Today I received the first of a couple batches of PCBs from OSH Park. This set is a simple MOSFET w/ driver. The MOSFET can be logic level driven and all my testing was done at 5VDC however the driver had a wide voltage range so running this at 9V or 12V should present no issues (16V max! VGS). I do have a few minor adjustments I want to add (silk screen additions and adjustments). I built this board as an electric scooter motor drive and I tested it with the motor I was interested in using and thankfully all testing worked out well including loading my power supply at max load for over an hour straight 🙂 This board can be used for an assortment of load control applications.
The build was easy enough, next board I think I’ll just skip the chipquik and stick to straight soldering iron… and I think I also “figured out” the trick to solder the heat sink without making too much of a mess.
I tested the load for 60 minutes at 66% duty cycle and 1.5KHz PWM with total load at 140watts (45Volts 3.1 Amps).. Â the heat sink temperature on the MOSFET raised from ambient (78 deg F) to 90.7 deg F (32.3 deg C) steady for an hour. Â The MOSFET is rated for 55V (max) at 30 amps but the board isn’t designed to handle that load. I used 50mil power traces which is roughly 3 amp handling according to the google machine; 5A if you don’t mind a 40 deg C rise (though there is a fan) so I’ll call this a 140W MOSFET board (47×3). I personally feel I could squeeze more current out of this in short bursts but I would need to do more testing.. and I can’t promise anything along those lines. The PWM signal was generated and directly fed from a PIC 18F16K22 microcontroller running @ 5V. You’re NOT going to be able to drive this at 3.3V.
I will likely stick a 75VDSS MOSFET in the next revision…
.. I’m considering selling these on Tindie. If you’re interested in getting on of these “BETA” boards at cost let me know before they’re gone.
Took a while to clear off the bench tonight. I got a little happy with the ORDER button last week through weekend. I’ve filled two more parts storage containers and luckily one of my new packages with another parts storage container from Amazon; I couldn’t find any locally, I hope they’re still making them.
The last thing I got to was this “Gauge Stepper Breakout” I got off The Rengineer (The Renaissance Engineer)’s Tindie store. Adam took this very nice stepper motor, put a gauge needle on it (which I believe he 3D printed, or at least it looks like it) and some nice diodes for voltage protection. What’s great about this board is you can drive it right from your microcontroller. I’m sure I don’t need to tell you how weird that is.. I worry about big LEDs.. but here this stepper was happy as a clam being powered by my PIC. I still might consider at least buffering this if I was to place it in a permanent circuit.
Adam’s provides links to some Arduino libraries but he was saved me a ton of time and just happened to have some PIC sample code! Not before I had about 75% of a ECCP program completed. (I’ll link some of that code if you’re looking for an example of Enhanced PWM output). Once I had Adam’s code I ported it to the TAUTIC 18F26K22 dev board because I already had one on the breadboard. That was mostly changes in the TMR0 (timer 0) code. The only thing of note is check out my IO comments for wiring. You have to cook up 5V/GND to this board then I drove RC3 –> A1 RC2 — A2, RC1 –> B1 and RC0 –> B2. When I had it flipped around the stepper moved in the opposite direction of how I wanted.
A little video of the code in action:
This code is a little sloppy but I just testing this out and it did the trick:
/*
* File: main.c
* Author: Charles M Douvier Contact at: http://iradan.com
* Core Driver Code by Adam F. of http://www.therengineer.com/
*
* Created on April 4th, 2014
*
* Target Device:
* TAUTIC PIC 18F26K22 Dev Board
*
* Project:
*
*
* Version:
* 1.0
*
*/
#ifndef _XTAL_FREQ
#define _XTAL_FREQ 4000000 //4Mhz FRC internal osc
#define __delay_us(x) _delay((unsigned long)((x)*(_XTAL_FREQ/4000000.0)))
#define __delay_ms(x) _delay((unsigned long)((x)*(_XTAL_FREQ/4000.0)))
#endif
#include
#include
#include
#include
//config bits
#pragma config FOSC=INTIO67, WDTEN=OFF, PWRTEN=OFF, CP0=OFF, CP1=OFF, BOREN=ON
#pragma config STVREN=ON, LVP=OFF, HFOFST=OFF, IESO=OFF, FCMEN=OFF
//WRT=OFF, FOSC=INTOSC, MCLRE=ON
#define _XTAL_FREQ 4000000 //defined for delay
#define MIN_STEPS_LEFT 23
#define MAX_ACCEL_INDEX 6
#define MAX_STEP 945 /* motor can move 945 steps from stop to stop*/
int an8_value, an9_value; //value for a/d
char buf[10]; //buff for iota
long int fvar; //long for format math
long int tens; //left of decm
long int decm; //decimal places
int tempi; //to add leadign zeros..
