4 digit 7 segment display usage?

Other Hardware Displays
1

I can't figure out how to use this display; it's a very unusual setup:

Pin 1 is the common cathode for two indicator lights, segments D, E, F, and G of digit 2, segments A, B, and C of digit 3, segments D, E, F, and G of digit four, and the colon.
Pin 2 is the common cathode for one more indicator light, segments B and C of digit 1, segments A, B, and C of digit 2, segments D, E, F, and G of digit 3, and segments A, B, and C of digit four.

This is perfectly acceptable, until you notice the following:

Pin 6 is the common anode for segment E of digit 2 and segment C of digit 1
Pin 7 is the common anode for segment G of digit 2 and segment B of digit 1

And so on. Every pin from 6 to 16 is a common anode; only 3, 4, 5, 17, and 18 are anodes for only one led.

The problem is that if I need a segment with pin 1 as its cathode, and another with pin 2 as its cathode, and the two don't share an anode, two other unrelated segments light up! How am I supposed to use this display?

2

Is that data from a data sheet or your measurements. a display that complicated just doesn't make sense. Do you have a pert # perhaps stamped or marked in some manner? More information or an accurate schematic would be appropriate to an accurate answer.

Doc

3

It's from my own measurements. It's labeled as "CT-9227A" in three different places (and oddly, two different fonts) on the breakout board it's mounted on. I took it from a clock I disassembled that said it was manufactured in Oct. 1999, so it's not exactly the most modern device out there.

And yes, it doesn't make sense at all.

EDIT: would you like a complete pinout? Because I could give you that, if you need it.

4

well when the data doesn't fit the problem them is the problem bad or the data? sometimes one can miss the pins... or mid read them. How did you go about discovering which did what? If I remember right I advised you to get a battery and a resistor and some clipleads or equivalent and go probing, making a diagram as you go along. remember that the positive lead will light an Led when it is more so than the anode. Try again? or try to find the data sheet, either method is likely to have some possibly different results if done again as now you know now there is something amiss.

Doc

5

That is indeed what I did; attached a resistor to the negative terminal of the battery, plugged it into one pin and went along all the rest of them with the positive lead.
After repeated tests, it does exactly what I expect it to when I connect any given pins. Which is fairly useless, since unwanted segments light up, exactly as expected. This was a perfectly functional clock--I tested it out, as I do everything, before taking it apart.

6

Well sir I refer you to my last post again... When the data doesn't fit??? Yes please post the diagram by all means. I think you might be fooled by sneak path's places in a design where diodes (LED's provide un wanted or accidental biasing were all the segments of equal brightness except the decimal point? if not that might be an indicator... unless there is... POST the diagram and I will do my best.

Doc

7

All LEDs are the same brightness (except when I ran out of 100R resistors and started using 220R ones instead. I need more resistors.)

Here's the schematic I tested it as having: (my first time making a schematic so sorry if it's sloppy)

7seg.jpg
7seg.jpg1217×396 4 KB

8

Two suggestions 1. leave out the junction circles except where they join more than 2 wires. makes the diagram hard to read. 2. label the segments. the usual is A seg at top rotating clockwise around the display looking at the face of the display where the G seg is the middle bar and label the DP and Colon identifiers. What you appear to have almost looks normal sometimes the common cathodes or anodes (common) are split, diallight 7660 is an example of that kind of display... easier to fab the substrate I guess... one less wire or via I guess. with that information both you and I will know, had you mapped the seg's... You would have seen it your self, i am sure.

Doc

9

I know how the segments are mapped; they're in order from A to G on each digit.

And I'm afraid I don't see what you're thanking about, nor do I see how to light certain segments without lighting extraneous ones as well.

10

I've looked at that drawing three times now and I still don't see the circles I mentioned the 'wire junctions removed' (they are distracting because one doesn't generally use them on a bend in a wire) and I still don't see the segments marked. I read your comments about multiple segments and I have only one question, are they on the same cathode or anode? if so you made a mistake if not don't worry about it as the LED is scanned no single set of segments is on for very long. I wanted to look and verify that the seg leads made sense.

Doc

11

While you are redrawing your diagram please expand it vertically so the horizontal lines are further apart and easier to follow.

You should look into the possibility that the display uses 'charlieplexing'.

Don

12

I'm sorry; I haven't had access to my computer since recording your reply. And I won't until tuesday, unfortunately.

The junction dots were added automatically by the software I used.

There are multiple segments on the same anode and on each cathode, but I may have gotten the terms mixed up.

And scanned? What exactly are you referring to there? Does every segment get turned on and off rapidly as others are lit?

floresta:
While you are redrawing your diagram please expand it vertically so the horizontal lines are further apart and easier to follow.

You should look into the possibility that the display uses 'charlieplexing'.

Don

I don't believe it uses charlieplexing; from what I've observed, nothing acts as an anode for some lights and a cathode for others.

13

... from what I've observed, nothing acts as an anode for some lights and a cathode for others.

I'll have to take your word for this since you don't seem to want to provide a diagram that older eyes can use. Perhaps you could redraw it by hand and take a picture of the drawing.

Don

14

I'll do a diagram when I get back home tuesday; sorry to make it so hard to read.

15

xolroc:
I can't figure out how to use this display; it's a very unusual setup:

It is by no means unusual at all - it is how almost all LED clock displays work, corresponding of course, to a particular series of clock chips.

xolroc:
I took it from a clock I disassembled that said it was manufactured in Oct. 1999, so it's not exactly the most modern device out there.

And they are probably still making them. It is modern enough. Why change a simple, cheap design?

floresta:
You should look into the possibility that the display uses 'charlieplexing'.

Definitely not Charlieplexing. It is even more clever than that. :smiley: I have more recently explained the design here.