Craftycoder's Automotive PDU - Update

I've been busy working on this project since I last wrote about it. I have a new PCB designed and built up and another that has since been designed but not yet built. Designing electronics is a process. The first design I had built had an analog mux on it that I was using to push all the current sense signaling to a few ADCs on the board. Later designs needed those pins on the MCU for other purposes so I added an 8 channel ADC to the design and removed the mux all together. I then found that the ADC was too fast to measure the current in a useful way because the PWM I use to adjust the current would give the ADC values that were either 0% or 100% instead of a working average. I knew this going in, but assumed I'd be able to get a reliable average in a reasonable amount of time. For whatever reason though, it took about 300 samples per channel to get an reliable average value. Multiply that by 8 channels and then assume a single byte to store the value and you can see that it consumes at least 2400 bytes (I only have about 500 to work with). It also took about a 12ms to get a value (there was a lot of overhead in the serial communication layer I had not accounted for). Multiple that by 2400 and it would take almost 30 seconds to get reliable current values from all the channels. That was a total epic fail. So, I went back to the drafting table and drew up a circuit that did all the averaging in the analog domain which operates WAY FASTER than any digital circuit can. What I drew up is a three stage low-pass filter for each channel. Image below.

Blue: PWM
Green: 1st Stage
Red: 2nd Stage
Celeste: 3rd Stage

As you can see, the circuit is not exactly trivial (especially if you consider I needed 8 of these added to an already crowded board; 56 more components). The PWM input signal is turned into a low impedance DC voltage source (with the help of an op amp) that is proportional to the duty cycle of the PWM signal and whose DC offset is proportional to the current draw. The second part of that is done with another op-amp and a very large but low impedance resistor and then I do a differential measurement to get the voltage drop across the resistor. I multiply that drop so that the upward bound is just below the top DC rail of the ADC so I get the most bits of resolution. Look up high side current measurement for more details on this. I used LTSpice to help select the values for the RC filter network. The values have a lot to do with the frequency of the PWM input and the speed at which you want it to settle. You can see from the above graphic that I settled on a filter that gives me a decently flat, DC like, signal that settles within 50ms of a current change. This is plenty fast for the purposes of a circuit breaker and should provide excellent user feedback as well. When they turn up a circuit they should see the current like move very fast. This is in fact the case and the video below will show that.

I had another exciting and problematic feature that has been a challenge to create. Market feedback suggested that some constant current capability would be appreciated. At first I thought I may be able to shoehorn this in with a software fix. It turned out to be just too slow to work. When you require constant current you really don't want to overshoot too far or you can fry your load before the circuit has a chance to moderate the current. On the latest iteration of the I added some pretty clever hardware constant current stuff by taking advantage of some features on the MOSFET driver that were for another purpose but could support a constant current function with a little external help. I thought about it for a while and drew up a circuit and had boards built but I never bothered to do ALL the math required to fully understand the circuit I designed. It turned out when it was finished that I didn't understand all the variables that controlled the timing of the circuit. I ended up with a perfectly adequate constant current source, but it oscillated at 20Hz instead of the 200Hz I expected. At first I thought I just dropped a decimal when I was doing the math because 20Hz and 200Hz look suspiciously similar. I redid the math and I had not screwed that up so I had to look more closely to see what was happening. I put it on the scope and then it became quite clear. Remember that 50ms number I mentioned about the settling speed of the 3 stage RC filter I created? Well, if you do the math, you will find that 20Hz is a synonym for 50ms. That was the problem. My filter was too slow to support the constant current feature I wanted to offer. I needed another 10x increase in performance. I knew this was going to be a big challenge and I also knew that my LMV324 op-amps (that were part of the circuit) were not going to be up to the task of moving so fast. I completely redesigned the constant current circuit with high performance op amps and much faster filter. I then went ahead and used LTSpice to model the circuit to confirm the design. This is a somewhat time consuming process, but these 4 layer boards and accompanying stainless steel stencils I am buying cost $200 and take about two weeks to get so its worth the effort to prove out the design before wasting all that time and money.

