Steam Engine Correspondence/Josh Jordan

=Mark Norton, July 3, 2011=

Hi Josh,

I got your request via Pivotal to join the Steam Engine project. How would you like to be involved in the project? I could certainly used some help. Some question follow that relate to how you might contribute:

How familiar are you with physics simulations? Do you think you could model the steam engine in an application (there is math available)? Do you have any experience with microControllers like the Arduino? Do you have any interest in building the engine or parts for it? Do you have experience making such things?

If none of the above, how open are you to discussion other tasks (documentation, marketing, fundraising, etc.)?

- Mark Norton

=Josh Jordan, July 3, 2011=

I have no experience modeling physics or making things out of metal. My skills are in electronics. I can help with picking what sensors and actuators to use if that is part of the design. I can pick out chips and design the printed circuitboard. Let me know what you need and I will write a specification document and do the design.

-Josh Jordan

=Mark Norton, July 4, 2011=

Ok, fair enough. Here's something that you could do for me. Do a bit of research into electronically measuring pressure in the range of 0 to say 300 psi, dynamically changing. I want to capture samples at 50 times a second and store the data. Can you find an off the shelf product to do this?

- Mark

=Josh Jordan, July 4, 2011= Maybe this will work, its a little expensive @$75: http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=MSP6864-ND 50 readings per second will be no problem. What do you want to do with the data? Activate something when pressure reaches a point?

=Mark Norton, July 4, 2011=

This is a good start. Can you find others, perhaps less expensive? Please add data to: http://opensourceecology.org/wiki/Steam_Engine_Specifications/Pressure_Sensor_Sources

What do you want to do with the data?

Several things. Pressure readings can be used to create a feedback loop to optimize performance of an engine with dynamic (computer controlled) valves. See http://opensourceecology.org/wiki/Steam_Engine_Design/Rotating_Valve. Also, pressure readings vs. distance can be used to create an indicator diagram (http://opensourceecology.org/wiki/Steam_Engine_Specifications/Indicator_Diagrams) from which can be calculated total work performed by the engine and other interesting statistics.

Activate something when pressure reaches a point?

In a dynamically controlled steam engine, pressure levels (both high and low) can be used to trigger events in a controller program.

This is good stuff you are doing. Get me a few more sensors and add them to the page I mentioned above. A range of prices and capabilities are useful.

- Mark

=Josh Jordan, July 4, 2011=

$75 is a good deal for 300 psi. Maybe it can be built without needing a pressure sensor. Low pressure sensors up to 20psi still cost around $10. Can you make a 20psi engine?

=Mark Norton, July 4, 2011=

There are really two use cases for a pressure sensor:

1. As a measurement instrument - not always on the engine. Expense matters less in this case. 2. As a data collection device for dynamic control. Not sure if needed or not, but let's source them now and figure need later.

Can you make a 20psi engine?

Yes, of course. However, if the solar concentrator (SolarFire) delivers 150psi steam, the engine needs to be able to operate at that level. 150psi has a lot more energy in it that can be harvested as electricity.

- Mark

=Josh Jordan, July 4, 2011=

It would be essential to have this sensor in the development engine whatever the cost. It may not be required for a production engine. We can measure temperature very accurately and cheaply so it may be possible to calculate accurate enough pressure data from temperature readings.

=Mark Norton, July 4, 2011=

I agree. Can you find two more - even if they are more expensive? When sourcing such things, we need to show a certain amount of diligence. Stopping at the first one, even if it looks good, will not reveal possible better valves elsewhere.

- Mark

=Josh Jordan, July 4, 2011=

I found some alternatives by searching for 300psi and up. I added 2 more one goes up to 500psi and the other goes up to 5000 psi.

=Mark Norton, July 4, 2011=

Thank you. I've added a bit of formatting to the page, but it's good to have the links in place.

I've sent you a request to join the Pivotal project as well.

Why don't have you have a look at http://opensourceecology.org/wiki/Steam_Engine_Design/Rotating_Valve. This is a design that we're playing around with since identifying several problems with the Bump Valve design. Any comments you have are of interest to me.

- Mark

=Josh Jordan, July 4, 2011=

I can't really comment on the mechanical design. I would like to learn more in that area. I can help this project by providing pcb designs and firmware. I would like to manage my own project, I am considering the kitchen oven. The mechanical drawings should be simple and the electronic oven controller is really the heart of that project.

=Mark Norton, July 5, 2011=

I think that's great, Josh. Taking on one of the GVCS projects is a wonderful thing to do.

Ironically, the first thing you need to do when starting up a project like that is to do some research and capture it into the wiki, pretty much like what I asked you to do with the pressure sensor. If you haven't done so, have a look at http://opensourceecology.org/wiki/Managing_an_OSE_Project. It will give you some tips on being an OSE project manager.

I'd recommend doing some of the background work, first. Collect some data, put it in the wiki, and start organizing it. Then let Marcin know that you want to take on this project (what ever it is). The pre-work will give him more confidence that you can do this job and will stick with it.

Meanwhile, might need some help with PCB work and electronics, so I'll keep you on the Steam Engine project for now - if you still want to participate, that is.

