Pulse trains producing "s-curves" for sign wave motion

[osmanmom]

your size of simulator will depend on linear shaft whether in unit kg/f , newton ,lbs

please check the spec of linear shaft motor

the price will depend on force and accuracy

 

[Lamboom]

Thanks Again! I'm on it!

 

[osmanmom]

http://www.linmot-usa.com/

http://www.youtube.com/watch?v=Ru69MRC4Vos

 

[Lamboom]

osmanmom - Ya know what you Are telling me... I've spent a long time trying to get mechanical actuators to do linear sine motion ... when a more practical solution may be the linear motor (something i wasn't even aware of ... sigh!) Thanks a lot for this information. I will seriously look into this alternative technology ... by the way, "billions" of cycles is what one needs in wave generation .. "Millions" of cycles isn't good enough .. and that's the best the ball bearing actuators can do. ... to be continued Regards, Michael

 

[Lamboom]

Hi Peter
Doesn't Ohmset's wave generator use a hydraulic driven wave maker? Your Motion Controller sounds more like a computer if it is "generating" the random signal for the hydraulic servo controlled actuator(s) at Ohmset. I don't need something that sophisticated. As a simple tool for Education, or exhibiting an ocean project, the random waves can represent an addition of only 8 to 15 regular sine waves. Each carefully selected to approximate a desired spectral distribution. It's not truly "random", but the visible sea looks good, and can be used to illustrate performance of model structures. I think you understand what I'm trying to do. I have to keep the finished system at an affordable cost for institutions that might want it. As you know, the United States doesn't spend much money on education or science, so .. I may be on a fool's errand....

 

[Peter Nachtwey]

Moog, formerly California Linear Devices has something like that

http://www.moog.com/products/motors-servomotors/linear-motors/
It is basically a solenoid. The problem with these kinds of systems is that they take power to hold position but there are no bearings to wear out. They can replace hydraulic actuators on smaller applications.

The pdf that osmanmom posted is great stuff. I download this pdf and forwarded it to the person here that worked on this project.
https://www.sfos.uaf.edu/directory/faculty/kowalik/Tsunami_Book/book_sum.pdf
The document shows equations with hypderbolic functions. I remember that we had to add hyperbolic functions to our controller to do the Ohmsett application.

I just saw your post. The Moog-CLD linear drives are not cheap. I don't think anything like this will be cheap but they are what you want for smaller loads.
Yes, Ohmsett used hydraulic cylinders because the loads were high. The Ohmsett wave generator is pushing a lot of water around.

Your Motion Controller sounds more like a computer if it is "generating" the random signal for the hydraulic servo controlled actuator(s) at Ohmset.

yes, the equation like or similar to those posted in the pdf are executed on our controller to generate the motion profiles. Like I said, we had to add the hyperbolic functions for this project. The sum of wave function results in a motion profile for the actuator to follow but to do a better job our controller also generates the velocity and acceleration feed forwards to reduce following error.

I don't need something that sophisticated.

But so much is already done. If you bought a PLC you would be starting from scratch. You would spend a lot of time reinventing the wheel. If you use a PLC you will only be able to execute pre-calculated wave forms and I doubt you would be able to execute them well without a motion controller anyway.
In the end you would NEVER be able to do close to what the Ohmsett system can do. It is a trade off, money for time and capabilities.

What about the software for analyzing data? Is there any need for data logging? We can send real time trends to a PC at 0.5 millisecond time increments.
See this:
http://deltamotion.com/peter/Videos/PneuMove2.mp4
It is a pneumatic application but we are generating sine waves.
What are you going to do with your data?

 

[Lamboom]

Thank you, Peter ... The Moog is a beast of a linear motor. I will check it out (looks like the smallest one would be perfect .. at a weight of 31 pounds ! ...

 

[Peter Nachtwey]

I was just talking to the engineer here that worked on the Ohmset project. This engineer got the formulas from Neal Brown and did all the programming. The NDA applied to Neal Brown's formulas. If you have access to Neal Brown and can use his formulas you could save a lot of time.

We looked at your website. It also looks like you get into special effects but there is one short video where a rocket goes a long distance and blows up something. That doesn't look fake.

 

[Lamboom]

Peter ... Neal did offer to modify the calc's for my little tanks; however, I didn't want him to do all that work until I had a finished design, and an actual customer for a model ocean flume suitable for serious science. Over the years I have built such systems for myself and a few of the big players in the oil industry (Exxon, Earl & Wright .. Brown & Root) .. I would like to develop a system that a High School might afford, or one a college could certainly afford. The Tsunami tank at The Museum of Science and Industry was ridiculously expensive … and yet, is very limited. That's what you get with a "design-by-committee". My little 1/3 MSI scale system can out perform that tank in every way imaginable, for 1/25th the cost.

As for that RPG... I'm sorry that spurious stuff is on the Micro Oceans site... it will soon be removed, or made non accessible to the public without password. I spent only a few hours building the site ... it's out of date, and not very professional.