LTSpice is an incredibly useful tool for these sorts of electrical engineering problems. You can see that I am able to get reasonably fast PWM signal that is at a constant current over a roughly 100Hz time frame. The 100Hz speed is a compromise between the slow speed of the former version and fast speed that I desire. My secret sauce here is that I am using only 2 stages of the filter for this instead of 3 and I am adjusting the the first stages a bit. This is the compromise. I could other wise pretty easily get any speed I wanted but I'd have to have even more passive components on the board and its already packed with parts.
I suppose I"ll leave with a couple videos showing the circuit in action.

3D Printer Meetup, Program Line-Up - 02/09/2013

Freeside Atlanta and My Inventor Club will be teaming up on Saturday, February 9th to bring you a daylong hackathon in 3D Modeling, Design, Marketing, Theory and more.  This is a free, educational meet and greet event which will be hosted in Freeside Atlanta and My Inventor Club's space.

This is the (draft) speaker line-up and schedule of events!  Like so many of these things, this is subject to change.  Watch this blog or follow us on Twitter for updates and other announcements.

Cost: Free
Location: Freeside Atlanta, 675 Metropolitan Parkway Suite 6066, Atlanta, GA 30310
Directions, Parking, and other Information
Date: February 9th, 2013
Time: 1200-1700 EST
Contact: pr@freesideatlanta.org or info@myinventorclub.com

Speakers Track 1100~1500 Eastern Time (UTC−05:00)

This will be a Google Hangout invitation on the day of the event.

Nicholas Giovinco, Moderator - Freeside Atlanta
Greetings and Introductions: Welcome to Freeside
Time/Length: 30 mins 1100-1130 (UTC−05:00)

Shane Mathews - My Inventor Club
Topic: Prototyping Management: Strengths, Weaknesses, and Expectations
Time/Length: 30 mins 1130-1200 (UTC−05:00)

Shane Matthews, product development guru and founder of My Inventor Club will talk about the professional use of 3d printers, the difference between hobby and professional grade 3D printers and how 3D printers, MakerBot Industries, have impacted the professional community. In addition Shane will have sample prints from some of these printers to show the difference between the prints as well as prototypes casted from 3D printed part. Come see first hand some of the differences in 3d printers, ask questions and learn.
http://myinventorclub.com

Colleen Jordan - @colleeniebikini
Topic: 3D Designer Wearables and More!
Time/Length: 15 mins 1200-1215 (UTC−05:00)
http://www.colleenjordan.com

Clint Rinehart
Topic: 3D Scanning with Open Source Software and Microsoft Kinect
Time/Length: 30 mins 1215-1245 (UTC−05:00)

3D scanning is a common practice among 3D printing enthusiasts.  This technology is no longer a limited modality with research or corporate funding, but now a practically attainable practice with a VERY low cost of entry.  This talk will feature the use of a Microsoft Kinect device with ReconstructME software.

Chris Caswell - [carrythewhat?]
Topic: Replications: a case study on the basic economics of Distributed Manufacturing of Free & Open Source Hardware. 
Time/Length: 15 mins 1245-1300 (UTC−05:00)

For 18 months now, our desktop 3D printers have been producing commodity prints and sold on the market. With 3 printers now, we are approaching viability as a self-sustaining business. But we have shown there is this small repertoire of digital goods, capable of desktop replication, and valued by the market -- set only to expand as replication technology advances and the software catches up.  Along with our experiences so far, I will share my thinking on free & open source hardware, distributed manufacturing, and an open economy.
http://carrythewhat.com

Jason Webb - Graduate student, creative technologist, OSHW engineer
Topic: DIY mold-making with 3D printing
Time/Length: 30 mins 1300-1330  (UTC−05:00)

Learn how to design and 3D-print your very own low-cost molds, which you can fill with Jell-O, chocolate, silicone, resin, wax and more!
http://jason-webb.info

Emmett Lalish - Engineer or an Artist?
Topic:  Designing 3D-printed mechanisms
Time/Length: 30 mins 1330-1400 (UTC−05:00)

