Author Options:

Robot troubles. Answered

I am trying to build a mini-sumo edgebot, as featured in Junkbots, Bugbots and Bots on Wheels. Below is a quick scan of the schematic I'm following. It is supposed to drive forwards until the edge sensor closes, at which point it will reverse and turn before carrying on.

Trouble is, it doesn't work.

The symptoms:

I have breadboarded the circuit, and when I connect the battery, it runs. But when I make the connection to simulate the edge sensor, both motors kick into reverse, but only for a fraction of a second, and then never again - further touches of the edge sensor elicit no reaction at all.

When I disconnect the battery, and trip the sensor, the LEDs light and the relays click, showing that the capacitor is still charged.

The only things in the breadboard not to "spec" are the motors - I haven't bought my intended motors yet, so I am trialling with a pair of un-geared 3V motors (these are the motors I intend to buy: http://www.ebay.co.uk item number 5855781458 )

How can I fix it so that it will react to the edge more than once in a combat?


wat type of relay u used and does this use a controller or does it runs by itself

(Woah, flashback!)

They are 5V DPDT relays, and there is no controller - this is a BEAM project.

. Oh. I just noticed that, according to Figure 12-3, the relay on the right is not getting full voltage to the coil - it's dropped through the LEDs. Probably doesn't affect anything, but I'd disconnect the right leg of the LED string, jumper from the left coil lead of the left relay to the right coil lead of the right relay and attach the LED leg to the right coil lead of the left relay. . And why two relays? One will do the job, since the motors aren't independent.

The voltage-drop is supped to make the robot turn in reverse, so that it then drives away from the edge next time it drives forwards.

The only way to do that would be if the LEDs were in the power to the motor, not the coil. As shown, both motors get full voltage.

But one coil switches off before the other, so the last part of the reverse only has one motor running.

. Hmmmmmmmmm. I can't imagine it making that much difference in pickup or dropout time, but I'm no engineer. Makes sense if it works and explains why two relays. . If I were gonna do something like that, I'd put the voltage drop in line with one motor's power during reverse, but, once again, I'm no engineer. . I'd also wire up the relay contacts different (power to C), but that's just me and would have no effect on operation.

i think NachoMahma is right. you only need one relay, because you are only using one switch.  Btw, the also need to be connected in the middle (where they both touch, right above the negative of the coil.) also, if you want the revers to last longer, you can use an NPN transistor, with a cap and resistor to control how long the base of the transistor if getting power, aka relay is engaged.

Are you reading my suggestions? I may be an electronics noob, but I'm a systems troubleshooting pro, and I can tell you that simplifying your circuit WILL help you narrow down what your problem is.

OK, then, TroubleShooterPro, exactly how should I simplify my circuit?

What should I remove, and how do I compensate for its removal?

Basically the same as NachoMama has suggested. In fact, if you'll scroll down the page you'll see a comment I left a couple weeks ago to that effect.

. I would suggest starting with battery, power sw, edge sw, and only one relay - no LEDs, no motors, and no cap. When you get that working, add the 2nd relay; then the cap; and finally the LEDs. . Royalestel may have (probably has) a better way to do it.

Please post whether it works after you get the right motors! I have the same book, and another geeky friend who knows slightly less about electronics and I are planning to build a pair of sumobots and have a deathmatch. We just need to find him gear motors and both of us relays.

Will do, they're ordered, but I won't be able to do anything if they arrive after Thursday, as I'm going away Saturday morning. (I got my relays from ebay as well - I just searched for DPDT)

Do you know of a support website for the book, or, more specifically, if there are any bugs in the circuits? I got it and the Absolute Beginner's Guide to Building Robots, and now own the former, but I want to make sure I am actually doing it right before I start building out of it.

. If I'm reading your pic right (good chance I'm not; I'd have to sit down and sketch out a schematic to be sure and I'm too lazy to do that), the cap is for time-delay-off operation of the relays and may be your problem. Try operating the circuit without the LEDs (or install a current limiting resistor); they may be draining the cap too fast.

The cap isn't draining at all - the circuit will reverse once, then not again. After I switch off, if I close the "sensor" switch, the LEDs light and the relays click, showing there is still a fair bit of charge in the cap.

. Let's see if I understand what's going on. The power switch closes, which starts the motors (LEDs off). When the EDS closes, the relays are energized (LEDs on) and the cap is charged. The relays reverse polarity to the motors, causing the EDS to open, which allows the cap to hold the relay for a time period. . Is the relay de-energizing? That is the only thing I can see that would cause your problem. After an encounter, short the cap and see if you can hear the relay de-energize. The cap should not be charged when in forward motion - only when the relays are energized.

> brain ache < I'm going to buy some fresh batteries for my camera, then I can show you what I've breadboarded. I'm pretty sure I'm doing something wrong at a really dopey, kick-myself-in-the-pants-when-I-see-it sort of level.

. I joined the Really Dopey, Kick-Myself-In-The-Pants-When-I-See-It Club a long time ago.

. Hmmmmmmm. I'm gonna have to make time to draw this out as a schematic. . Let's make sure I understand what I saw. Powerup is OK. First reverse/forward cycle works OK. Subsequent LS closures have no effect. After power is turned off, there is enough charge on the cap to energize the relays. . . What happens if the LS stays closed longer? . What is the ticka-ticka-ticka sound after you turn the power off (and the motor sspins down) and before you energize the relay? . The only time I can see the LEDs light up is when you energize the relay after power down. Is that right? . . Try the circuit without the motor or LEDs. I don't think the motor is having any effect, but the LEDs may be. . Try the circuit without the cap. If I understand what's going on, the cap should hold the relay in for a short time after the LS opens and I don't see that happening. My first guess is that the cap is wired wrong.

The constant ticking is the camera's focus hunting (it's a stills-with-video-capability camera). Thanks for the other suggestions, but I probably won't have time to try them until after next weekend (we're away on holiday early Saturday).

. Awwwwww. You're gonna make us wait? How gauche. LOL . Have fun on holiday.

Ah, so you hadn't fixed it yet ... when troubleshooting anything, you swap suspected bad with known good. In your case, I would simplify the circuit as much as possible first, then start swapping out parts, and adding back in parts.

This looks like a similar diagram to Mousey the Junk Bot. I had this same problem- the motors you are currently using to test your problem do not have enough voltage for the rest of the circuit. Your new motors are for 3V-6V, which are good because you want motors that are closer to 4.5V.

So that's all it is? Use the right motors and everything balances/works? Cool, thanks.

You can't see in the scan - the capacitor is 4700uF @ 16V

What purpose does the capacitor serve? (spare me the effort of trying to work it out) L

It provides the current for a few moments' reverse.

You're not running the reverse off the battery?

I happen to have a motor mounted on a board with a logical controller. It came out of a stereo and drives the volume knob via remote contol. If you'd like it I can post the whole thing to you.


>doh< The capacitor provides the juice to make the relays switch the polarity of the supply to the motors for a moment. The LEDs use a fraction of the current to one relay so that it reverses in a turn.

Oh wait, you solved it. nevermind . . .

Oh, and "yes please". I haven't a clue what I'd use it for, yet. I've PMed my address.

Well, I do know a bit about troubleshooting. I'd recommend making the circuit simpler (leave out LEDs, one motor, and a relay, add normal resistors as needed to keep voltage at proper level) and testing that. If that checks out, swap out relay and motor with the other. If that works, add in the LEDs, then test. Then complete the circuit and test one more time. If a part is flaky this should help you identify that. Otherwise, sometimes merely reseating everything will solve the problem. In retrospect, I'd try that first, actually. Hope this helps.

> bump < This is still doing my head in.