Show Posts
Pages: 1 ... 4 5 [6] 7 8 ... 375
76  Topics / Robotics / Re: [Guide] How to make a three wheels moving platform on: March 20, 2014, 01:30:12 pm
Interesting tutorial, but this build doesn't appear to be inexpensive... you have a complete BOM for reference?
77  Using Arduino / Motors, Mechanics, and Power / Re: How to increase amperage from seed studio motor shield. on: March 20, 2014, 01:24:43 pm
You cannot get much more than 1 ampere out of the L298 chip before it starts to overheat and shut down.

Sure you can; you can get close to 2 amps (per channel - close to 4 in parallel mode) - but it has to be properly configured - mainly (besides the electrical connections) a fairly large heatsink (much bigger than the one on that board), and quite possibly forced air (fan) cooling.

From what I understand, the L298 (and the L293, to an extent) was originally targeted at the automotive market, for electrical actuators, in which case it was very easy to apply a large heatsink (thus no fan needed).

It is ancient technology. I recommend modern motor drivers, like those sold by Pololu.

I suppose "ancient" in an electronics sense, if not a human since (if that were the case, I'd be ancient, am I'm only 40 years old, dammit!)...

I will agree with the need/want to use a more modern mosfet driver approach (though it's a pity from a breadboarding perspective that none of these seem to be available in thru-hole versions, outside of a discrete mosfet and DIP mosfet driver h-bridge design).

I guess I am just a curmudgeon when it comes to certain aspect of electronics and robotics; lately it seems that so much of it has become "Lego-like", with the focus more on the software aspect rather than the mechnical/electronic aspect. I see the pros and cons for both, and thus can't see either way as better or worse...
78  Using Arduino / Sensors / Re: Potentiometer that snaps back into position? on: March 19, 2014, 12:58:05 pm
Are there potentiometers that are built so that they snap back after the person lets go, or is that something I'm going to have to make on my own?

This would probably be something you would have to build.

Motorized potentiometers have already been mentioned, but what you are describing sounds more like a spring return system; maybe a pre-tensioned coil spring around the pot's shaft might work? Or a tension spring with a cord of some sort wound on the pot's shaft?

At any rate, this is similar to how some hobby RC transmitter joysticks, as well as some older computer joysticks (as well as current thumbsticks) are setup, except they are designed to "return-to-center". Something to keep in mind, should you pursue this mechanism, is that your initial beginning value may not be the same as the later values when the pot is returned back to the beginning by the spring. You will need to take care of this in software, or some other fashion, most likely. Back in the earlier PC days, you had to go thru a "calibration" routine with joystick games to set the proper max/min positions and center of the joystick so it would work properly with the games.
79  Using Arduino / Motors, Mechanics, and Power / Re: Need help with connecting gear motors to arduino uno for robotic dog on: March 19, 2014, 12:50:00 pm
Here I was hoping for someone building a re-incarnation of Frank DaCosta's "pet" (TAB Books 1141):
80  Using Arduino / General Electronics / Re: Need urgent help for my motor drive design. on: March 19, 2014, 12:42:11 pm
I'm looking at a datasheet here for the ULN2003A:

Page 12 shows some sample app schematics. If you look at the output side of figure 19 or 22, you'll see a couple of differences to your schematic.

First - the high-side of your coil isn't connected to the ULN2003A common. Also, your ULN2003A ground - pin 8 - isn't connected (actually, pin 8 isn't even on the diagram?). Finally, you probably won't need the flyback diode across the coil, if connected as in the sample application schematics (the schematics don't show one).

I would also second the suggestion of going with a "true" h-bridge design using two SPDT relays instead.

If you need the ability of free-wheeling, a third SPST relay can be put in-line with the low-side ground to enable that ability (otherwise, you always have "brake" with the dual SPDT design). If you instead want speed control (while still having a relay h-bridge), you could put an n-channel mosfet driver on the low-side before the ground, and PWM that (make sure to pick a proper mosfet with the appropriate current handling - and don't PWM the speed while switching the relays, unless you want to weld your contacts).
81  Topics / Robotics / Re: Arduino 6 wheel rover Help! Urgent for school!!! on: March 19, 2014, 12:01:36 am
Ok - based on what I could see on that HobbyKing product listing, that chassis has the 1:34 motors, which are detailed with specs at Pololu:

The motors are 6 volt motors with a stall current of 6 amps.

