5hp autonomous robot (FHR) -with drivetrain vid

P!-Ro

For a couple weeks now I've been making some plans and gathering parts for this robot, and last week I've actually started. But, before I got too far into the building of the project I thought I would get some advice from some of you first.

The robot will run off of a 5hp·Briggs & Stratton·engine I have. It will use chain drive to the wheels and uses a clutch that ketches when I speed the engine up to the correct rpm. However, this won't give me much leverage in the speed adjustment. Because of this, I'm planning of steeling some of Ken's robot power plant·plans (the link can be found here). I am hoping to use an alternator to generate power to run 24v motors for low speeds, and at high speeds use the chain drive. Because of this, I wont be able to use differential drive, so I got a small 18v screw driver for turning the front wheels (got the idea from a picture of a robotic lawnmower or something, they attached it to the front wheels). I also hope to get disk brakes like the quadrover, but I haven't done any searching for them yet.

This whole project is going to be controlled by·the propeller. Although I'm not good with it yet (as you can easily tell by this post) I am doing my best to learn how now. I'm hoping to give the robot r/c capabilities as well, hopefully making it safer so if something goes wrong, I can snatch control from the prop. Also, for those of you who are worried about the safety of this beast and the likelihood of my neighbor ending up with a hole in the trailorhouse, I plan to only let the robot go speeds at which it is able to stop in time if the sensors detect an obstacle. And if it miscalculates, oh well, good thing it will have a steel frame (maybe not so good for the obstacle though).

Any intelligent advice you have to offer as I build this thing would be greatly appreciated, and wish me luck that it works!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PG

Post Edited (Pi Guy) : 6/30/2009 2:25:52 AM GMT

Jim Cook

Hello Pi Guy,
I've been kicking a similar idea around for about a year now. I'm still in the thinking stage.
I hope to have a 10 -12 hp engine running a hydraulic pump. The output of the pump would be directed as needed to a hydraulic motor in each wheel. The controler would use servos to move the hydraulic valves to direct the flow.

P!-Ro

Although it would be fun, I don't think I'll be able to get a hydrolic system. I'm not quite old enough to get a regular job (coming March I will) and I just about emtied my savings on the engine and the clutch, not to mention I have yet to buy the wheels (about 70 bucks more). However, once I get the initial bot out, I'm planning on doing a lot of work with it to make it better. Plus, I'm hoping I will have enough money from my *potential* job by then. Right now though, I'm just going to stick with whats cheap til' I get more money in my pocket.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The Pi Guy

Ken Gracey

I LIKE PI,

Great project, for certain!

My advice is to skip any hydraulic drive robot, which you don't propose anyway. As you know, we've got some experience with both now and I find the hybrid approach (electric + gas) much easier to design. After all, so many of us are familiar with electronics not to mention the cost is much lower.

How are you going to couple the engine and alternator? If you don't have a lot of machining capabilities, just use a belt drive with the alternator. It's soo much easier than a direct connection with no slop.

Let me know how we can help. You're welcome to anything and everything I've done with robotics. What you're going to make should be a lot of fun. And a great way to learn the Prop, too.

Sincerely,

Ken Gracey

P!-Ro

Your help will be greatly appreciated Ken. As for your question about how it will be connected, belt will probably be best. I don't want my bot being too wide, so most large parts such as the alternator will probably be added to it's length or height. Although I probably won't get into the alternator very soon (I have yet to buy it) a question that has been on my mind for some time is how I will be able to control the power created by the alternator. This is because the setup will almost have to be completely opposite of the usual. Instead of needing the most power at high speeds, I will need it at the slower speeds before the clutch ketches, when I am using the DC motors.

Although I'm not sure what I'll do yet, I was thinking of disconnecting it at higher speeds to reduce friction and prevent frying my electronics, and instead use a smaller DC motor to make power. If you have any better ideas, tell me so I can make plans now before buying parts.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The Pi Guy

P!-Ro

Hey Ken, how did you control the power to the motors? Did you regulate it and run it through a motor controller, or did you do it some other way?

