Using LM18200 as an H-Bridge for AC Output: How to Configure and Drive a Coil

[anthonyjames9]

Hey guys,

I'm attempting to use this chip, http://www.national.com/pf/LM/LMD18200.html#Diagrams with a simple 555 circuit as the input to drive an AC signal out.

From my understanding I can have the 555 going into the PWM input, a relatively high current connection to Vs, this will then generate a high current (relatively) switched output at output 1 and 2?

I'm just not too sure how to configure it, the data sheet is confusing me.

The idea is I'm driving a coil, if i can have up to 3A being switched at 500kHz it'd be perfect, I'm just clueless.

Thanks in advance.

AJ

 

[Bobbywhy]

The data sheet of the LM/LMD 18200 is used for driving DC motors. Its output pulses are averaged (felt) by the motor. More negative-going pulses drives the motor one direction and vice-versa. When the positive and negative pulses are equal the motor senses ZERO drive signal and does not move. Nowhere in the data sheet is an AC output mentioned. You have proposed an AC output switched at 500kHZ. What does that mean...do you suggest turning the AC on and off at that rate? This does not make sense. Also, in the data sheet you will find a graph of "supply current vs frequency". The graph stops at 100 kHz, so it is unlikely that the IC would ever operate at 500kHz.

What coil do you plan to drive with 3 Amps? If you say what you overall objective is, it may be possible to select an IC perfect for your project.

Suggestion, read the data sheet section which lists "applications". It seems this is not the IC you need.

Just a guess: if you are thinking of making a power supply, consider Pulse Width Modulator (PWM) ICs.

 

[jim hardy]

hmmm looks to me like it might drive a coil
provided duty cycle is 50% so there's zero DC component.

it's difficult to get 50% out of a 555 so you might follow it with a divide-by-two to make it symmetric.

and as previous post observed - try it on bench to see if it's fast enough.

they claim it'll maybe do 500K by adding bootstrap capacitors

page 9

External 10 nF capacitors, connected from the outputs to
the bootstrap pins of each high-side switch provide typically
less than 100 ns rise times allowing switching frequencies up
to 500 kHz.

 

[anthonyjames9]

Thanks for your replies guys.

I'm trying to create a small wireless induction circuit, I've managed to use a signal generator to create the 500kHz signal, but obviously it's not capable of driving much current. Maybe AC was the wrong word, but more "Choppy DC". I was hoping the 555 would act as the PWM signal, 'telling' the chip to turn it's outputs on and off to give this choppy DC.

I'm not too sure how I'd do this with the 555?

it's difficult to get 50% out of a 555 so you might follow it with a divide-by-two to make it symmetric.

I've used some basic 555 calculators such as this one:
http://www.ohmslawcalculator.com/555_astable.php

Setting C to 0.001nF, R1 to 1kΩ and R2 to 1.5MΩ it seems to reduce the duty cycle but keep a large frequency. I'm assuming perhaps this isn't the best way of doing it though?

 

[Bobbywhy]

Now that you have said you are building a wireless induction circuit "driving a coil" others may post useful and helpful comments. Recently a wireless power transfer thread appeared here on Physics Forums:

https://www.physicsforums.com/showthread.php?t=556235&highlight=wireless

I suggest you read all sixteen posts.

Yes, the (venerable) 555 will supply the PWM signal to your LMD8200. See the data sheet http://www.national.com/pf/LM/LMD18200.html#Diagrams (be sure to download the pdf data sheet) for a description of types of PWM signals for controlling the 8200. If no joy is found there, there is a ton of 555 application notes to help with configuring the 555 using a Google search.

The LMD18200 is a 3A H-Bridge designed for motion control applications such as driving DC and stepper motors. The time average of the output square wave pulses results in an average direct current delivered to the load (coil) and is proportional to the pulse width of the PWM control signal.

IMHO your choice for the power driver IC seems to be not the best one for this application because, as I read the data sheet, its output is “square waves” (not “choppy DC”). When an inductive load like a coil is driven normally a sine wave is used. An oscillator (Colpitts.or relaxation, or crystal controlled, or ect.) generates a sine wave which is amplified, and that drives the coil. I reserve the right to be wrong about this.

 

[anthonyjames9]

Thanks for your reply again. The post you linked to is actually one I created a while back when I was looking for initial help. Through discussion with various people it was deemed that the Colpitts (or similar) weren't stable enough and a much more reliable method for this purpose would be a 555 and a driver. (Similar to that used in the commercial powermat system).

For my purpose I also can't go about about 1MHz (as discussed in the linked post) due to RF problems, which is why around 500kHz has been chosen. Ideally, from research it would seem 8MHz is a nice frequency to play with, but the problem I then encounter is things such as the breadboard won't work properly, and the cheap diodes I have lots of won't work.

Thanks once again.

 

[jim hardy]

that circuit on h-bridge datasheet is using 555 as a pulse width modulator to establish the DC output level.

if you just hand it 50% to pin 5, choose a direction and i THINK you'll have what you describe .
i never used that device so that's just how i read datasheet..

try this with 555 - tie trig and reset together to your timing capacitor, connect timing resistor to 555's output pin and see if you don't come real close to 50%.

i hope you post results of your experiments. i'd like to have a small induction furnace too, and saved some salvaged drive transistors from an industrial one.