So - for three motors (per side) - that would be 18 amps of stall current. Here's what I would do:

Rather than purchasing an h-bridge driver setup for each side, instead purchase a couple of these:

Then run them off of 5-cell NiMH R/C packs (6 volts) - these may have to be custom made; you could also try some kind of LiFePo or LiPoly packs (though the latter will be an issue because a 2 cell pack will output over 7 volts - which will cause the motors to wear out much faster).

Using a pair of ESCs rather than an h-bridges means you can easily control the speed of the motors using the Servo library, plus such ESCs will likely be a cheaper option. Something to think about, anyhow...
82  Using Arduino / Sensors / Re: Very low friction potentiometer or encoder on: March 18, 2014, 11:39:28 pm
One option might be to use what is called a "servo potentiometer". This is a potentiometer which can be rotated continously 360 degrees; there is a small "dead-band" gap or break that the wiper will travel over. Bourns is one well-known manufacturer of these potentiometers:

Be aware, though, that these kinds of pots aren't cheap (even used/surplus - unless you get really lucky) - they are precision devices, and typically will have ball-bearing construction. You can also order them "ganged", with multiple such devices on a common shaft (so you could place two 180 degrees out of phase, to help counteract the "gap").
83  Community / Bar Sport / Re: Bitcoin miner - Is it possible? on: March 16, 2014, 01:43:36 pm

That's the thing about cryptocurrencies - they are just as ephemeral as what we already use, so why can't we all just agree to use such a thing? It's the same "gentleman's agreement" that we already operate under, but it does have some upsides...

A modern high performance graphics card has thousands of times better performance, memory and bandwidth available and even so will only earn cents per day.

For bitcoin - maybe (it seems like it is even less than that, honestly - which is why people have switched to ASIC miners); for other (newer) cryptocurrencies, you can actually make a fair amount per day from what I understand. I have a friend who has set up a small rig (30 or so GPUs) in his garage mining some of the alternate cryptocurrencies that are easier to mine; he's managed to make a few hundred dollars (if he is lucky, he may break even on his investment).

The second way is to sell mining equipment. Just like a gold rush - the most reliable way to get rich is by selling shovels.

You definitely have that right; my friend is mainly in it to understand how all of this works - with maybe a minor hope that he hits on the "next bitcoin" (that is, the next cryptocurrency to have a bubble) - by mining for many different alternatives. In other words, maybe gold is played out for most mines, but maybe copper is easier to get and will become worth more. Divestment of efforts, resources, etc - in the hopes that maybe one or more hits just as big. I don't really see him getting lucky like that, but in the meantime he'll probably learn a lot in the end.

For what it's worth, I don't see Bitcoins ever being accepted for general use in their current form because they are completely unaccountable - it's the electronic equivalent of leaving banknotes under the park bench and hoping that the person who takes them decides to send you the goods you were promised.

Well - then it's no different than any other payment system, then, is it? I mean, if you buy something from someone, then send cash in the mail, nothing says they'll send you your item. Or even if you pay with paypal, or a credit card, or a check.

That's what an escrow system is for (arguably, that's what Silk Road - and I think Mt Gox also provided - an escrow system, along with ratings - so that both buyer and seller were protected and that the exchange actually happened).

But yeah, if you transfer cryptomoney from your wallet to some other guy's wallet, there's nothing that says he'll send you what you were buying.

Unless/until the transfer can be bound into a contract you would be mad to trust that as a payment system.

Again, what "contract" really exists with current electronic payment systems? People get ripped off all the time in the same kinds of manner - and sometimes there isn't even any way to show accountability. Arguably with cryptocurrency, you can at least show who made what payments when (there is a ledger for all wallet transfers), which isn't easily the case with other systems (certainly not cash - and even with credit/debit, individuals can't see all transactions that have occurred to/from accounts - not without court orders and such - so in effect, cryptocurrencies are more open - which would be a lot nicer if they were in broad use, because then we could easily see who is donating to whom in political campaigns and such).

