If you are 'into' electronics these days, you'd better have good eyesight: have you looked at the SMT technology? I just purchased a Netduino and the soldering must have been done by 6" elves.
Printable View
If you are 'into' electronics these days, you'd better have good eyesight: have you looked at the SMT technology? I just purchased a Netduino and the soldering must have been done by 6" elves.
lol, these boards are not done by hand, they are done by machines/robots. There are companies where you can submit your schematic and PCM board design, and they make the whole thing for you on mass. They can even give you an estimate before you give it a "go"... You can do the submission online.
Do you see any 6" elves walking around? No, because they are slaving away in these factories ;)Quote:
Originally Posted by CVMichael
haha, that's a good one :D
By the way, here is an example of what I was talking about in the previous post:
http://www.futurlec.com/PCBService.shtml
Under the "Services" you can see "PCB Manufacture", "PCB Design", "Board Assembly".
I'm currently designing a PCB :)
http://1.bp.blogspot.com/-TuZt70hu0a...rotopeller.png
http://github.com/AdamWard/Protopeller
It's Open Source too, so help yourselves. :)
I use DipTrace to make my schematics & PCBs... http://www.diptrace.com/ It's free for 300 pins.
What I like about it is that it's very easy to use, and when I want to make a PCB for my schematic, it "links" them together, so that if I make changes to the schematic, I can tell it to update the changes to the PCB.
I've got the full version of Diptrace :)
It's great but there are some annoying bugs I could do without.
Yep, i do some electronics. I love PIC's. I made some FM radios (i was so happy when i switched them on and i heard our local stations), then did some LED / LAMP flashers, some power supplies (regulated 0..12V, 5V), now planning to make dual +-15V for OP amps projects.
Also i love building amplifiers, i have made C70 amplifier (search in Google) designed PCB's by my own (because there isn't any public PCB for it) and now im looking for good VU meter, etc. Lot's of work to finally complete it. For my ears it sounds incredible. I Tested it with the best speakers i had around and sound is so good with such a good bass. Even preamplifier isn't needed, which is good actually. It will keep sound cleaner and nicer.
I will try to get my site running asap and can show some pics and vids.
Using 2N3055? I love those transistors!
Yep Michael, with 2N3055.
Do you also love building amps or amps in general?
I have built plenty of them, also when i started with electronics, some simple, some harder. Not all of them started working at first time or worked at all.
Years passed and now i built C70 and it started working with the first time.
I carefully designed PCB to avoid any errors, it's not best, but it works.
Actually i already started that project on summer 2010, but stopped at some time, now i picked it up again.
I have to do some things for it. VU meter, tone regulators, speaker protection and maybe some other things i forgot.
Im putting this all into old amplifier case.
I don't know exactly atm, how i will do the front panel.
I hope you guys are OK if I revive this thread?
Here is the latest thing I made: http://www.youtube.com/watch?v=L3Dpe1v6G-w
Soon I will be working on a laser range finder... can't wait! Tomorrow I am getting the Linear CCD Arrays, and I am hoping to have the laser finder done in a few weeks.
That is awesome! Wish I was good at electronics then I could create a robot to keep our kitten entertained.
I used to have a cat, so I know what you mean. I wish I had a cat now :cry:Quote:
Originally Posted by Nightwalker83
Well, something relatively easy you can do is get an Arduino (because I hear it's easy to learn for beginnners), and 2 servo motors.
Connect the shaft from one servo to the body of the second servo. This will give you pitch and yaw control. Then connect a 5mw laser pointer to the second servo pointing forward.
It will look the same as the servos I have in front of my robot, except instead of the ultrasound sensor, it will have the laser pointer.
[edit] Take a look at this: http://www.odec.ca/projects/2004/kra...ml/laser1.html
Tilt the whole thing to point the laser on the floor, and write a simple algoritm to move both servos randomly.