I'll give some pointers on using OpenSCAD to design complex, printable mechanisms. I avoid using support at all cost and in my book, the order of coolness for mechanisms goes: bolt together, snap together, preassembled. The beauty of 3D printers is that they make preassembled mechanisms possible for the first time, but designing them requires thinking about shapes in a whole new way.
http://www.thingiverse.com/emmett

Mark Ganter - The Solheim lab Open3DP
Topic: AM in the early morning hours...
Time/Length: 30 mins 1400-1430 (UTC−05:00)

The Mechanical Engineering Department at the University of Washington is host to the Solheim Additive Manufacturing Laboratory.   This lab hosts the Open3DP website.  Open3DP's mission:to disseminate information and foster a community of people interested in an open sharing of 3D printing information. We hope that you find useful information on this site and that you will feel free to comment and share your expertise.
http://open3dp.me.washington.edu

Henry Thomas - “Wingcommander” WPThomas
Topic: General experiences getting my Replicator 2 to print reliably
Time/Length: 30 mins 1800-1830 (UTC−05:00)

Breakout Workshops 1500-1800 Eastern Time (UTC−05:00)

These courses will be held at both Freeside and My Inventor Club’s physical spaces.

Scanning with Microsoft Kinect
Instructor: Clint Rinehart
Location: Freeside Atlanta Worktable
Requirements: Self

A demonstration of Microsoft Kinect Scanning and Reconstruction with ReconstructME.

CAD, CAM, Print!!: Module 1
Instructor: Nicholas Giovinco
Location: Freeside Atlanta Classroom Zone
Requirements: Laptop and Creativity

This is part 1 of a two part workshop, centered on Computer Aided Design (CAD).  In this, we will walk through a very simple design process in Google Sketchup of creating your very own Keychain.  Choose a word or name and lets begin!

CAD, CAM, Print!!: Module 2
Instructor: Nicholas Giovinco
Location: Freeside Atlanta Classroom Zone
Requirements: Laptop and Creativity

This is part 2 of a two part workshop, centered on Computer Aided Manufacturing (CAM), where we work our brains around physical creation of our idea.  In this, we will discuss the best methods for printing our designed keychain in a MakerBot Replicator 2.

zCorp 3D Printing via Cold Sintering
Instructors: Buddy Smith & Patch Trowell
Location: Freeside Atlanta Woodshop Zone
Requirements: Protective Eyewear is supplied and must be worn at all times

Additive manufacturing has many, MANY, applications and modalities.  Hobbyist machines such as Makerbot and other Repraps utilize a derivative of Fused Deposition Modeling (FDM), but a lesser known modality is Sintering.  The zCorp technologies found at My Inventor Club and Freeside Atlanta are capable of just such methods.  In this workshop you will see and experience the basics of this technique applied.

Startup Your Prototyping Idea
Instructor: Shane Mathews
Location: My Inventor Club
Requirements: Interest in how to develop an idea

Discuss the first steps in taking an idea to a physical prototype, deciding on the best process to make a prototype including materials, fit, finish, number of evolutions, and other details of prototyping before moving to the next step.

CNC Milling Station #1
Instructor: Eldon
Location: Freeside Atlanta

CNC Milling Station #2
Instructor: Brian Cribbs
Location: Freeside Atlanta

Craftycoder's 8 channel automotive PDU

I have been working a Bluetooth 4 automotive power distribution unit for a couple months now. Here are some videos showing the capabilities. 

The goal is to have a device about the size of a deck of cards that can distribute up to 100A @ 13.8V for automotive applications. 8 individual circuits can be controlled. Control will include constant current, circuit breaker at specific currents, and PWM duty cycle. You will be able to program dedicated functions to external buttons. Lastly, you will be able to get real time feedback from the system on a Bluetooth 4 device. This data will include total current, current by channel, and system voltage among other things. There are a lot of features on the drawing board still.

This is going to have application in show cars, jeeps, motorcycles, and boats I believe. It's been a really fun project so far. I'm going to need about $120,000 to bring it to market so I want it to be pretty perfect before I write that check. I do hope I get that far.

