Let's Make Robots!

theory vs. soldering

Well...my robot is far from being able to be presented.

In fact I try to build/solder some parts on my own, which seems to be more challenging than expected. Actually I´m soldering a h bridge with its controller based on oddbots h bridge - without the current sensing part - and I just hat to face, that it doesnt work. I guess its not due to oddbots design, but to my soldering skills and or adapting to a simple board design, since I solder though-hole.

I hope I can figure out the problem soon...

This is his version: http://letsmakerobots.com/node/18633


well...and this is my design, which seems to have some problems...


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

At the end of the day, H bridge designs can be hard to debug without a dual trace oscilloscope. If you are testing on a breadboard then only test at low PWM frequencies because the added capacitance of the breadboard can affect the circuit.

I got everything working about the h-bridge controller and it´s pretty impressive, now that I understand the way it works as I read about the datasheets and items used in general, as you included that many features and safety in an elegant way. I learned quite a lot and hopefully I will be able to make my robot drive soon (lack of time) and take care of other circuits.

When searching for bugs, I had to hope that the items being soldered were ok as I don´t have a osziloscope, but am currently limited to a simple multimeter.

Thus i detected the poor voltage behavior across the 2,5V lines, which would only increase to 3,5V.

About testing PWM it is really a pain with insufficent measuring tools (who wonders). Thus I disabled PWM.

So thanks. Great support and making one deal with more circuits. It´s quite more fun this way, although there´s so much more to learn...but thats the fun.


Check your work for continuity where it should and shouldn't be.

Have you considered breadboarding the circuit first?

Trhanks for your reply.

Well, as oddbots schematic was well organized and clear I actually didnt use a breadboard before. I think I will give it a try with the breadboard soon. However I rechecked a couple of times, whether something went wrong with soldering and couldnt find any short-circuit or whatsoever. 

The first p-channel mosfet connected to the LM339 situated in the bottom left corner becomes extremely hot as soon as I try to power the circuit like:

(Arduino UNO)

digitalWrite(pinA, HIGH)

digitalWrite(pinB, LOW)

PinA and B are on the left-hand side.

However the mosfets are still ok, as they work with another h-bridge controller circuit, which I always use to test as soon as this happens.

I replaced oddbots BC847 transistors with BC547 ones, which shouldnt be a problem referring to the specifications. 

damn...looks like i found the bug.

Dealing and learning about all the items (voltage divider, LM339) etc., I got fed up with everything and used a volt-meter to measure every trace.

Guess I built something like a divider regarding the arrangement of the first 100ohm resistors falsely connected to the capacitor and the "following 1.2k ohm resistor.

I should be more patient.

Guess I should´ve spend more time on a breadboard and the following layout.