This will keep your cat in a good shape for sure :D
As for the laser range finder: Would it be cheap enough for hobby work? I haven't tried the new ultrasonic sensor I picked up (after concluding the SRF08 was wildly useless for anything beyond about 50cm), since I am currently working exclusively on brain software, but the key to the brain will be some kind of sensor that can tell that object O is N distance away with some kind of accuracy. This can be one O in front of the bot, or pretty much any number in any direction. There was some rumor of a low cost, 360 degree laser range finder coming out based on some new vacuum competitor to the Roomba, but as of a month ago, I see nothing on the market. Low-power laser range finders are all $1000+, which is a bit much for a toy. The XV-11 rangefinder looks great:
http://www.ros.org/news/2011/02/neat...er-driver.html
http://www.hizook.com/blog/2009/12/2...neato-robotics
Except that the device is not nearly as cheap as it was supposed to be. The vacuum is only $400, and the laser range finder can be picked up for $200, but that means that it could be considerably cheaper, and that's what I'd like to see.
Can you explain that a bit more? because it's first time to hear about it.Quote:
Originally Posted by Shaggy Hiker
About the laser range finder. I did not do this yet, I will get the linear CCD arrays tomorrow. But here's how I plan to do it.
The CCD array is ELIS-1024A-LG, it's a surface mount, 1024 pixel linear array. It works at 20MHz. I will use a microcontroller to read the 1024 pixels, and calculate the distance based on triangulation. It will have to see where the laser dot is, and based on that to do the calculation.
Because the CCD array works at 20MHz, I am hoping to read the distance at a few hundred times per second. Of course it all depends on how fast the microcontroller is. So basically the top speed will be 20,000,000 / (1024 + (however long it takes to calculate the distance)). So the distance read speed really depends on microcontroller, but I'm hoping something between 300 and 1000 readings per second.
I will make the PCB board with the laser finder, and assemble it on a stepper motor. When the stepper motor moves each step, it will do a range reading. This way I can read the distance at 0.9 degrees, in a full circle (360 degree), therefore 400 steps from the stepper motor and therefore 400 range readings.
So this will be horizontal reading. Next step will be to add another stepper motor that will move the laser range finder vertically.
This way I am hoping to be able to map in 3D everything around the robot.
By the way... In terms of price: I bought the ELIS-1024A-LG for $13.50, but including shipping & handling it's about $30 a piece. All parts added together I don't think it's more than $100 for the entire range finder.
Yeah. What's the power demand?
Frankly, the device you describe would be incredibly valuable in the hobby robot world (which probably means that there isn't enough demand to make it worth selling, unfortunately). I'd sure like one, but I'm not at the level with hardware that I could build one myself without some serious instructions.
As for the brain, this thread discusses the basics:
http://www.vbforums.com/showthread.php?t=598244
however, you can see that there was a HUGE gap in the middle. That was due to me working on a different project that was absorbing all of my time...and then it was canceled. So, now I am back to hobby work. Inline with the last post in that thread, I began working on a program to design a pseudo-neural net. While I am still adding occasional cosmetic features, that part is pretty much complete. That brain designer will allow me to rapidly (and mostly graphically) design brains, such that the code underlying the brain will already be tested, which makes it more reliable. That just leaves two difficult problems:
1) How to design the brain.
2) How to talk about it.
Last night, as I was drifting off to sleep, it occured to me that I need to be able to have single neurons be replaceable by entire brains. That's a remarkably easy change to make, and will allow layers of brain activity. For example, the first brain, which will test the concept, will just be focused on training the bot to figure out how to turn, how to turn on a slippery surface, and how to turn when one track is blocked (such as by a carpet threshold, which will block this bot if it hits it at an oblique angle). That's a relatively simple problem that can be a discrete part of the brain, so it makes a good case to test whether learning can happen, as well as testing performance of the brain (a brain with N neurons takes X milliseconds, so I can figure out the timing cost of a much larger brain).
However, a higher level could then be written completely independently, then hooked into the existing Turning brain such that it would be saying "instead of this input, perform this whole brain load worth of stuff." That way, I can take an existing, working, brain, and build a whole other layer that can be dropped in dynamically to radically change how any particular decision is arrived at.