I believe it's technically possible using the same PPK mutual signing approaches proposed for secure exchange systems, but I don't see any sign of it happening yet.)

As I said - Silk Road was doing some kind of escrow-based system (where you had a wallet on the system, and transfers happened between wallets on the system itself once parties agreed - rather than the more direct personal-wallet-to-personal-wallet); I don't know if there were any PPK signings or whatnot happening, but I could see that being done.

Bitcoin realm seems rife for ripoffs too. So many cases of bitcoin thefts in the news recently.

Much of this has happened because of insecure wallets - it would be like leaving your physical wallet out on a park bench or something; also, people have had their wallets stolen, but not their keys (so it would be like leaving a very secure safe out in the open public - someone could steal it, but they may never be able to open it). This is what happened mainly on the Silk Road takedown by the Feds - they got a ton of wallets (stored as files on the servers) that had a ton of money (in Bitcoin) in them, but that money is essentially lost to the market because no one (not even Feds - not yet, at least) can open them and use them (and if they did manage to crack them in some manner - the Bitcoin community would likely see it).

At least, that is how I am understanding things. I am sure that some people left behind wallet files and also the keys to them (like taping the combination to a safe) - people can be very stupid about things like this.

Lastly - if it isn't apparent from any of the above - I am no expert on this whole cryptocurrency thing, but a lot of people don't seem to understand it at all. Which isn't surprising - most people have no real concept of how baseless our entire monetary system (worldwide) as a whole is. For some reason, people think these bits and bytes flying through banking computer systems have some kind of intrinsic value, when in reality, the only value they have is the value we humans assign to it, for whatever arbitrary reason. Indeed, it has always been this way (even with barter, of course). So if everything is being done with a computer - why not an alternative system? If enough people give it value - then it has value. What governments don't like about cryptocurrencies, I believe, isn't that issue - it's more the fact that such currencies are more open (you can see all of the transactions - nothing is hidden from anyone), plus they don't require any middle-men (just like cash) - but are just as easy to use and transfer (because the internet and computers) as credit or debit systems, and they don't require banks, and of course - because of all of this - in theory taxes and/or other fees can be easily avoided.
84  Community / Bar Sport / Re: Bitcoin miner - Is it possible? on: March 16, 2014, 01:43:15 pm
Yet there doesn't seem to be any better logic in their own system which, if I understand it from a quick read, just creates new bitcoins for validating existing bitcoins.

That isn't quite how it works.

Cryptocurrencies (because that is what we are talking about - bitcoin was just the first) are "mined" by a "computer" (CPU, GPU, FPGA, ASIC, etc) solving an equation. When that equation is solved (generating the "hash") - the equation gets harder to solve for the next "round". It takes a lot of processing power to create the solution; it is also possible for multiple machines to solve a round at the same time. There are checks and balances to make sure each "winner" gets the number of "coins" they have mined based on the solution and when it was submitted.

Sounds like getting a new $10 bill whenever you run the special validation pen across 1000 other $10 bills. Oh, wait, you have to swipe the pen across 100,000 smiley

No - that isn't it - mining takes resources, just like mining for a precious metal (which doesn't really have much intrinsic value either, when you think about it - most are just "pretty/shiny" things to oooh and ahhh over, though some have certain physical and chemical properties that make them valuable - but not to the level that they are valued at on the monetary market): you need equipment (the computing power), energy (to run those systems and to keep them cool), space (to house the systems), time, and maintenance.

Mining a cryptocurrency is much more akin to physical mining; at time goes on, it takes much, much more effort to get at the remaining "coins" than it did in the beginning of the mining - just like a physical mine "plays out" and takes more effort, time, and money to get at any remaining resources. This effort is purposefully built into the system being solved by the mining hardware.

In fact - the bitcoin cryptocurrency is already pretty much "played out" - at least to the point where an ordinary person can't hope to be able to mine anything on his or her own; they would easily end up spending more on their mining rigs than they would ever be able to hope to recover. It's even beginning to be doubtful that even large rigs (large liquid-cooled ASIC server systems) will be able to recover their costs. That of course doesn't stop people from trying at any level (just like people still pan for gold, or re-open supposedly "played out" mines).