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The Pi Guy

Tom C

I LIKE PI,

This thread will answer all you questions concerning Ken's hybrid robot: http://forums.parallax.com/showthread.php?p=612409

Regards,
TCIII

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
If you are going to send·a Robot·to save the world, you·better make sure it likes it the way it is!

P!-Ro

Much better! The site I was originally looking at (the link to it is in my original post) seemed to skip the information I need. I'm not planning on getting to the electronics yet, but I would much rather have time to think about it first before I build it. Before I ever even started getting my parts like the engine, I wanted to thoroughly think through it first, and as a result it was about 3 weeks later that I finally bought the engine, and even later for the clutch and chain. It will probably be the same for the electronics as well.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The Pi Guy

P!-Ro

Here's a heads-up for how my project is doing. I got my sprocket ready for the axle, witch required me to attach the two sheared pieces,·and shrink the center·hole so it would fit the axle. Once that was done, I started cutting pieces of metal for the body. Then, today, I finished·cutting the rest of the metal and I started tacking it on. It's not done yet, but I hope to finish it by tomorrow and give it a good weld.·I'm hoping once I get this done I'll be able to mount·the engine and the axle, but it all depends on time. If I do get it done though, I'll probably get a picture of it for you guys.

Next week, I'm going to try and get the steering done and mount it on the front of the robot, as well as giving it a buggy-like top with two bars on either side going up and over, and having a link between the two supporting it so if the robot rolls, it will stay attached and not bend. Then I'll just give it·some·finishing touches and start working on the electronics part of it.

I hope this helps a little for those who are still confused at what I'm doing, and I'll send more information as this project comes along to help even more.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The Pi Guy

P!-Ro

OK guys, sorry I don't have the picture today, but I didn't feel like showing a bunch of parts on the shop floor. All I managed to build today is a rectangular frame and I cut a few pieces of metal so they are ready to be welded on for attaching my engine. I also bought two pillow blocks (at least I think that is what they are called, but these ones are a little different, with bolts attached for connections to the body). Anyway, I got them and will attach them to the frame Monday, along with the axle. Once I do this it should actually look like something, so I may take a picture of it then, for those waiting in anticipation.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The Pi Guy

P!-Ro

Now that most of the frame is done, I'm starting to wonder about that steering system. Everything up until now has not been a problem since my welding teacher could help me, but he doesn't have a clue on how to steer a vehicle with a screw driver (probably for a good reason, too) but I'm going to try anyway. Right now I'm worried about how I can #1 quickly turn the vehicle and #2 do it without causing the motor to stall. If you have suggestions or have heard of ways to do this, I would greatly appreciate it, because right now I'm fairly clueless, and I don't feel like testing and re-testing over and over just to find the best way to turn two wheels.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The Pi Guy

Lab Rat

i like your hydraulic idea.
i have always wanted to take a full size riding tractor (the hydraulic drive type) and make it remote controlfor fun

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
I`m not prejudice I hate everyone equally

P!-Ro

It's not actually going to be hydrolic, but chain drive. You'll see when I get the pictures in soon.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The Pi Guy

P!-Ro

Another question: how can I measure rpm? I know I can use this sensor http://www.parallax.com/Store/Sensors/PressureFlexRPM/tabid/177/CategoryID/52/List/0/Level/a/ProductID/90/Default.aspx?SortField=ProductName%2cProductName
but it is restricted to 7500 rpm, and I will also have to figure out how to attach the magnet so it won't go flying off. Anyone have better ideas for measuring the rpm?

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Pi Guy

P!-Ro

Ok guys, there has been a delay. Chapman, my teacher, doesn't have anymore 3/4" rod for me to use as an axle, so I will have to put off the building the rest of the drive system until the new order comes in. But, it won't restrict me from building the steering (or at least starting it) so I will be starting it soon. However, I still need to hear some ideas on how·I should build it before I start, so you 7+ people who I know visit this post daily need to turn on those light bulbs and give me some ideas so I can get started. I would really appreciate your help; 'cause things are dark right now.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Pi Guy

SRLM

For the magnet use some sort of mechanical mounting. A couple of bolts and a some fender washers would work. And is it likely that the motor will go over 7500 RPM? I think you may have a case of over engineering...

