7219 driver (was Transistor calculation)
Archiver
Posts: 46,084
Ok, maybe I need to explain my whole project alittle more (neglected the
gory deals to simplify things). What I am trying to do is design a driver
board for the MAX7219. I am trying to drive a 8" display that consists of
12 LEDs per segment (3 series groups of 4 parallel arranged LEDs), 7
segment, 4 digits. Currently I am using op-amps for the source (segment)
drivers but am still 'blowing' 7219's. What I would like to do is design a
driver that could power ANY format of display (up to a limit, say 500mA,
24v). My 99th idea was to use a ULN2803 for the source (segment) drivers
and use TIP transitors for the sink (digit) drivers. I hope this explains
my need for the PNP for the sink drivers. For those not familiar with the
7219, it multiplexes between sourcing the segments and sinking the digits
to display the correct output to the LEDs. I was hoping for a simple
solution but I see I need to do my homework with transistors and maybe
FETs. I would love to opto-isolate the 7219 completely from the LEDs but
that may have to be revision 101. The new datasheet from MAXIM shows a new
driver circuit (using FETs) for higher power buy it inverts the common
cathode output of the 7219 to common anode (I would rather not rewire my
336 LEDs). I am slowly learning why simple circiuts need additional caps,
transistor, diodes, etc. Thanks for the help and thanks Al for the
links(especialy on the FETs). Back to the books. Any insightful
direction is appreciated. Dan [noparse]:)[/noparse]
gory deals to simplify things). What I am trying to do is design a driver
board for the MAX7219. I am trying to drive a 8" display that consists of
12 LEDs per segment (3 series groups of 4 parallel arranged LEDs), 7
segment, 4 digits. Currently I am using op-amps for the source (segment)
drivers but am still 'blowing' 7219's. What I would like to do is design a
driver that could power ANY format of display (up to a limit, say 500mA,
24v). My 99th idea was to use a ULN2803 for the source (segment) drivers
and use TIP transitors for the sink (digit) drivers. I hope this explains
my need for the PNP for the sink drivers. For those not familiar with the
7219, it multiplexes between sourcing the segments and sinking the digits
to display the correct output to the LEDs. I was hoping for a simple
solution but I see I need to do my homework with transistors and maybe
FETs. I would love to opto-isolate the 7219 completely from the LEDs but
that may have to be revision 101. The new datasheet from MAXIM shows a new
driver circuit (using FETs) for higher power buy it inverts the common
cathode output of the 7219 to common anode (I would rather not rewire my
336 LEDs). I am slowly learning why simple circiuts need additional caps,
transistor, diodes, etc. Thanks for the help and thanks Al for the
links(especialy on the FETs). Back to the books. Any insightful
direction is appreciated. Dan [noparse]:)[/noparse]
Comments
>Ok, maybe I need to explain my whole project alittle more (neglected the
>gory deals to simplify things). What I am trying to do is design a driver
>board for the MAX7219. I am trying to drive a 8" display that consists of
>12 LEDs per segment (3 series groups of 4 parallel arranged LEDs), 7
>segment, 4 digits.
Can I suggest that you drop the MAX part entirely? Take a look at the
TPIC6595 from TI. You will need one of these per digit: 4 total for a 4
digit display.
The '6595 is functionally similar to the 74hc595 but contains high current
MOSFET output drivers that will switch up to 60V at a maximum of 250
mA. Its what I now use in all of our large 7-segment displays.
It still takes 3 lines from the Stamp: sclk, data, rclk. You connect sclk
and rclk to all of the TPIC6595 chips and data to the input of the last
'6595. The data out from that '6595 feeds the data in of the 2nd last
'6595; the data out from that one feeds the data in of the 3rd last, etc.
Note that I am suggesting that you feed data to the the digits from right
to left. The reason for this is so that you can easily add digits without
having to affect your code. Say you have code that writes data to 8 digits
of display but your current client needs only 6 digits. The same code
works with both: the extra bits just spill the end of the last digit's
shift register.
My last suggestion is that you put more LEDs in series if you can. Here's
how we work the numbers: assume a 12 Vac power transformer is used to power
the project. After you rectify and filter that, you will have somewhere
between 15 to 17 Vdc unregulated sitting on the filter capacitor. Each
red LED drops about 1.7 Vdc (green & yellow LEDs drop about 2.1 Vdc, white
& blue LEDs drop about 3 Vdc). You want to have about 5V drop across the
current limit resistors: this ensures that changes in the supply voltage do
not affect LED intensity too much. Finally, there is less than 1V dropped
across the TPIC6595 when it is on.
Given those numbers, (16 - 5 - 1) / 1.7 = 5.88 LEDs --> 6 LEDs in
series. Now re-spin the numbers to find out the value of the resistor: (16
- 1 - (6*1.7)) / .02 = 240R. Power rating is: I*I*R = 0.096W --> 1/4 W
resistors are fine.
So your strings would have 6 LEDs in each where possible. If you need more
LEDs, use more strings. Each output on the '6595 is rated for 250 mA: that
means that you can drive up to 12 strings of LEDs per output pin.
I've assumed that you are operating the LEDs at around 20 mA - adjust the
resistor value as required.
Finally, some segments look better if you have fewer LEDs in them. That is
no problem: just figure out the voltage drop you will have and calculate
the required resistor value. The displays that I used to build had 3
different resistor values because the strings were different lengths.
One last thought: you can make a very good looking display if you control
the corner LEDs separately. I just used a value of 1 LED in the string
when calculating the resistor (its only the corner), then used 2 diodes to
feed that corner from the adjacent segments. Works like a hot damm!
dwayne
--
Dwayne Reid <dwayner@p...>
Trinity Electronics Systems Ltd Edmonton, AB, CANADA
(780) 489-3199 voice (780) 487-6397 fax
Celebrating 18 years of Engineering Innovation (1984 - 2002)
.-. .-. .-. .-. .-. .-. .-. .-. .-. .-
`-' `-' `-' `-' `-' `-' `-' `-' `-'
Do NOT send unsolicited commercial email to this email address.
This message neither grants consent to receive unsolicited
commercial email nor is intended to solicit commercial email.