We worked with Tompkins Industries to get hydraulic fittings and hoses in-house to complete the power unit, and spent our week doing hydraulic assembly. In this update, we’ll show you our progress so far, explain a little bit more about the hydraulic system, and teach you how to assemble hydraulic fittings into a working system. Check it all out here:

-Gui

I’m going to detail a subtle problem we hinted at a few months ago. To understand the problem one needs a basic understanding of the system layout and a few key concepts, so I will quickly review.

Each actuator in the leg has a control valve attached to it. The control valve is opened by an electric current. In an ideal world, a constant electric current will result in a constant size orifice through which hydraulic fluid will flow.  The simplified topology is displayed below.

The symptom is very interesting and counter-intuitive; when a control valve to an actuator is opened with a constant electric current, the system goes in to a self sustaining oscillation. This can be extremely dangerous since the oscillation in question is a 350 lb leg swinging around!

It took us a while to figure out exactly what was causing the oscillations. A poorly tuned control loop in software can easily cause similar symptoms, so a lot of time went in to tuning our control loops trying to get the oscillations to stop happening before we discovered the oscillations were implicit to the system and occurred even with constant valve command.

Why does this happen? One would naively expect that a constant current to the valve would result in a constant orifice, which should give a smooth motion at a constant speed at the actuator. When you turn the handle of your faucet, you open a constant size orifice from the water mains to your sink, and a constant flow results. You would be surprised if you turned on your sink and the water came out in spurts.

There are two concepts you need to understand to grasp these oscillations. The first is the concept resonance, the second resistance or damping.  Click on the words if you want the wikipedia articles, but I’ll try to give a shorter, application specific explanation below.

Resonance

How many of you have played with one of those little springy doorstops?

In the mechanical paradigm, a resonance is created whenever a mass is attached to a spring.  The mass and spring are both forms of energy storage – the mass stores kinetic energy in its motion, the spring stores potential energy in its deflection.  The resonance is simply the mass and spring passing the energy back and forth to each other.  All mechanical resonances behave this way – from guitar strings to pendulums to swaying bridges (gravity provides the spring in pendulous cases).

Resistance/Damping

Why does that little springy doorstop stop moving?  The doorstop has many forces acting on it, but for the sake of simplicity let’s pretend only air resistance is acting on it.  When you move through the air, the air pushes against you.  If you move forward, air pushes backwards.  So the air is always pushing against the motion of the door spring, taking a little bit of energy out of it every cycle as the end swishes through the air.  The air acts as a damper on the doorstop.  Since the air is taking energy out of the doorstop resonance, we would say it is a positive resistance/damper.

Changing the amount of resistance changes the behavior of the resonator.  Let’s say we raised the resistance on the doorstop by putting it underwater… the doorstop would probably swing back and forth far fewer times, as energy leaves the system much faster.  What if we put the doorstop in a vat of syrup?  The doorstop would probably not resonate at all, it would just slowly swing back to the center position.  These are both cases of increasing the resistance.

What would happen if we changed the resistance from positive to negative?  That is, instead of the air pushing against velocity, it pushed with velocity.  It’s a contrived example, but you can intuit what the behavior would be – instead of the oscillations diminishing over time, they would grow until some other factor started limiting them.  In the doorspring case, this might be when the end of the spring starts hitting the walls.  The controls guy would say “the door spring has entered a limit cycle“.

What’s this have to do with the Protoleg shaking?

This is oversimplifying somewhat, but think of the Protoleg like a giant springy doorstop.  It has mass (about 350lb of it) that can store energy in its motion.  It has a spring at its base (the natural springiness of the steel, the expansion of the hydraulic hoses…) that can store energy in its deflection.  That combination of mass and spring creates a natural resonance.  This resonance also has inherent damping (mechanical friction).  If the system is just sitting there and you shove the foot in one direction, the foot will bounce off its joint stop, bounce back off the other joint stop and settle.  The natural resonance dies out after 1-2 cycles.  The controls guy might say “in this regime, the leg is positively damped”.