3D Printing & Design workshop February 9th, 2013 @ Freeside Atlanta

Come one; come all!

Freeside Atlanta and My Inventor Club will be teaming up on Saturday, February 9th to bring you

a daylong hackathon in 3D Modeling, Design, Marketing, Theory and

more.  This is a free, educational meet and greet event which will be hosted

in Freeside Atlanta and My inventor Club's space.

What to Expect:

-Several Machine Technologies on display with instructional demonstrations

-Show and Tell examples of select prints and design prototypes

-Lectures ranging from Design Theory, Modeling Techniques, Printing Strategies, Troubleshooting & Community Experiences

-3D printing workshops and Hands on immersion

-Special Guest Speakers

Cost: Free

Location: 675 Metropolitan Parkway Suite 6066, Atlanta, GA 30310

Date: February 9th, 2013

Time: 1200-1700

Contact: pr@freesideatlanta.org or info@myinventorclub.com

-Nick

3D Printer Meetup recap

We've had two really great 3D printing meetups lately, one to talk about and demo 3D printers, and another to help people build and debug them. Here's some of the stuff we saw -

  Jorge brought in his Rostock Prisma to show it off and debug a few extruder issues. It's a really cool and fairly robust design. Plus, it's a lot of fun to watch! You can also check out the Thingyverse page that Jorge has put together for it.

 We also had Ant's 3D printer and the Thing-o-Matic going, so people could benchmark or run off a few parts. James Stephens also brought in his RepRap Mendel to debug some extruder issues.

Randy did some work on his CNC, building a cabinet with removable side-walls for it, while Van continued working on his 3D Printer, the same variant as Ant's and the one that I am building for the space. I made slow progress on that, too, and it "only needs to be wired up" now.

That's all for now! Watch our Meetup page for more great events like these!

Precision Low Current A/C Measurement

I got an interesting question on my YouTube channel yesterday asking if my transimpendance ammeter can measure A/C in addition to D/C and if the output of the instrument is A/C. The answer is yes to all of the above. Good question and a fun little thing to test on the bench on a Saturday morning.

Full power tests of motorcycle PDU

I pushed 185 watts through my custom DC relays last night. They ate up that power like a champ. Likewise, I've also been testing it in the off condition to measure its steady off power consumption. The cards burn 150 nano amps or 1.8uW when off. That's means it's about 80 mega ohms which is OUTSTANDING. This project is really moving along nicely. I have some serious burn in testing to do and then it's time to get the main board encapsulated and do some real life testing. That's going to be fun!

Reflow Oven Demonstration

Arduino Compatible Mini USB Host Board

Well, its time to finally get this project "out there". I've been sitting on it for months putting together a bunch of cool little projects for this board and the video above shows off some of the cooler ones. Check out the Kickstarter page for more details as well.

Ever since I saw Kickstarter the first time I've been wanting to give this a shot so I could learn the ins and outs of doing small scale manufacturing from home. Because this is not a product, but a platform or hacker tool I am not super confident there is a substantial enough market for it but I cannot know if I don't try.

The plan if this succeeds is to give all the boards a final value engineering assessment and put the project out to bid. I've already done this with a Florida shop called Tropical Assemblies as well as with Seeedstudio. I'm not super excited with Seeeds PCB quality so I think I'll skip them but the good news is that the local fab shop had similar prices so at least I'll know that I can call and complain if anything goes wrong.

An issue has already cropped up with the Joystick shield. A part I used in the BOM is no longer available anywhere. I'll contact the vendor and verify its availability or lack thereof and reevaluate. I've already got a backup plan but it will require a two sided board which is going to kill the bottom line. No matter, this is more about learning this home manufacturing process then about any profit motivation so we are just going to drive it hard and fix what breaks.

Many thanks to Freeside for all the support and camaraderie.

Offroad Wheelchair Update 1 - To Retrofit or Reinvent?

Recently, we at Freeside Atlanta teamed up with the Alchemical Arts Alliance and My Inventor Club to design and build an Offroad Wheelchair so that our friend Robin can get around their events. Together, we've raised about $2,000 for the project.