- Mark

=Josh Jordan, July 7, 2011=

For the development of the steam engine it will pay to install as many sensors as possible. You never know what sensor data combinations will be useful to the optimal control process.

Here are some specs I suggest: arduino-compatible 3 pressure sensor inputs for each of the suggested pressure sensors, possible buffered, level shifted, or filtered as appropriate. 4 thermocouple inputs- highly accurate, high speed temperature sensors, requires a buffer chip and thermistor each. button inputs- use lever buttons to log moving parts accelerometer input- measure and log vibration and acceleration

Client interface specs: Log all of the sensor readings in a moving timeline like "system monitor" Javascript if virtual com port works in win/osx/linux, otherwise cross-platform.

We might be able to implement control systems and safety features later on. It might be possible to make a steam engine more mechanically simple than ever before by using digital control systems. If we had 1 well designed electric valve, it could be used in all the high pressure valves and in other designs with high pressure valves. I would still want a mechanical pressure release valve. I used to work at a place that manufactures steam cleaners for jewelers so I have a little bit of experience with high pressure steam applications.

-Josh Jordan

=Mark Norton, July 10, 2011=

I agree with the need to fully wire steam engine prototypes for data collection. You suggest a broader coverage than I, but that's ok - the more data, the better, IMO.

I like the thought of modular designs that can be assembled in different ways: a few different control valves, mechanical and electronic valve drivers, different cylinder bores, etc. It's like building a steam engine construction set that allows different models to be built. This allows us to experiment with different characteristics and work towards optimizing both the parts and the assemblies.

Regarding high pressure steam applications, what should we know? What are the challenges, safety considerations, operating parameters, etc. I'd like to capture this into the wiki into a new general section on steam engineering.

- Mark

=Josh Jordan, July 11, 2011=

It will be better to design the electronics with more inputs than we need because it will be useful in other applications. I think only a handful of electronic designs will be needed for all of the OSE projects if we take this approach. This design might only cost $10-15 in parts, its the sensors that will be most of the cost.

It would be interesting to see what computer modeling can do for steam engine design, if it is possible to beat the best engines from before diesel took over. I don't think it will be easy for the same reason it will be impossible to design the apollo rockets again. The most talented engineers of the day devoted themselves to that technology.

There is a certification that you need for high pressure applications, that is for industrial certification only, for consumers you need a different more strict certification. I am not sure of all the details, I worked with manufacturing jewelry steam cleaners over 10 years ago. For jewelers or pawns to have these in their store it was considered industrial use. It is a simple device, a cylinder tank wrapped in a heating element sleeve, and then wrapped in insulation. There is an electric valve controlled by a foot switch. Heating controlled by a simple thermostat. The electric valves failed all the time, that was the most sensitive part. The only other part is a pressure release valve for safety, this was required for the industrial certification. We tested blowing it up on purpose without the safety valve, the 1.5" steel cap on top where the water goes in came off and shot up thru the ceiling and stuck in the roof probably still there today. It would be smart to design a failure point like this so if it has to it can explode in a predictable way.   Later on they started making steam another way that did not require any certification at all.  They use aluminum tanks, heat them up, and pump water into the hot tank which instantly turns to steam without high pressure existing.  It still shot out the nozzle and cleaned jewelry good, maybe this way can still move a piston with this kind of low pressure steam and not need any certifications.

= Mark Norton, July 11, 2011=

I view the sensors (etc) as part of the prototyping effort. Some may still be needed on production machines, but the cost of instrumentation for experiment and evaluation purposes won't be applied to the eventual cost of the machine.

Regarding certification - this is an interesting aspect that will need to consider. Could you locate some links that point to what the cert requirements actually are? OSE steam engines will likely be found in two situations: consumer and industrial. The later will be neighborhood or village level power generation efforts.

How much pressure do you think the aluminum tank approach generated? Heat+Pressure=Energy, after all.

- Mark

= Josh Jordan, July 11, 2011=

I called them up but they stopped making the steam cleaners and don't know anything about them. The regulation is for pressurized pipes, and a steam tank is considered a pipe. If I remember correctly, we could not get the cleaners certified for consumer use because they just wont certify high pressure steam applications. I did a quick search for pressurized pipe regulations, it might be under OSHA but will require some more digging.

The aluminum steam engine could be attractive because aluminum is obtainable from clay and a lot of electricity. It is not a strong metal and probably would warp even at lower pressure. Have you looked at steam turbine technology? Maybe spinning a dc generator would be simpler than moving a piston and will work under more variable pressure and temperatures.

=Mark Norton, July 11, 2011=

I really want to capture the certification and regulation parameters as these could seriously shape the design and how it is used. I've created a story in Pivotal to capture this information. See https://www.pivotaltracker.com/story/show/15602059. Could you take this on for me?

- Mark Norton

= Josh Jordan, July 11, 2011=

Sure, I can look into the regulations. I won't be able to do much while I am out of town until Friday. Don't have my pivotal account with me.

= Mark Norton, July 11, 2011=

That's fine, Josh. I travel a fair bit, too (like this coming weekend). Thanks for taking this on, I appreciate it.

- Mark