int vtxdata; //volts int for TX
int itxdata;
unsigned short defaultAccelTable[][2] =
{
{ 1750, 3},
{ 1149, 3},
{ 926, 3},
{ 794, 3},
{ 709, 3},
{ 666, 4},
{ /*629*/450, 4},
};
unsigned int currentStep;
unsigned char currentState;
unsigned char stateMap[] = {0x09, 0x01, 0x07, 0x06, 0x0E, 0x08};
unsigned char serialBuffer[10];
unsigned char serialByteCount;
static const unsigned char stateCount = 6;
volatile unsigned int uart_data; // use 'volatile' qualifer as this is changed in ISR
/*
*
*/
void interrupt ISR() {
if (PIR1bits.RCIF) // see if interrupt caused by incoming data
{
uart_data = RCREG; // read the incoming data
PIR1bits.RCIF = 0; // clear interrupt flag
}
}
void init_io(void) {
TRISAbits.TRISA0 = 0; // output
TRISAbits.TRISA1 = 0; // output
TRISAbits.TRISA2 = 0; // output
TRISAbits.TRISA3 = 0; // output
TRISAbits.TRISA4 = 0; // output
TRISAbits.TRISA5 = 0; // output
TRISAbits.TRISA6 = 0; // output
TRISAbits.TRISA7 = 0; // output
ANSELA = 0x00; // all port A pins are digital I/O
LATAbits.LATA0 = 0;
PORTAbits.RA0 = 0;
TRISBbits.TRISB1 = 0; //P1C output
TRISBbits.TRISB2 = 0; // P1B output
TRISBbits.TRISB3 = 1; // AN9 speed control 0-5V
TRISBbits.TRISB4 = 0; // P1D output
TRISBbits.TRISB5 = 1; // RB5 = nc
TRISBbits.TRISB6 = 0; // RB6 = nc
TRISBbits.TRISB7 = 0; // RB7 = nc
ANSELB = 0b00001000; //RB3, AN9
TRISCbits.TRISC0 = 0; // output to B2 .. reversed to stoke the right direction
TRISCbits.TRISC1 = 0; // output to B1
TRISCbits.TRISC2 = 0; // output to A2
TRISCbits.TRISC3 = 0; // output to A1
TRISCbits.TRISC4 = 0; // output
TRISCbits.TRISC5 = 0; // output
TRISCbits.TRISC6 = 0; // output
TRISCbits.TRISC7 = 0; // output
ANSELC = 0x00; // all port B pins are digital I/O
}
void uart_xmit(unsigned int mydata_byte) {
while(!TXSTA1bits.TRMT); // make sure buffer full bit is high before transmitting
TXREG = mydata_byte; // transmit data
}
void serial_init(void)
{
//9600 8N1
// calculate values of SPBRGL and SPBRGH based on the desired baud rate
//
// For 8 bit Async mode with BRGH=0: Desired Baud rate = Fosc/64([SPBRGH:SPBRGL]+1)
// For 8 bit Async mode with BRGH=1: Desired Baud rate = Fosc/16([SPBRGH:SPBRGL]+1)
TXSTA1bits.BRGH=1; // select low speed Baud Rate (see baud rate calcs below)
TXSTA1bits.TX9=0; // select 8 data bits
TXSTA1bits.TXEN = 1; // enable transmit
RCSTA1bits.SPEN=1; // serial port is enabled
RCSTA1bits.RX9=0; // select 8 data bits
RCSTA1bits.CREN=1; // receive enabled
SPBRG1=25; // here is calculated value of SPBRGH and SPBRGL
SPBRGH1=0;
PIR1bits.RCIF=0; // make sure receive interrupt flag is clear
PIE1bits.RCIE=1; // enable UART Receive interrupt
INTCONbits.PEIE = 1; // Enable peripheral interrupt
INTCONbits.GIE = 1; // enable global interrupt
__delay_ms(50); // give time for voltage levels on board to settle
uart_xmit('R'); // transmit some data
}
// All this motor and timer code is from Adam with very minor changes to fit the processor
void t0Delay(unsigned int usec)
{
unsigned int t0ticks; //16 microsecond timer0 ticks
unsigned char t0Preload;
if(usec<16)
{
t0ticks=1;
}
else
{
t0ticks = usec/16;
}
t0Preload = 0xFF - t0ticks;
INTCONbits.TMR0IF=0; //clear the flag
TMR0 = t0Preload;
while(INTCONbits.TMR0IF==0)
{
;
}
}
void zeroMotor()
{
unsigned int i;
for (i=0; i < MAX_STEP; i++)
{
LATC=stateMap[currentState];
currentState = (currentState + 5) % stateCount;
t0Delay(1900); //2200 in datasheet
}
//now the motor is zeroed, reset our state variables.
