Post by thread:Driving 7segment displays

John; First, I'm not sure what ou mean by a 4Hz clock. Most digit multiplexing is done over 30Hz. This refers to the digit to digit scan rate. You should take care not to go too high, though, since changing digits takes some time, and if you go too fast, a larger and larger portion of your time is spent changing and not displaying. The displays sure have no flicker left in them, but the brightness will then fall off. I like to set a byte for scan rate in software and let timer0 run the scan. Then I tinker with the scan value to get the best quality scan. Scientific, huh? Chris Eddy Pioneer Microsystems, Inc. >This might sound like a stupid question, but at what speed can i drive >7segment displays?? In my current analogue design i use a 4Hz clock to pulse >the displays, but using a pic this is a little difficult. > >Will the 7segment display anything at all if i switch them at say once every >100 instuctions ( on a 10 Mhz clock )? I suppose it all depends on the duty >cycle. > >Also, do i have to connect anything to the two clock lines in a 16c84 if im >using the internal clock ? > >Thanx > > John

The question isn't what rate CAN you scan them, it's what rate SHOULD you scan them. The rate of fusion for your eyes is in the range of 45 flashes per second (any slower and you'll see flicker). LED's have little (if any) persistance, unlike a CRT, for instance. So, you should scan the entire display at least 45 (or so) times per second. I usually shoot for around 60 complete scans per second to be safe. LEDs react very quickly, so you can scan a lot faster without affecting the display, but it doesn't buy anything unless you're trying to do something fancy like control the brightness by pulse-width modulating them, and it costs processor overhead (a greater percentage of your total time is spent updating the display). I suspect by the wording of the question that something may be missing, though.... {Quote hidden}> >This might sound like a stupid question, but at what speed can i drive > >7segment displays?? In my current analogue design i use a 4Hz clock to > pulse > >the displays, but using a pic this is a little difficult. > > > >Will the 7segment display anything at all if i switch them at say once > every > >100 instuctions ( on a 10 Mhz clock )? I suppose it all depends on the duty > >cycle. > > > >Also, do i have to connect anything to the two clock lines in a 16c84 if im > >using the internal clock ? > > > >Thanx > > > > John -- Andrew Pines spam_OUTapinesTakeThisOuTcosmodog.com http://www.cosmodog.com

Ok, just to make sure you know that Toshiba produce a nice and low cost chip to control 4 x 7seg displays, they do decode, multiplex, driver scan with brightness control in 16 levels, serial interface and more. Forget about scan rates and software overhead. :) The problem is; sometimes you just don't have more processor time to do it, and even so, it will involve transistors or some other driver chips... sometimes expensive than this Toshiba solution. For common anode: http://www.marktechopto.com/tb62709f.htm http://www.marktechopto.com/tb62709n.htm They also produce 4 digit x 7 seg in a single package with 12 pins (6+6, 100mil x 600 mil spaced) code MTN 4456R-11A, so save you some pcb work, perfect for the above chips. http://www.marktechopto.com/fourdigit.htm Marktechopto people are very nice and they deserve this free ad. They also have several other optoelectronics products, including drivers for 5x7 dot char displays, and the displays. If you know some other manufacturer similar chip, email me or post here, I think more people are interested. Wagner.

At 12:15 AM 10/3/99 -0400, Chris Eddy wrote: >them, but the brightness will then fall off. I like to set a byte for scan >rate in software and let timer0 run the scan. Then I tinker with the scan >value to get the best quality scan. Scientific, huh? The essence of science is experimentation, so the answer to your rhetorical question is 'Yes!'. ================================================================ Robert A. LaBudde, PhD, PAS, Dpl. ACAFS e-mail: .....ralKILLspam@spam@lcfltd.com Least Cost Formulations, Ltd. URL: http://lcfltd.com/ 824 Timberlake Drive Tel: 757-467-0954 Virginia Beach, VA 23464-3239 Fax: 757-467-2947 "Vere scire est per causae scire" ================================================================

As stated by others a good scan rate is greater than 50Hz, therefore you should switch digit on a 4 digit display once every 5mS, Normally with a 4MHz clock I set TMR0 on an internal prescalar of divide by 16 and use the overflow interrupt to switch digit. The 16c84 does not have an internal oscillator - it always needs external components, those PIC's which do have an internal oscillator do not need a connection to the external pins - often these pins are dual purpose and are I/O ports when not used for the oscillator. Robin Abbott - robin.abbottKILLspamdial.pipex.com ************************************************************************** * * Forest Electronic Developments * http://dspace.dial.pipex.com/robin.abbott/FED * ************************************************************************** {Original Message removed}

> The 16c84 does not have an internal oscillator - it always needs external > components, those PIC's which do have an internal oscillator do not need a > connection to the external pins - often these pins are dual > purpose and are > I/O ports when not used for the oscillator. Oops, missed that bit. What crystal and cap values should i use then to get a nice stable 10Mhz clock; or should i use an RC circuit instead? I need it to be pretty accurate though. My knowledge on PICs is still very much in its infancy, but im learning. My electronics is even more rudimentary, but i know what i want and i can learn how to do it. And learning how to code in assembler again makes me sleepy :-) Basically what im trying to do it read 2 frequencies and display the values. One frequency is the pulses from one of the coils on a motorbike. The other is for it's speedo. The problem with the revcounter one is that i have to multiply the freq with a constant value in order to still display the correct value. Should i do this in assembler or with a PLL? Thats all for now. John