Like regular money ($, £, €) bitcoins will only work as long as people trust them. I will stick to the Government validated stuff. Bad as they may be, at least they are elected from time to time.

What is your government's money backed by? Mine (USA) - is backed by a promise (oh - and lots of nuclear weapons, I suppose) - your's is probably not much different; that's why they are called "promissory notes". They are only so many pieces of sometimes pretty paper (or plastic for some currencies) that people have decided to collectively pretend have value (and let's be really honest - those pieces of paper mean nothing - I mean, if you have a bank account, use paypal, or a credit/debit card, etc - we're all just shuffling ephemeral bits in a computer system around, which makes all of this even more virtual, arbitrary, etc). be continued...
85  Community / Bar Sport / Re: Looking for United States published Electronics magazines on: March 15, 2014, 07:21:34 pm

...and the sister magazine - Servo (for all things DIY robotics related)...
86  Using Arduino / Motors, Mechanics, and Power / Re: L293D Motor Supply Voltage Drop [SOLVED] on: March 15, 2014, 07:12:31 pm
For some reason the L293x's seem more prevalent.

Mainly because they are cheap, well understood, and there are a ton of them out there - basically, the L293 (and the L298) are the "555" of motor drivers...
87  Using Arduino / General Electronics / Re: L298N is not working?? on: March 15, 2014, 07:08:59 pm
Purchase an H-bridge rated for that amount of current. I don't think the L298N would parallel well by adding more L298N chips.

No - but a single L298, in parallel mode (driving a single motor using both channels in parallel) should be able to supply up to 4 amps of current.

BUT - you will need one helluva hefty heatsink (and forced cooling - stick a fan on it!).

That said, there are better alternatives out there, but if all you have is an L298, and your needs don't exceed 4 amps, and you don't mind the inefficiency...
88  Using Arduino / Project Guidance / Re: Demonstration code for several things at the same time on: March 13, 2014, 10:18:25 pm
any ideas?

I don't know about the code, but you have a serial.print() line in the loop() function that I would like to encourage you to edit, or the moderators may do it for you.

/remember, this forum caters to many age groups and such; let's try to keep it somewhat in line with that...
89  Using Arduino / Project Guidance / Re: Plug and play - Glitch art webcam on: March 12, 2014, 11:47:56 am
I wonder id you could just strip the webcam wire ground and use interference to corrupt its frames instead.

I really doubt that would work; likely it would just cause the camera to disconnect from the driver, or the driver itself to crash - it wouldn't likely corrupt the image data itself. You couldn't even hack the camera, as most current web cameras have everything (cpu, encoder, usb, etc - including the camera ccd in many cases) on a single chip - that is, the image sensor is the entire package, with essentially a usb port to the outside world.
90  Using Arduino / Motors, Mechanics, and Power / Re: RC car servo with 6 wires ;;; on: March 10, 2014, 04:05:16 pm
Yeah the best will to replace it with an 3-wire servo. The thing that piss me of is that i took it from modeling shop and told them
"i want a car with a real servo" !!!!!! I paid 100euro for that sh1t.

Well - technically you did get a "real servo" - it's just that the control electronics weren't in the servo, but on the external control board...  smiley-razz

I guess next time specify a three wire hobby-grade replaceable servo, and either bring along or link to an example of the servo or something.

That's a good servo to use; there's only likely to be a couple of potential issues, since this is the steering servo:

1. You may need to custom mount the servo - hopefully both the original servo and this new one are the same size or close, and that the existing mounting points can be used. Otherwise, you will need to do some custom fitting to get it to work properly.

2. It's unclear whether the original servo (or the vehicle itself) had a "servo saver" shock absorber mounted; sometimes it's part of the steering mechanism, sometimes it's a special spring-loaded control horn that's mounted to the servo. If the car is small enough, or is meant to be driven on flat surfaces - then it may not have anything. The purpose of such a device is to act as a spring-loaded shock absorber so that if the front steering wheels are bumped by an external force, the servo control horn (or the gears in the servo) aren't damaged. You typically see such a device used on larger 4WD buggys/truggies and such.

The first item will be the real potential issue here; the second I wouldn't worry about too much (just something to be aware of).
Pages: 1 ... 4 5 [6] 7 8 ... 375