If the valve just acted like a small hole through which fluid flowed, it would still be a positive resistance in the system, pressure opposing flow.  But the valve is more complex than that and has a flow profile that looks like this:

This is a chart from the valve manufacturer showing flow (y axis) vs pressure (x axis) when the valve is commanded with a constant electrical current.  Notice that for a large portion or the graph, the slope is negative.  This means that in those portions of the graph, more pressure you put across the open valve, the less flow you get through it.  This creates a large negative resistance that causes leg oscillations to amplify until they enter a limit cycle. Yikes!

How can we fix it?

Engineering is all about tradeoffs.  If we had infinite money we would probably just replace our valves with a much more expensive type that doesn’t have an implicit negative resistance. But (a) we don’t have infinite money and (b) super expensive solutions are antithetical to the mission of Project Hexapod. The simplest, cheapest and fastest solution for us is to just add a positive resistance in to the hydraulic circuit to cancel out the negative resistance.  This can be accomplished by just sticking a small hydraulic orifice in the circuit.  We’ve been using orifice disks mounted in the bottom of our valves, but we’ve had problems with them floating out of place since we mount our valves upside-down, so we are switching to orifice fittings that go on the end of our hoses that can’t float out of place.

This has the down side of reducing the maximum output power of the joint, but it’s a relatively small price to pay for a more controllable system. The video below shows some preliminary testing with and without an orifice in place. Note that the orifice is slightly undersized in this video.

Thanks for reading, and thanks again to our sponsors at HydroAir and HydraForce who have sent us several rounds of hardware to help us experiment with solutions to this problem.

We also continued working on the control systems for the leg and the hydraulic power unit. In order to develop code faster, we went ahead and concentrated on updating our simulation to reflect the mechanical realities of the leg – specifically, we added backlash and quirky valve response to the simulation, in the hopes that we can iterate much faster on the control system at home (instead of on-robot). James will have a long, in-depth control systems update soon about this.

The hydraulic power unit is coming along – we’ve mounted all our components, and we’re now working on connecting them together. This week, we’ll show you some of the mounting and fitting assembly work we’ve been doing, and we’re going to take the time to step you all through how the hydraulic power unit works. Let us know if anything was unclear!

Tune in next week for more work on the power unit, and updates on some interesting control systems we’ve been working on to address our mechanical and hydraulic issues on the prototype leg.

-Gui

In this update, we cover the building of a Torchmate 2 computer-controlled plasma cutter with a custom-built water table, the continuing development of the hydraulic power unit, and continued testing and debugging of the prototype leg. Check it all out here:

This coming week we’ll be working a lot more on the powerplant – we have a hydraulic heat exchanger coming in that we need to mount (and that we’ll talk more about), and we’ll be starting work on connecting all of our components hydraulically. We’ll also continue development work on the leg, and figure out what courses of action (both mechanically and in control system development) we need to take to reduce the leg judder that you see in many of our videos.

-Gui

In this coming week, we’ll be doing a lot of tuning of the control loops that govern the leg’s motion, continuing the final design process for the legs, and pushing ahead with the hydraulic power unit integration. Stay tuned – we’re starting to pick up some serious speed!

-Gui

We have come to realize that a) we are bad at this continuing-to-write-blog-posts thing, and b) even so, we really need to be documenting this process for the world to see. Thus, we’ve decided to change gears a bit and offer weekly video updates instead of blog posts. Check out our first update below!

Let us know what you think, and if you have anything you’d like to learn more about!

-Gui

We are happy to report our prototype leg’s first movements!

We are equally happy to report that it takes roughly 600 pounds of force to crush a Macbook G3.

Now that we’re past the orgy of destruction, let’s talk about the fun stuff… safety! When pressurizing a newly constructed hydraulic system for the first time, one must take extreme care. Our systems are pressurized to ~2300psi. That’s 2300 pounds of force per square inch. These are levels of pressure that most people don’t ever deal with and don’t have intuition for.

