![](templates/subSilver/images/CCSLogo.jpg) |
![CCS C Software and Maintenance Offers](templates/subSilver/images/forumAd6.jpg) |
View previous topic :: View next topic |
Author |
Message |
crystal_lattice
Joined: 13 Jun 2006 Posts: 164
|
Half Bridge configured as a Push-Pull inverter driver |
Posted: Sun Jun 22, 2008 7:57 am |
|
|
Hi there, I have come up with a quick solution for a push-pull inverter/half-bridge controlling two power MOSFET's.
The operating frequency is around 25Khz - This will be a fixed frequency, but needs to be externally tweaked to get the circuit to resonate where after it will be left alone. I might have to incorporate the dead band function to prevent any shorts as the load is somewhat inductive, so it would be nice if this could also be adjusted. I have i 18F4620 at hand and was thinking of using this.
I've never used this functionality of the pic and was wondering if anyone could point me to the right appnotes/ examples. I'm currently searching the net/ microchip site but there are hundreds of notes to read and not all are applicable and this solution is very urgent so any help would greatly be appreciated!!!
This is for an ultrasonic bath inverter and the existing circuit uses thyristors which is obsolete (unit older than 10 years) and new ones i got were not fast enough and am struggling to get the frequency up to get the transducers to resonate without the conduction halves overlapping. Hence the move to MOSFET's.
There are 7 generators of which 4 is currently down and i suspect they won't last long as the broken 4 broke within days of each other.
Any help would be greatly appreciated!!! |
|
![](templates/subSilver/images/spacer.gif) |
PCM programmer
Joined: 06 Sep 2003 Posts: 21708
|
|
Posted: Sun Jun 22, 2008 1:57 pm |
|
|
I don't have an 18F4620. I do have a 18F4550, and I think the ECCP
code for Half-Bridge mode will probably also work for the 18F4620.
You should see output on the P1A and P1B pins. Refer to the pinout
diagram in the PIC data sheet to find the pin numbers for those pins.
Quote: | #include <18F4550.h>
#fuses XT, NOWDT, PUT, BROWNOUT, NOLVP
#use delay(clock=4000000)
//=====================================
void main()
{
setup_ccp1(CCP_PWM_HALF_BRIDGE | CCP_PWM_H_H);
setup_timer_2(T2_DIV_BY_1, 255, 1);
set_pwm1_duty(64); // 25% duty cycle
while(1);
} |
|
|
![](templates/subSilver/images/spacer.gif) |
crystal_lattice
Joined: 13 Jun 2006 Posts: 164
|
|
Posted: Sun Jun 22, 2008 11:13 pm |
|
|
Thanks PCM, i will give it a try, in the mean time i've just used a software solution by setting pins high/low with delays in between. ![Smile](images/smiles/icon_smile.gif) |
|
![](templates/subSilver/images/spacer.gif) |
Ttelmah Guest
|
|
Posted: Mon Jun 23, 2008 4:07 am |
|
|
And on the 'tweaking', you will need to change the timing value put into the timer2 setup, and make the duty value always be 1/4 this. So, something like:
Code: |
int16 divisor;
divisor=400; //set this using your external input
if (divisor>512) {
//error here
}
if (divisor>256) {
divisor/=2;
divisor--;
setup_timer_2(T2_DIV_BY_2, divisor, 1);
}
else {
divisor--;
setup_timer_2(T2_DIV_BY_1, divisor, 1);
}
divisor*=2; //since divisor is an int16, the duty will want to be
//*2, rather than /2
set_pwm1_duty(divisor); // 25% duty cycle
|
Now, the steps for the frequency will be:
divisor frequency
397 25252
398 25125
399 25125
400 25000
401 25000
402 24875
403 24875
404 24752
etc.
If you need finer tweaks, you will have to think again.
Note that you can slightly detune those transducers, by altering the loading across them, reducing their effective 'Q'. This is commonly done now, where units have multiple transducers.
You could even consider detecting the voltage being produced across the transducers, and adjusting the frequency automatically to find the best resonance.
Best Wishes |
|
![](templates/subSilver/images/spacer.gif) |
crystal_lattice
Joined: 13 Jun 2006 Posts: 164
|
|
Posted: Mon Jun 23, 2008 9:05 am |
|
|
Thanks Ttelmah, do you have any experience with these transducers? Not sure how "by altering the loading across them" can be done, are we talking about the same transducers used on ultrasonic cleaning bathes?? They smaller ones usually have a allen bolt in the middle, but think that is only for mounting purposes.
I've contacted the original system installers and they don't really have anymore tech info on the system and say they are using new tech in the new systems eg. MOSFET/IGBT drives. But now my next problem comes in with finding MOSFETS with a >350V rating @ about 110A.... any suggestions for these?? |
|
![](templates/subSilver/images/spacer.gif) |
Ttelmah Guest
|
|
Posted: Mon Jun 23, 2008 2:43 pm |
|
|
Just as with any resonant circuit, you can 'pull' the resonance by changing the load. The amount of adjustment that can be done electrically is small, but you can move the peak a little, with external capacitance/inductors. You can move it far more, by altering the mechanical load as well...
If the company concerned, is like the one I know, the old design was poor, and the new design is worse... Design changes brought in by accountants, who don't understand that while you can cut costs by changing to a cheaper low temperature rating pump, the effect on reliability at the temperature these run, losses what was saved, in service calls.
Remember that you don't need 110A continuous, only a pulse rating. Look at the MicroSemi FredFet range. I have used their APT97H50J, in a similar application in the past.
Best Wishes |
|
![](templates/subSilver/images/spacer.gif) |
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
Powered by phpBB © 2001, 2005 phpBB Group
|