H-Bridge

[jsmoker]

I'm trying to use the bridge found on this webpage:

http://robotroom.com/HBridge.html

I've set it up just as shown but I also have an LED (and resistor) between the two inputs to show me that I am getting a signal.  I also have the input voltage common to both the micro controller and motor power supply (as the motor can only handle around 5 volts anyways.

Unfortunately when even I plug in the motor, the microcontroller stops running.  Any ideas what the problem might be?  Is there anyway to fix this without totally separating the power supplies?

-JSmoker

[jsmoker]

Oh, and I'm using a dsPIC30F6011A, the mosfet is the one shown TC4424 only a SOIC version.  The motors are small simple hobby motor that run with an input up to 6V (I'm gonna run them at around .4 amps) and I'm using a 1amp voltage regulator.

[Robotboy86]

Without looking at the links, and only hearing that the microcontroller hits off when you attach motors, my GUESS is that you're sharing a common power source, in which case I would STOP that and seperate them. 

Your MCU should only share one link with the h-bridge, and that is hitting on the mosfets on/off.  H-bridges can drain a TON of power..  So.

[jsmoker]

yeah, as I mentioned in my post, that was my assumption too, but was wondering if there was a way around that without separating the power sources, i.e. big cap or something.  And just out of curiousity...why is this happening.  If the voltage regulator is able to put out 1Amp, the motor takes .2 amps (.7 amps under high loads) and the controller takes .1 amp, that's under the 1 Amp supply.  Far under for no load.

[jsmoker]

P.S.  the reason I'm trying keep the power supply how it is, is 'cause the whole circuit is on a custom PCB, so it's a pain to try to attach leads to traces and disconnect traces. 

[Joe]

P.S.  the reason I'm trying keep the power supply how it is, is 'cause the whole circuit is on a custom PCB, so it's a pain to try to attach leads to traces and disconnect traces. 

A capacitor would be the way to go but I couldn't tell you what size. If you can take the pic out of that board you should try rebuilding as much of the circuit as you can in a solderless breadboard. Also are you sure about that load current? Have you measured the voltage accross the leads under full load to make sure it hasn't dropped? .7A sounds kind of low.

[jsmoker]

Does anyone know a higher amperage H-Bridge chip than the L298?  I need to be able to pull around 4-5Amps.

[Admin]

Does it need to be an IC?

http://www.dimensionengineering.com/
has several good motor drivers . . .
(links at top left)

[Admin]

(I just merged the above topics because they were very related)

I normally use a big cap for my stuff . . . Id make sure you arent shorting anything, because it sounds like you are having a lower power dropout. An H-bridge also requires two PNP and two NPN type mosfets, not four of one type.

This explains it decently:
http://www.societyofrobots.com/robotforum/index.php?topic=1625.0

In any case, as you've probably noticed by now, I wouldnt recommend building an H-bridge. You can buy drivers fairly cheaply now.

[jsmoker]

Don't mean to be picky, but anyone know of a dual motor driver for less than what's on

http://www.dimensionengineering.com/

I'm looking in the $5-$20 range.  you know, I wouldn't complain about $0-1.  Building one with mosfets is more of a pain, but they run $1-3 each.

Thanks

[Admin]

Put two L298's in parallel to double the current limit.

[Rebelgium]

I've had that problem to (MCU reset when you start your motors)
This is caused by the fact that all motors are inductive loads, and inductive loads ask high currents to start.

So if you start a motor, you will see a current peak. This causes the voltage to drop (even with a voltage regulator). Just for a very short time though, but long enough to be a bother for you MCU and other circuitry to.

The two solutions are:
*Place two capacitors on your logics power and GND. One electrolytic of about 100?F, and one ceramic of about 100nF
If this doesn't help, keep adding capacitors.

*Use a separate power source.
This is the easiest and most reliable solution.