The design phase is usually the most difficult part of the projects, and it is often the most expensive place to make a mistake. Committing the resources to a poorly-designed project can cause the entire thing to be wasted, so we designed the Offroad Wheelchair project very carefully.

We started with the constraints – We need ease of maneuverability, an ability to overcome obstacles and take fairly steep inclines, longevity to make it through events up to a week long, and recoverability in case it gets stuck. Also, biggest two constraints – Budget (~$2000) and time to prototype (2 months)

From that, we landed on 3 design options –

1 – Electric motors with onboard generator for periodic battery charge

2 – Modifying an existing lawnmower or outdoor vehicle with hydraulic controls and automation

3 – Modifying a zero-turn lawnmower to suit our purpose

After weighing the options, we chose option one for its low noise and high efficiency. Next step - simulate the design. I used the website Study Physics to figure out the torque requirement to pull a simulated wheelchair + passenger up a given slope. Then I simulated it in a spreadsheet starting with the worst-case scenario so I could play with the numbers. The inputs are in orange and the rest are calculations.

In other words, the force required to take a vehicle up a 15 degree slope is a bit more than half of the force required to lift it all the way off the ground. To find the torque requirement, the force needs to be distributed around the wheel, since the wheel radius affects the leverage exerted by the motor.

Therefore, the minimum torque of the motors is about 200 pound feet. Is that feasible for an electric motor? Let’s look at the best motor example I could find - 

This means that the system required two very powerful electric motors, running near peak torque continuously, and geared down 60x. They exist, but they are about $600 each and don’t usually match each other’s output exactly. Plus, the motor controllers are $200-$400 each! That means that this project is neither feasible nor practical, as the entire budget could be spent on 2 motors with their respective transmissions and controllers. It won’t work. But, this is why we simulate. It’s time to drop back and try again.

The hydraulic modification of an existing vehicle seems interesting, but also time-consuming. Plus, it wouldn’t be able to pivot in place like the other two design options, which is important to delivering the rider to exactly their target. Not to mention that the drive of the wheels will probably be locked together, so there is nothing stopping the thing from sliding down a slope. It also requires the modification of a ~$1000 platform with up to $800 in hydraulics and most options aren’t configured for somebody to easily get into from a wheelchair.

Then, Shane from My Inventor Club sent me an email about a zero-turn mower that was available in our price range. Zero-turn mowers use hydrostatic transmissions to convert the high-rpm, low-torque energy from an engine into low-speed, high-torque power with hydraulics. The mower we chose is a Grasshopper 725k. The actual torque isn’t listed in any of the data sheets, but the horsepower (25) and wheel diameter (22in) are. So now we just need to find the torque and update our simulation to see if it will work.

If we know how to find torque then we just need to find the rpm at the wheels at the top speed, discount it by about 30% for conversion losses, and that will be near our actual torque.

First, get the top speed to feet per minute – 

Then, turn that to rpm using (the diameter of the tire) * (pi) to get the ground it covers per revolution.

Finally, you can do the torque conversion – 

 The zero-turn mower is 800lbs, so it is much heavier. The torque required to haul it up a 15 degree slope is 530 pound-feet, so it might be able to crawl along if it can keep traction. However, rolling over would be a concern at that angle anyway. Therefore, the zero-turn is feasible and reasonably meets the requirements of the project.

We can look into supplementing the power of the motors with a winch if the vehicle gets stuck somewhere that the torque/traction can’t overcome and reduce engine noise, but we're in the ballpark now. So, we’ve settled on a design – a retrofit of a zero-turn mower with a winch. It will be faster and more cost-effective to retrofit a used device that has been engineered to a similar task than to design and build something new from scratch. However, this platform could serve as a prototype for some future design, so we can work out the issues through iterative prototyping.

Until then, we’ll do the most effective and cost-effective Offroad Wheelchair that our constraints allow and document it so others can duplicate and improve the idea.

Check in on the Offroad Wheelchair page for links and info, including the simulation spreadsheet that I used (you get to point out my mistakes!) and the first video of the mower that we selected.