As for ideas, you have to provide a summary of what you need ideas on. You have quite a few posts in this thread, so if you could give a nice, ordered list perhaps a reply would be forthcomming.

P!-Ro

Right now I'm hoping to just use IR sensors and a white line on my clutch to reduce this speed and mounting problem. With the steering, I've been thinking about it lately and I've come up with an idea, but if any of you think of something different, please tell me.

What I have thought of for the steering system is attaching a long bolt with a nut on it to the drill chuck. Then I'll weld a small rod to it so when the drill spins it moves it back and forth. Once I get this motion out of the drill, I can hook it up to the front wheels the way it is normally done.

I'm sure this design will have many issues, which is why I want some input, but if I don't get responses I'll probably just try it and modify it as I go, depending on the problems I have.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Pi Guy

Ken Gracey

Hey I LIKE PI,

I've seen the drill approach work successfully for steering, but I think there are alternatives. Maybe you can find a low-cost linear actuator? Have you looked in eBay for linear actuators? I think it could look more refined if you found one of them instead of hacking a drill.

Ken Gracey

P!-Ro

Wow, a little pricey there, Ken, but it gave me an idea to simplify my design. Since I have some old 12v motors sitting around, I might just attach the motor straight to the bolt. This is will make things simpler, and since it will automatically gear down when you use the screw and bolt, I shouldn't need to use the gear train of the drill.

I'll work on these plans this weekend while I'm on the bus going to a marching band competition on Saturday. Then I will start working on it Monday if I finally get the drive train done. (I got virtually nothing done today since the stupid attendance office mixed up my name and thought I was supposed to leave during fifth hour, so after I solved the mess I had run out of much of my time in welding. When I finally got back, I still had to change the oxygen to use the oxy-acetelyne torch. I needed it to cut the piece I wanted, but since there wasn't enough time, I never getting anything done.) Anyway, hope the drive train gets done tomorrow, because there is no school for me on Friday.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Pi Guy

P!-Ro

I've now made the devise I want to detect RPM. It's like the QTI sensor found here. Basically all it does is count how many times the white stripe on my clutch passes the sensor. Since my current prop isn't working, it is connected to my BS2px. When I have time I am hoping to create the code for it and try it out on the engine.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Pi Guy

P!-Ro

I've got a problem guys. I've been asked by many people in my welding class what I want to do with my robot and I don't know what to tell them. Any Ideas?

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Pi Guy

theloneduck

kill all humans? look at some of the completed projects and replicate those? i.e. mow the lawn, carry golf bags, etc.

PJAllen

pi Guy said...
I've got a problem guys. I've been asked by many people in my welding class what I want to do with my robot and I don't know what to tell them. Any Ideas?

I don't get it.·

Do you not know your objective?·

Post Edit -- If you don't want to give details, then don't.

Post Edited (PJ Allen) : 10/23/2008 12:11:39 PM GMT

Ken Gracey

Your objective needn't be to solve a particular problem using a robot. It could be simply to learn about what's possible in this project, perhaps considering your newly acquired skills for your next project. Do you have any idea how many robots I've made that serve no particular purpose? You don't have to provide an answer just because you are being asked to. Curiosity is an attribute as valuable as motivation.

Ken Gracey
Parallax, Inc.

P!-Ro

Great advice, Ken. Beats PJ's mocking (just joking, PJ, I don't mind). The only reason I started the project was because I wanted a robot like the Quadrover. The only problem is, just because you have a Quadrover, doesn't mean you know what to do with it. That is why I asked the question in the first place.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Pi Guy

PJAllen

Mocking?· (Oh, please.)· What mocking?

You didn't ask what you should do with your project.· You asked what you should tell others about it, that·you didn't know what to tell them.