Do you know a www addy for a datasheet. http://www.maxim.com = some pharmaceutical company. {Quote hidden}> -----Original Message----- > From: pic microcontroller discussion list > [.....PICLISTKILLspam.....MITVMA.MIT.EDU]On Behalf Of David J Binnington > Sent: 03 October 1999 12:20 > To: EraseMEPICLISTspam_OUTTakeThisOuTMITVMA.MIT.EDU > Subject: Re: Driving 7segment displays > > > Wagner wrote: > > > If you know some other manufacturer similar chip, email me or post here, > > I think more people are interested. > > The MAXIM MAX7219 works very well and only needs one resistor to > select the > maximum segment drive current. Drives any combination of up to 8 > digits or > discrete LED matrices. Requires 3 I/O - clock, serial data & chip select > but the clock and chip select can be multiplexed with other I/O. > Brightness, > number of digits, decoding etc. etc. are all under digital > control. Serial > register also has an output to allow devices to be chained. > > Dave >

John De Villiers wrote: > One frequency is the pulses from one of the coils on a motorbike. The > other is for it's speedo. The problem with the revcounter one is that > I have to multiply the freq with a constant value in order to still > display the correct value. Should I do this in assembler or with a > PLL? What's the PLL got to do with it, and why's it an alternative to assembler? Do you mean a HLL? I'm applying traction to your femur ;-) Actually, I think I know how you are muddled. If you *were* to use a PLL, you have to code the PLL in assembler (or if it suits you, "C"). It goes sort of like this: At a given rate (determined by your timing loops) you add a phase increment to a register which is your phase accumulator. Every time you receive a timing pulse, you look at the phase accumulator and see how close it is to the "zero" point. The error is scaled, and added to/ subtracted from the phase increment. This can be made to work as long as your internal "flywheel" (PLL) responds faster than the *real* one. The phase increment is scaled closely (binary factor) to the actual RPM value. There is some risk of locking to a (sub-)harmonic with this phase detection algorithm, perhaps you should use the "digital" algorithm from the 4046 chip instead... Interesting concept really, you do the measurement "backwards" using successive approximation of your presumed frequency to your (sub- multiple) real one. -- Cheers, Paul B.

John . . . . anything from static operation , up to micro-seconds is possible for the LED drive. You're right , the question is * what is your duty cycle going to be * and how many * digits * are you going to drive ? The 4 hz. pulse you are currently using can easily be reproduced by creating a * time base * and using the interrupt ( tmr0 ) to enter an * incrementing * routine. The bit count ( 2,4,8,16,etc) can be used to select the * multiplier * of the * time base * and hense the timer period. Don't forget that the PIC is capable of * D.C. * operation up to the upper frequency limit. So , if you find that selecting a lower clock frequency would work out better, then go for it . There are numerous application notes on the PIC website that give good examples of multiplexing several LED digits and most are written with a clock speed of 4 mhz. in operation. Personally , I like to use a 5 millisecond , TMR0 interrupt loop , then during the service routine , update all of the digits data , then select the digit * drive * , for the LED which is next in the multiplexing routine. So , this produces 5 milliseconds of * on * time , per digit and is repeated each 20 milliseconds. This will produce a * flicker-free * update to the LED digits and would still allow you to use a *servo * to blink the display if desired. Use PORTB to drive all of the seven segments , plus the decimal point ( if used ) , in parallel . Use PORTA to drive the LED digits , either directly ( if below 25 mA , common cathode ) or I recommend the use of drive transistors ( 2n2222 or equiv. ) placed in series with the common cathode of each LED digit and ground. Using the transistors ( instead of direct drive ) will keep the LED digits off , when PORTA is low and allow for a higher drive current , per digit. Don't forget the current limiting resistors , in series with each LED * segment * and PORTB. Remember , you are multiplexing the eight segments and need to specify the current limiting resistor for a single segment . Also , the resistor will need to be reduced in value , due to multiplexing , to achieve the same LED illumination that you are calculating using a static formula ( I=e/r ). The individual digit duty cycle is 25%. With the above timing , expect a 50 hz. refresh rate from the four LED's . . . . . If you dig through the Microchip applications , on the website , you'll find a few that are using the 16C7x as an example for the multiplexing . . . . . if you have any trouble getting these to work . . . . then email me * off-line * and I'll give you some working examples. About the internal clock . . . . . . I'm not sure what you are referencing to , but you will need to provide either an RC , Xtal , resonator or * external - oscillator * to drive the chips clock. You then can divide this frequency by four to obtain the frequency of the * internal clock * . . . . . which is unaccessible by you , but determines the time required to execute instructions. Some require one , some require two. Oh . . . . almost forgot . . . . no such thing as a stupid question , get as many views as you can and pick the one's that are reasonable to you. Although I'm not one of them . . . .you'll find some very talented PIC people on the list , who really know their stuff ! Regards . . . . Steve Kelley {Original Message removed}