There are many inherent dangers to working with fluid at these pressure levels. If a fitting blows, the pressurized fluid can send pieces flying with great energy. Jets of the hydraulic fluid itself can also cause terrible injury… a jet of pressurized hydraulic fluid will cut flesh and inject the wound, giving the double whammy of slicing trauma and poisonous injection. Here’s the kicker: these jets are often so thin that they’re effectively invisible. Google “pinhole leak injury” if you want to know what this looks like, but be warned that it’s pretty gruesome.

To try to avoid these potential problems, we pressurized the system to full pressure and just let it sit for a while with everyone standing far away. This gives oil a chance to accumulate in places where it is leaking. The HPU is then turned off and any leaks are addressed. We found a few slightly loose fitting with this method.

Next, we did a “rag sweep”. This involves tying a loose rag to stick and just sweeping the airspace around every hose. If there are invisible pinhole leaks, they will make their presence known by deflecting the rag. We found no pinhole leaks, which makes the following video very boring (but we feel it should be included so you can see the process).

Our first moves were very slow and cautious – we want to explore our full range of motion and make sure we didn’t have any hose or cable pinching happening. We had the system at roughly full pressure (~2300psi supply, ~250psi return) but very limited (1.5gpm) flow, to guarantee very slow motion.

Here are a few range of motion pics:

And the full range of motion video.

Next we’ll be coordinating multiple joints under computer control. Sweet.

Just a quick update to announce that all swag (t-shirts, bumper stickers, wrist bands) bound for US addresses has shipped or will ship by end of today. For US supporters – your shirts and bumper stickers should arrive before Christmas.

Here are a few pictures of the team packing up the thousand+ orders. Our deal with the shipping company fell through, so we finally just decided to get everything done the old fashioned way…

International swag has not yet shipped and we’ll let you know when that goes out. We will definitely be shipping international orders before Christmas, but I wouldn’t count on them arriving before Christmas.

Thanks again for your support!

It’s been too long since our last update, and for that we apologize. We’ve gotten a ton done. Most of it has been the not-so-glamorous support work that needs to happen. Since our last post, we’ve:

• Acquired insurance for prototype leg and robot operations
• Worked out a safety agreement with the Artisan’s Asylum
• Received all of our swag!
• 800+ T-shirts of many sizes
• 1500+ bumper stickers
• 1500+ wrist bands
• Negotiated a deal with a shipping company to ship all our Kickstarter backers their rewards (without breaking the bank)

The T shirts arrived on Friday and are looking great:

So to all our Kickstarter backers, your stuff should be arriving in the next month.  It’s taken us longer than we expected, and for that we apologize.  We still need to pack and address over a thousand packages, but we have all the equipment in place to do it.

On the robot side of things we’ve also hit a few milestones.

We’ve put together a mobile, electrically powered hydraulic power unit (HPU).  This HPU will make it easy for us to do controlled, indoor testing for the prototype leg.

That cart may look familiar: we repurposed the custom-built cart we used for Gimpy, our first prototype leg.

“What did you use to power Gimpy?  Why don’t you use that to power the big prototype leg?”  Great question.  Gimpy’s hydraulic power unit could only supply 1000psi at a very low flow rate.  We need at least 2200psi to realistically test the full scale protoleg.

We’ve hydraulically plumbed the leg: all the necessary valves and hoses are in place.  The entire leg hydraulic system comes back to a pair of quick-disconnect fittings that connect to our mobile HPU.

We’ve also built up the joint controllers we’ll need to actually control this leg.  These joint controllers are much more advanced than the ones we used on Gimpy, and feature:

• Fast Ethernet (100Mbps)
• Power over Ethernet
• 12-bit rotary magnetic encoder (Austria Microsystems AS5045-ASSU) for reading joint positions
• Dual 1.5A, 12-bit, dithering current controllers for driving valves

Next up: packing and shipping over 1000 units of swag, then wiring the leg and *gasp* turning it on for the first time.

Thanks for reading and have a happy Thanksgiving holiday!