the way into the country, and that your products won’t be confiscated on the way out.

Open a company bank account


So after months and months of work we finally have a Chinese company! But wait, it isn’t really useful with out a bank account that can convert foreign currency to RMB. This took nearly three more months.

Here’s the strange thing: Chinese banking rocks. Fees are non-existent or super duper low. Foreigners can walk in off the street, open a personal account, get an ATM card on the spot, and sign up for internet banking, all for free with a small initial deposit (~$20USD, ~120RMB).

Business banking is a whole other thing. First we went to Bank of China, cause, you know, they’re huge and international. They wanted to schedule a call in a week to setup an appointment for next month, not ok. We visited PingAn, ICBC, Communications Bank, and a few other smaller banks that weren’t even licensed to work with foreign owned companies.

Eventually we landed with China Merchant’s bank, simply because they would actually meet with us. Pro tip: choose a bank close to home or office, you or your employees will be spending a day a week there for as long as the company operates. Almost every major transaction needs to happen in person.


After more than a month we received permission to open a bank account from the People’s Bank Of China central banking authority. At this point the import export permit was finished and we entered another month of waiting for approvals before the account was open.

Capital injection


Even with a bank account and import export license we still can’t run the damn company. We have to “inject” the 400,000RMB of capital, then convert it from USD to RMB.


The 400,000RMB in foreign currency is wired from the business owner, the HK company, to a special single-use capital injection account. When the money arrives, appear in person to convert up to 300,000RMB per day for operating expenses.

If you want to withdraw and convert capital to pay a supplier, say for PCBs, you have to submit already-paid tax receipts. Our accountant describes this as an “incomplete system”: you can only use company money to buy products, but you can’t get company money until you can prove you paid for the products and taxes. We had pay for stuff with personal money pulled from an ATM machine so we could get the tax receipts so we could get money out of the company. Which came first: the chicken or the fapiao?

Online banking


China Merchant’s Bank has reasonably workable English (likely Windows only) crapware for managing accounts. It comes with two USB certificates, one for the accounting department and one for the administrator.

Each transfer is first entered by the accounting department login, then the administrator has to login and approve. It’s pretty burdensome for a small business.


At least the developers seem to care about the user experience. “Check if there is enough fun”. Indeed.

For the big ones, not the small ones

One theme that keeps popping up: China is still built for big business. The plus side is a real company with real no-bull expenses and tax deductions. But, while the bank software would be great for an organization of 100+, but it stinks for a couple hackers who want to export a few PCBs. Similarly, it is easy to export a shipping container load of stuff, refund VAT, and convert payment to RMB, but you gotta hack the system to ship a $14 PCB order.

Maybe this is why such a large gray market export economy is allowed to thrive in China. Foreign and local agents are exporting products, which China encourages, but the system is incomplete and overly burdensome for small companies and individuals to be fully compliant.

A blind eye approach could be much more effective than reworking the whole system. After all, currency control largely exists to prevent big (foreign) interests from speculating and manipulating the Chinese economy. The spirit of the law isn’t to bust an eBay Arduino seller for illegal exports.


For those of you following along on WeChat #shenzhen_hacker_bei, some of this might seem familiar. We reviewed our business plan with the accounting firm multiple times, but still had a moderate crisis last week.

The accountant specifically told us, in writing and on multiple occasions, that a fapiao tax receipt was sufficient to get a VAT refund. We show up with fapiaos from the PCB factory for a refund. They pull out the example export declaration form – no refund without it, and by the way, no currency conversion either. It was obvious our accountant simply had not done it before. Rather than ask the head CPA, she just parroted incorrect conventional wisdom from her colleges. Idiots.

For most of the week we thought we were tanked. After multiple visits with lawyers and accountants we pieced together the full picture described in this post, and it seemed impossible to continue without renting a Hong Kong warehouse.

Yesterday we finally found the logistics agent willing to handle the export for us and provide an export declaration form. Hopefully these extended write-ups save someone anguish in the future.

Company. Check. Import export license. Check. Bank account with money. Check and check! But wait! We still need the work permit and residency permit! Add another 2 months before we can actually run the company. More on that process next week.


Massimo Banzi hangout at Fabacademy

via Arduino Blog


Fab Academy is a distributed educational model directed by Neil Gershenfeld from MIT’s Center for Bits and Atoms. Students view and participate in global lectures broadcasted every week and on February 1st Massimo Banzi was invited to give a lecture to an audience of  students from all over the world. You can watch the 50-minute recorded lesson in the video below: