Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Voltage Reg Heat?

Status
Not open for further replies.
LED drive current

What current are you wanting to feed the LED's with?.
That's really the key question to answer first. What is the per-segment recommended current for the display you're using? You could drive a high-efficiency display directly, normal LED displays require drive transistors to handle the 6x current (or a driver chip like the ULN2003 - note, VCEsat is higher on these than using individual transistors because they're darlington arrays) on the multiplex lines.

I couldn't find a datasheet for the 74S138, but the 74LS138 has a maximum sink rating of 8mA per output. It would be an acceptable choice for high-efficiency, low-current displays, allowing you to multiplex up to eight 7-seg displays with three lines.
 
Well, I layed out a new circuit. I started out using the 74hc138 and the 4511, but the display wasn't getting enof power.

From what i can gather from the old design the displays are using around 40ma a segment. I calculated that out for the whole thing and it was around 1.5A, ?? Wow :shock:

The displays that I am using are Lumex LDD-C812RI. It says stedy current of 25ma.

Is this what I should be pushing into them?

My new design uses portc for the segments and porta is hooked to 3904 transistors to the CC of each.
The only problem is the display is dim. What is the best action?
I think I have a 2803 somewhere.

How fast should I cycle the display, #inst cycles? Right now it is about 850.

Thanks
Kent
 
kentken said:
Well, I layed out a new circuit. I started out using the 74hc138 and the 4511, but the display wasn't getting enof power.

From what i can gather from the old design the displays are using around 40ma a segment. I calculated that out for the whole thing and it was around 1.5A, ?? Wow :shock:

Which was a bit naughty whent he 4511 is rated at 25mA!.

The displays that I am using are Lumex LDD-C812RI. It says stedy current of 25ma.

Is this what I should be pushing into them?

If you're multiplexing them you can use 150mA, 6 times their maximum power (as they only get 1/6th of it) - but, as you have concerns over power consumption, I would suggest running them less, and adjusting the series resistor values for minimum current with sufficient brightness.

My new design uses portc for the segments and porta is hooked to 3904 transistors to the CC of each.
The only problem is the display is dim. What is the best action?
I think I have a 2803 somewhere.

Try posting a sample section of the circuit (no need for all of it) so we can see exactly how you are connecting them. Don't forget to include the series resistor values!.

How fast should I cycle the display, #inst cycles? Right now it is about 850.

If it's too low you will see visible flicker, all you need to do is make it fast enough so you can't detect the flicker - probably the easiest way is to use a timer interrupt, which is what I used in my tutorial. You can then simply alter the timer interval to alter the refresh rate.
 
I have included a basic sch of what I am using now.

I did some more testing on the old design. I put another diode, and then a large 3.3Ohm 5W resistor before the regulator. Didn't help, still heated up to much. I even added another tant cap on the output of the reg. :(

It has a draw if .58A. with a 7805 Regulator.

What I don't understand is why this at a draw of 580ma on a 1A regulator makes more heat then a 350ma draw on a 500ma reguator in another project.

I think what I might do next is putting a 2803 in the seg lines.
If you know of any other Ideas, I am lisening.

thanks agin
Kent Kenison
 

Attachments

  • newcountersch.jpg
    newcountersch.jpg
    97.7 KB · Views: 423
kentken said:
I have included a basic sch of what I am using now.

I did some more testing on the old design. I put another diode, and then a large 3.3Ohm 5W resistor before the regulator. Didn't help, still heated up to much. I even added another tant cap on the output of the reg. :(

It has a draw if .58A. with a 7805 Regulator.

What I don't understand is why this at a draw of 580ma on a 1A regulator makes more heat then a 350ma draw on a 500ma reguator in another project.

The dissipation depends on the voltage dropped across the regulator, if it's got a lower input the dissipation could be far less. Are the two tantalum capacitors as close as possible to the 7805, they need to be close to stop it oscillating - if you have a scope, make sure it's not oscillating.

But you MUST fit current limiting resistors between PortC and the segments - maximum current is about 25mA per pin, there's also a total maximum for the entire chip, I should check that as well in the datasheet.

If you limit the total segment current to 25mA per segment, this gives 175mA maximum consumption by the displays - the rest of the circuit takes very little, so your current should be much reduced.

I think what I might do next is putting a 2803 in the seg lines.
If you know of any other Ideas, I am lisening.

You certainly require something to feed the segments, you don't even have any series resistors to limit the current, so you are essentially shorting the PIC outputs down to ground!. PNP transistors (and current limiting resistors) are all you require, or just resistors if 25mA is enough.
 
Try it without drivers first, just put resistors to the segments like nigel said, 150ohm or so, that'll give about 20mA per segment...

I do this a lot and the display is pretty clear...
 
Thanks, I now understand the need for the resistors.

I am also running into the problems that I knew I was. When I make tmr0 rollover small enof to make a brighter display it gets in the way of the other parts of the program.
I have included the sch and the asm file, please take a look and see if there is any tips that you might have.

I am also going to dig up the biggest heatsink that I can find that will fit in the box. Might Have to be a alum plate.

Please let me know what will make this work better.

Thanks
Kent[/i]
 

Attachments

  • newcounterpcb.jpg
    newcounterpcb.jpg
    110.5 KB · Views: 392
  • balecounterrev2.asm.zip
    2.9 KB · Views: 127
My first thought was 'what has timer0 got to do with how bright the display is?'.

I then looked at your code, you don't seem to have got the idea of multiplexing?. You are servicing all digits on every interrupt, resulting in reduced brightness of the display - you should service ONE DIGIT on each interrupt, leaving that display lit, until the next interrupt comes along when it moves to the next digit, and so on. This gives your program lots more time in between interrupts, and makes the display a lot brighter.

Also, you don't appear to be saving and restoring registers in your interrupt routine, it's rather important that you do so.
 
OK, that makes more sence :wink:
I gess I have been so stressed about this working that I missed that.

I will make the changes, and I will do the restore routine in the int.

Also, If you noticed I load the EEPROM will 0's, so when it starts there isn't FF's in it. The only problem is that it wont save to it in the program, and it did before, any ideas?

Thanks
Kent
 
kentken said:
Also, If you noticed I load the EEPROM will 0's, so when it starts there isn't FF's in it. The only problem is that it wont save to it in the program, and it did before, any ideas?

Do you mean it's not saving it in the HEX file?, if so you must have made a mistake somewhere - I'm presuming your PIC programmer software handles EEPROM data OK?. If you mean your program isn't saving any changes, you may have a problem with your EEPROM writing routines.
 
It has the problem writing in the program, I don't know what changed to make it stop working.

I Changed the program to scan through the displays, It works ok but not right, how would you sugest I do this?

also, the display is bright now.

Thanks again
Kent
 
It has the problem writing in the program, I don't know what changed to make it stop working.

I Changed the program to scan through the displays, It works ok but not right, how would you sugest I do this?

also, the display is bright now.

Thanks again
Kent
 
It has the problem writing in the program, I don't know what changed to make it stop working.

I Changed the program to scan through the displays, It works ok but not right, how would you sugest I do this?

also, the display is bright now.

Thanks again
Kent
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top