currentStep = 0;
currentState = 0;
LATC=0; //turn off coils
}
void moveMotor(unsigned int targetStep)
{
unsigned int dir;
unsigned int curDelay;
unsigned char speedIndex=0;
unsigned char stepsAtThisSpeed=0;
unsigned int stepsLeft;
if(currentStep<targetStep) { dir = 1; stepsLeft = targetStep-currentStep; } else { dir = -1; stepsLeft = currentStep - targetStep; } while(stepsLeft>0)
{
if(stepsLeft<=MIN_STEPS_LEFT) { //decellerating if(stepsAtThisSpeed==0) { if(speedIndex>0)
speedIndex--;
curDelay=defaultAccelTable[speedIndex][0];
stepsAtThisSpeed=defaultAccelTable[speedIndex][1];
}
}
else
{
//accellerating or steady state
if(stepsAtThisSpeed==0)
{
if(speedIndex<MAX_ACCEL_INDEX) { speedIndex++; curDelay=defaultAccelTable[speedIndex][0]; stepsAtThisSpeed=defaultAccelTable[speedIndex][1]; } //else we're at steady state - do nothing. } } //write step LATC=stateMap[currentState]; if(dir==1) { currentState = (currentState + 1) % stateCount; } else { currentState = (currentState + 5) % stateCount; } t0Delay(curDelay); if(stepsAtThisSpeed>0)
{
stepsAtThisSpeed--;
}
stepsLeft--;
currentStep+=dir;
}
}
int main(void) {
init_io();
serial_init();
// set up oscillator control register, using internal OSC at 4MHz.
OSCCONbits.IRCF = 0x05; //set OSCCON IRCF bits to select OSC frequency 4MHz
OSCCONbits.SCS = 0x02; //set the SCS bits to select internal oscillator block
ADCON0 = 0b00100101; //select AN9 and enable
ADCON1 = 0b00000000; //speed Vref=AVdd, VssRef=AVss
ADCON2 = 0b00111011; //ledft justified, 20RAD, FRC
INTCONbits.TMR0IE = 0;
TMR0=0;
T0CONbits.T08BIT = 1;
T0CONbits.T0CS = 0;
T0CONbits.PSA = 0;
T0CONbits.T0PS = 0x04;
INTCONbits.TMR0IF = 0;
T0CONbits.TMR0ON = 1;
__delay_us(5);
currentStep = 0;
currentState = 0;
zeroMotor();
__delay_ms(149); //this could be less messy
__delay_ms(149);
__delay_ms(149);
__delay_ms(149);
__delay_ms(149);
__delay_ms(149);
moveMotor(20);
__delay_ms(149);
__delay_ms(149);
__delay_ms(149);
__delay_ms(149);
__delay_ms(149);
__delay_ms(149);
moveMotor(940);
moveMotor(5);
while (1) {
//PORTAbits.RA0 = 1; //heart beat
//__delay_ms(50);
//PORTAbits.RA0 = 0;
//__delay_ms(50);
ADCON0 = 0b00100101; //select AN9 and enable
__delay_us(5);
GO = 1;
while (GO) continue; //wait for conversion
an9_value = ADRESH; //AN9 value
fvar = an9_value;
fvar = fvar * 10749; //calibration.. change to meet your needs
fvar = fvar / 256;
tens = fvar / 100;
//tens = tens % 10;
decm = fvar % 100;
vtxdata = fvar / 20;
uart_xmit(vtxdata); // -->RS232
moveMotor(vtxdata);
//moveMotor(5); //from sample code
}
return (EXIT_SUCCESS);
}
Some ECCP ( Enchanced PWM ) code written for the PIC 18F26K22 code I wrote following the screen shot of the output:
My wife is kicking all non-food items out of the Frig… time to find a new place for the Chip Quik!
Chip Quik is a brand of solder paste used often in hobbyist reflow; this is the product I used to solder SMD parts to printed circuit boards. The product can work six months (or longer in reality) if you keep it cooled.
What to do? Well I started by looking into small frigs.. too much power and they cost too much. Next was the obvious choice: A peltier cooling device and a box. It took me a while to find *anything* that was small and seemed like it would work on my bench as well as be able to be insulated. I went to Lowes and picked up a roll of foil bubble-wrap looking insulation, a trip to a hobby store scored me this wooden box. A little hot glue and a start with two surplus heat sinks (seen rejected and removed to the right in the photo)
The inner heat sink had a fan, exterior heat sink didn’t but didn’t get uncomfortably warm. It didn’t work as I had hoped though; I couldn’t pull off more than about 10 deg F delta T from the ambient temp in the room…. not going to be good enough. My workshop hovers around 78-85 deg F due to heat load of the items… Chip Quik isn’t going to last more than the six months rating in those temps.
The search for a solution continued, I picked up a little peltier kit off eBay to see if I could do better; easy enough I guess. Put it all together and now I’m getting about 25-30 deg F delta T from ambient. My little box settles out fairly quickly at about 58 deg F (15 deg C) which is good enough. Next I’ll do some clean up and power this with a PC power supply. I’ll use a spare PC power supply because I don’t want to use up a bench supply and like it or not these devices to draw some current. It’ll take about 25%-50% of the power of the smallest frig I could find to keep this thing cooling .. though the frig would be great for some beer pop, unfortunately I don’t have the room.
Alternative ideas to this problem are welcome.
I have a ton of projects right now and zero parts… wait wait wait.. I hate waiting for parts 🙂 … come on slow boat from China! I also have a board in OSHPARK I’m super excited about, but no sneak peaks…. it’ll be here soon enough. Wait wait wait wait…