Searider

I agree with Ken. My robot projects tend to be "experiemntal Platforms" in that I want to explore a sensor or feature or gadget and I add it to my platform. I tend to end up with a eclectic collection and sometimes I have to remove a previous addition to make room for what I am working on next.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔

---

Searider

---

P!-Ro

Pj, I didn't know what to tell them what it was FOR, so you can stop your mocking (despite your complaining, it is still mocking).
·
Anyway, you should start listening to people like Searider, who know what their talking about.
·
@Searider, I fully agree with you (and Ken) that it can be a real help to make project bases to help you make your next robot or have it easy to change to your needs. Personally, I am going for the second choice since this has already cost me over a hundred $ and I haven't even bought the wheels yet ($70) OR the electronics (I hope to be able to afford a prop protoboard by the end of the week). What also helps is since I am in welding, I can basically make what I want for it without having to find and buy it. Too bad it takes so long to do it though.
·
Also, for those of you that are mocking me because I didn't think about what I wanted to do with my project (no names being said here) I did. Only problem is that my robot outgrew what I originally thought it capable of. One easy way to describe it's capabilities for you is to compare it to the Quadrover.
·
-The Quadrover has a 2.5hp engine--mine has 5
-The Quadrovers top speed is 10mph with hydraulics--my engine has been used in go carts going 30mph with a passenger in it, and mine will not, cutting out over half the weight. It also has chain drive, taking out the friction problems of hydraulics.
·
I could probably keep comparing the two for hours, but I'd probably say lies since I still don't know what my robot can do and what I can afford to put on it. But that doesn't stop me having high thoughts for it! If only I knew what to make it do...
·
(p.s., listen to theloneduck, he has some good ideas...x^(

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Pi Guy

PJAllen

PiGuy, you didn't ask what you should do with your project.· You asked what you should tell others about it, that·you didn't know what to tell them.

i like pi said...
I've been asked by many people in my welding class what I want to do with my robot and I don't know what to tell them. Any Ideas?

You can go back and re-read my Reply.· I asked you whether you knew your objective.·

PJ Allen said...

I don't get it.· Do you not know your objective?

You've conceded that you did not.· At this point that does not surprise me.· Nevertheless, that's entirely your problem.

If you still need some ideas for what you should do with your project...··... I·could give you one, alright.

P!-Ro

Okay, I should be able to·buy the protoboard tomorrow, then work on it when it comes in. I'll get that rpm figured out and display it on the lcd if I can get it to work. I would like to use one of Parallax's nice motor controllers, but it would cost too much, so I plan on using transistors. So if any of you know a thing or two about transistors and H-bridges, it would be helpful.

I'm also looking into disk brakes right now, and it would be great knowing a thing or two about that also before I buy anything. I'll probably go to Elmers tomorrow and look into seeing if they have them (Elmers has many parts for small vehicles like go-carts and four-wheelers).

Since I never bothered to tell you earlier, my drive system is almost done now. All I have to do is finish attaching the axle and sprocket. I know it sounds like a lot, but I already got all the cutting and drilling done, so it won't take long.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Pi Guy

Post Edited (I LIKE PI) : 11/9/2008 8:03:04 PM GMT

P!-Ro

I was just watching some videos of the quadrover, and suddenly my project seemed much more ambitious than I originally thought, with twice the power and over twice the size. I'm excited though, because soon I'll at least have part of the bot working. My drive train is almost done, and once I get that I'll start looking into wheels as well. I still need to get in steering, but at least that is a start. I'll try to get you guys a picture tomorrow (if I remember) weather the drive train is finished or not. If it isn't, I'll probably just put the peaces were they go and take a pic anyway, just to give you guys some ideas so you can actually give me some help!

I'm sorry it took me so long to finally get this much of it done, but this is my first major project (besides my eagle) and the only one I've done in the engineering area. Wish me luck·that·I make good progress tomorrow,·or with my fortune·it won't get done til' next week!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Pi Guy