18f4685 6 seg clock - dim display

> I *think* this is roughly equivalent to the latches idea - I haven't
> redesigned/built this board yet, but my plan is:
> -Essentially leave the PIC program alone and the display multiplexed, with
> one exception
> - I put latches between the PIC and each 7 segment element, so as you
> suggested, the display elements themselves won't be multiplexed and the
> PIC won't be sourcing/sinking them.

Pretty much, my idea simply means you only need 2 I/O lines on the PIC
instead of 8 or 9 for the parallel latch + the multiplex lines to enable
the displays. Multiplexing the displays will also use less power - You
could combine the two ideas and use a serial/parallel chip for the
segments, and a mux to select which display (each with a transistor of
course) which reduces your I/O lines and saves power all while driving
the LED segments as hard as possible.

have fun,
DLC
> Thanks!
>
> --- On Sat, 5/30/09, dlc wrote:
>
> From: dlc
> Subject: Re: [piclist] 18f4685 6 seg clock - dim display
> To: p...
> Date: Saturday, May 30, 2009, 8:45 AM
>
> I'm a bit late to this story, but how about this idea...
>
> Use a series of serial to parallel chips daisy-chained together. Each
>
> segment driven but 1/8 of that one chip. The PIC doesn't drive any of
>
> them, none of the segments is multiplexed (solves the brightness issue).
>
> My apologies if this has already been suggested.
>
> DLC
>
> Eirik Karlsen wrote:
>
>>>> Yes I summed the display currents, which is wrong of course.
>
>> Other than that I won't give a penny.
>
>> Once you've tried proper driving you'll agree.
>
>>> Prune skrev:
>
>>>>> Just to be pedantic I don't think this is quite right. The circuit is
>
>>> multiplexing the 7-segs so only one display is on at a time so at 20mA
>
>>> the max will be 160mA. Still too much and I endorse the suggestions
>
>>> to use an NPN to source the current on each device. I use 560 ohm
>
>>> current limiting resistors on each segment and get a good bright
>
>>> display using a 1/2 duty cycle (2 digits) with either the MAN71a or
>
>>> MAN6730 (two segment) displays. No need for segment drivers.
>
>>>>> The OP's circuit uses 6 displays so some of the dimness will be
>
>>> because of the 1/6 duty cycle of each display - no easy way round that.
>
>>>>> Eirik Karlsen wrote:
>
>>>>>> No wonder your display is dim. You can't drive it like that.
>
>>>> Assuming this PIC is like the earlier 18series it'll have some
>
>>>> limitations:
>
>>>> Max 20mA sink on any one pin, and max 20mA on all pins summed.
>
>>>> You have 6x8 = 48 segments, and for top brightness expect 20mA pr
>
>>>> segment...
>
>>>> thats 960mA when all are lit (which should also be a design goal).
>
>>>> No way a PIC can deliver that without burning up, and certainly not
>
>>>> without
>
>>>> excessive voltage drop on outputs.
>
>>>>>>> You'll have to use both segment driver and display driver. Thats the
>>>> only
>
>>>> way to get it 'very bright'. I'd suggest using MOS drivers for both
>
>>>> digit and segmens
>
>>>> as this way you can drop all those base resistors.
>
>>>>>>> Here in this clock I've used bipolar drivers, good bright display but
>
>>>> with MOSFET
>
>>>> drives 22 resistors could have been removed (8 in the SIL).
>
>>>> >>>> 20circuit/>>>> HTH
>
>>>>>>>>>> Shadowrun Gamer skrev:
>
>>>>>>>> this is correct, no limiting resistors are needed. I can run this
>
>>>>> display at 100% duty cycle at 5V
>
>>>>>>>>> --- On *Fri, 5/1/09, Paul Laverick //*
>>>>> wrote:
>
>>>>>>>>> From: Paul Laverick >>>> Subject: Re: [piclist] 18f4685 6 seg clock - dim display
>
>>>>> To: piclist@yahoogroups .com
>
>>>>> Date: Friday, May 1, 2009, 5:23 PM
>
>>>>>>>>> Hi,
>
>>>>>>>>> Your diagram shows no current limiting resistors for each
>
>>>>> segment, is this correct or do you have some?
>
>>>>> I had a play with 7 segments once, and had a similar problem, it
>
>>>>> was a long time ago...
>
>>>>> I've heard of two versions of updating the segments, using the
>
>>>>> LSB of Timer0 to update each segment depending on the value (00
>
>>>>> = Segment 1, 01 Segment 2 etc), or every time Timer0 rolls over,
>
>>>>> change segment.
>
>>>>> I tried using timers but found the last segment was always the
>
>>>>> brightest, as all the calculations were happening while it was
>
>>>>> on, giving a longer on time..
>
>>>>>>>>> Hope to of helped,
>
>>>>> Paul
>
>>>>>>>>> *From:* sr_gamer >>>> *Sent:* Monday, April 13, 2009 9:31 PM
>
>>>>> *To:* piclist@yahoogroups .com >>>> *Subject:* [piclist] 18f4685 6 seg clock - dim display
>
>>>>>>>>> if you look in the files section under sr_gamer you will find
>
>>>>> the schematic and the program for a clock I'm building. The
>
>>>>> display is VERY dim.
>
>>>>>>>>> I'd like the display to be very bright.
>
>>>>>>>>> These elements are very bright when supplied with 5v, which
>
>>>>> should be what is coming off of the PIC, so I'm either suffering
>
>>>>> from low duty cycle or having trouble sourcing enough current. I
>
>>>>> tried some 1k pull up resistors between the pins labeled anodes
>
>>>>> and +5v, but this didn't seem to affect the situation.
>
>>>>>>>>>>>>>>> --
>
>>>> ************ ********* ********* ********* ****
>
>>>> VISIT MY HOME PAGE:
>
>>>> >>> LAST UPDATED: 23/08/2003
>
>>>> ************ ********* ********* ********* ****
>
>>>> Regards
>
>>>> Eirik Karlsen
>
>>>>>> --
>
>> ************ ********* ********* ********* ****
>
>> VISIT MY HOME PAGE:
>
>> > LAST UPDATED: 23/08/2003
>
>> ************ ********* ********* ********* ****
>
>> Regards
>
>> Eirik Karlsen
>
>>>>> --
>
> Question with boldness even the existence of a God;
>
> because, if there be one, he must more approve of the
>
> homage of reason, than that of blind-folded fear.
>
> Thomas Jefferson
>
> ------------ --------- --------- --------- --------- -
>
> Dennis Clark TTT Enterprises
>
> www.techtoystoday. com
>
> ------------ --------- --------- --------- --------- -
--
Dennis Clark
TTT Enterprises

Start a New Thread