Talking about the brain is almost as hard as designing it, though. The Brain Designer tool is pretty cool, but of what value is that if a person doesn't understand what it is building? It is like a neural net, and could create something like a neural net, except that it isn't a neural net at all. It is also somewhat like a neural net where every connection is, itself, a neural net, or at least could be. But why this should work is more a matter of faith, at this point. I believe I have to actually develop some test cases, such as that Turning brain, just so that I can figure out the language to describe what it does, and whether it is feasible. After all, I rather would like to put the source code out for anyone to use, but putting out the Brain Designer in the CodeBank would be almost worse than useless, as it would be a highly complex, specialized, tool with no suggestion as to where or how it should be used. Until I can discuss some techniques for using it to design a brain, it won't be of much value, and I can't discuss that until I understand it.
After all, the real goal is to make a learning engine, then let it learn. I'm not nearly as interested on WHAT it learns as I am interested in WHETHER it learns, and how flexible it can be even with a rigid mind.
Just FYI:
Lately I am working on projects that require fast processing. I was using PIC18 microcontroller but it's too slow for some things I want to do, so I decided to upgrade. I thought if I go next in line (PIC24) then eventually I'll have the same problem. So I got the best PIC there is at the moment... PIC32MX795F512L
I bought the chip from Digi-Key, I put it on a test PCB board, I made a small program (LED test), but when I was about to write the program to the micro, I discovered that my ICD2 does not work for the latest PIC32s... I bought the ICD2 back in 2007, since then they made a new one ICD3.
So basically my choice was to get a PIC that ICD2 can program or get the new ICD3... so... I purchased the ICD3, it was $244 (with shipping + tax), I will receive it tomorrow. Can't wait to get it :)
My next dilemma after this will be the compiler... Right now I use mikroC PRO for the 8 bit PICs I currently use, but for PIC32 the compiler is different. That's another $250 :(
PICs are expensive!
Shaggy Hiker
Are you still working on your "brain" project? Because my goal is to make a robot, and it sure could use a brain :)
Yeah, I'm on it more than ever. I've run into a problem, though, which will take some thought: It is possible to create a standard Neural Net as a single class. The class would take the number of inputs, the number of outputs, and the number of hidden layers as arguments to the constructor. The rest can all be done in code. That makes NN kind of easy to get your head around. I have built my Brain Designer, but I have yet to come up with even a vague rule of thumb as to how to use the thing, which means that I can't let anyone use it, yet, because I can't explain what a person is supposed to do with it. Heck, I'm not even sure, yet, myself. So, the result is that I have an idea, and some software based on the idea, and it is pretty cool, but I can't explain it, so it is technically useless.
My next step is to try it out in some scenarios to try to figure out some guidelines as to how to use it. It's an interesting situation. If you have an idea that you can't describe, is it really an idea? Perhaps a concept is only valuable when you are able to communicate it to somebody else.
http://cdn.memegenerator.net/instanc...x/15222587.jpg
I voted for... all of them..
Well, this is the Chit Chat forum... :ehh:Quote:
Originally Posted by Icyculyr
I recently got serious into electronics. The combination of taking things apart to see how they work while growing up, taking apart old video game consoles to fix em up, and my love for youtube has made me wanna self educate myself and jump into a new hobby. Now I have a huge project on my hands. I currently own 3 NES's, only I plan on modding one big time. With the controllers for example, I know where 3 power sources are. I used one of them and soldered an LED light and resistor on one to light up the Nintendo logo. It works. I plan on doing the same for the B and A buttons. Then Im gonna spray paint my controllers (using acrylic enamel) with a couple coats of primer, 5 coats of black color paint, and a couple coats of clear coat. Each coat will take 2 days to cure and in between after drying I gotta keep sanding it down with various grits of sandpaper. In the end its gonna make it so shiny, it'll be like a black mirror. The same will be done to the NES itself.
As for the NES electronically, I'm gonna convert the mono RCA to stereo. Then I'm gonna light up the inside with either LED lights or probably those lights they use inside gamer PC cases. I might even add a window on top if I choose. In the end. It'll be as shiny and pretty as this:
http://t2.gstatic.com/images?q=tbn:A...Py4x_yB2NAogOg
All this for the new JR television show: Pimp my Game System.
I was an electronics hobbyist in school - I wanted to become a electronics engineer.Quote:
Originally Posted by wossname
Unfortunately, I fell into the trap of programming.....:eek2:
I should =OQuote:
Originally Posted by Shaggy Hiker