I Did SFX for Television, Movies & Theater while working at the Model Ocean Basin, as a consulting engineer, in Escondido. About 11 years ago, I helped start this company: http://www.strategic-operations.com/ I developed the pyrotechnics, and managed the Battlefield Effects division, using my licenses with the state, and the Fed's. Needles to say... it seriously interfered with my work at the model basin.. (I also worked on several projects with Neal Brown) .. I am now trying to work exclusively with Micro Oceans.

If you watch the videos on the Strategic Operations web site.. you will see how the rockets work .. the idea was to make it look real, but not hurt anyone .. unlike film work, you never know what the military is going to do, so it was quite a challenge...

 

[Lamboom]

Peter-
What you are saying about the PLC is quite true. There is a lot of work in pre-design to get the motions required for the desired wave. What goes in, is not necessarily what comes out. Especially in an open system. I believe the same is true with hydraulics.

Then, there's your video on pneumatic motion control. That is truly an accomplishment. Very impressive. Your motion controller is a serious computer. And, I suspect, the valves and rod-less actuator are of a high quality. So much to compensate for: irregular friction, fluctuations in available pressure to the valves, inertia of the moving carriage, resonance .. and so on. Nicely done.

You have made your point. I drool on my keyboard watching that video.

Then again, you must appreciate that the water surface is quite forgiving. Generally all that matters is the height and frequency (or Period, marine engineers say) of the regular wave, when you are testing in only regular waves. Data such at wave height, wave period and any other important information as a function of time, about the object exposed to this ocean environment, is recorded. And, charts are generated showing each value or interest as a function of a range of periods. Motion data (pitch, roll, heave, surge & sway) force data: moments and structural forces..displacement, velocity and acceleration forces are logged. When the Random Sea is run, then things get tricky. Spectrum analysis must be run for the waves where the model will be placed, and adjustments must be made to bring the spectral curve into alignment with the desired, theoretical distribution of ocean energy.

There are limits to what reasonably can be had from model testing in the scale of 1:200. (Ohmset likes to work in 1:1 to about 1:50

I was going to design a tank for an Arabian prince, that was building a science museum in Saudi Arabia … my partners and I didn't take the bait .. but, that's another story. I'm not really interested in that scale of a project. I want to find out if there is any interest by any US institution (educational or other) in a way to accurately model the ocean surface for scientific demonstration or study … at a fairly low cost. Bigger and better, would cost more … and for that, I suspect I could afford your motion controller …

Thanks again for your time and interest.
Regards, Michael

 

[Highland Controls]

Any motion controller, even some indexers, should be able to do this. A triangular velocity profile produces a sine position profile. The calculations are pretty simple. The trick is to have a controller that can do the continuous motion. If it is an indexer, you need to be able to chain the moves together. Waiting for a move to complete then starting the next one(from some kind of host program) won't work.

 

[Lamboom]

Hi Highland Controls ... thanks for the info. I'll be looking at motion controllers. The trick is to select an actuator, perhaps a linear motor, with enough force to drive the wave generator (say, 8 pounds) in fact this thread is all about the "specific components" that might work together. I started it with the question: what's out there that has a pulse train output that will talk to the IAI PCON PL/PO driver? (it's an open ended driver that runs a linear actuator following a pattern of pulses which would produce sine wave motion. Probably not too precise, and would take half a lifetime to program, in order to obtain an accurate series of different regular waves.

Thanks to this forum .. I have learned something about motion controllers, and linear motors. It changes everything from the way I was planning to do this ... sometimes You just have to start over...

For now... I need to research specific equipment, based on power, force and velocity specs ... not to mention the cost .... to be continued

Regards, Michael.

 

[Lamboom]

Peter.. Are you suggesting the RMC 70, talking to the driver of a linear motor?

Michael

 

[Lemming]

Hi, have been watching this thread with interest... Haven't done it myself, but wonder if the control could be accomplished by using a Micro1400, and driving a couple of PTO instructions. They will provide the pulse count out, but would have to be driven on a X time sample of the wave period. Some maths in the back ground would generate the pulse quantity to sort the amplitude ??

I have a 1400 lying around, so will have a play. I am sure it wont be as nice as a dedicated motion controller or like..

 

[Lamboom]

Hi Lemming ... thanks for the input. I suspect that nothing will be as good as a motion controller ... with the exception of a variable speed DC motor, spinning a pin, with a remotely adjustable radius ... in a slot. You would call it a "scotch yoke" .. and it would produce perfect sine waves of any amplitude, and period... That's what I'll use after a massive CME takes out the grid .. and we're back to the stone age.

However, in the meantime, I'm going to continue with the plan to do this electronically, with nice computer screens, good data acquisition ... and touchscreen control of basic parameters.

I'm certain there is a linear Motor, and a Motion Controller in my project's future. I will have to spend serious time in the selection of required components. In the meantime ... it would be nice to see what could be done with pulse trains. Thanks much for looking into this.
Regards, Michael

PS: I assume you are talking about Rockwell's MicroLogix 1400 Control System?