3
\$\begingroup\$

I'm working with a board that has multiple 14 & 7 segment displays all being addressed using an HT16K33 where various segments are either not displaying or are being lit when they shouldn't.

In the case of the 14-segment, which has 3-characters, what I'm seeing is that when one segment in particular is set, a second one also lights up. I confirmed this by running a script which starts with an empty 16 bits and slowly shifts 1's up through the positions to cycle the individual segments. Also, this behavior is the same for all 3 characters in the component.

Using the segment wiring diagrams below, when segment H is lit, segment L also lights up and vice versa. It appears as though both are slightly dimmer than the rest because when both are toggled on, they appear to be the correct brightness. See the video of the issue happening below and note that it's on more than one character (it happens on all 3).

video of extra segment lighting up

I visually checked the board and the chip to see if perhaps there may have been some solder bridges on the HT16K33 or the LED but everything appears completely normal. I'll add that I purchased this device fully assembled as part of a set of 2 and the other board works fine. The boards are identical except for having different I2C addresses. The likelihood of this being an issue with the design of the boards is slim.

At this point, I'm completely baffled and, while I would first want to chalk it up to a faulty LED component, the fact that I'm having similar issues with other 7-segment characters that are also wired up to the same HT16K33 leads me to believe it might be something with the chip or one of the surrounding components that could affect those particular segments to affect each other (a bad pull-down resistor perhaps?).

I found this post which possibly describes similar issue (calling it a display artifact) but this is my first time delving into LED matrices and the HT16K33 drivers so I'm hoping that someone with more experience than me might recognize this behavior and point me in the right direction.

Diagrams from the product listing on Alibaba (for the same component but different colour) Pinout of 14-segment LED Wiring of 14-segment display

As an additional reference, here is a link to the circuit diagram of the board itself (design is "Time-Circuits-Display" by "CircuitSetup(John)"- not mine). Schematic is copied below in case the link dies.

schematic
(click image for full-size version)

\$\endgroup\$
2
  • 1
    \$\begingroup\$ @SamGibson I updated my question with the missing link to the other post and clarified my wording. I was referring to the segment letter image to illustrate which segments were affected. Also regarding the circuit design, while it's always a possibility, I I have another identical unit that functions perfectly fine. The fact that this is happening on multiple segmented displays leads me to believe it's either something faulty with the chip or something wrong with the assembly but I'm hoping to get more information before pulling out the hot-air station and replacing the chip. \$\endgroup\$
    – nageeb
    Commented Apr 10 at 16:23
  • 1
    \$\begingroup\$ The fact that segments "I" and "M" first illuminate together at lower brightness and then at full brightness would suggest a solder bridge. If the individual alphanumeric characters are multiplexed off a single driver, then the problem would manifest across all of them. I suggest reflowing all the joints in the "I" signal path. It's quick, it's hopefully relatively easy, and has a negligible chance of causing additional problems. \$\endgroup\$
    – vir
    Commented Apr 10 at 16:42

1 Answer 1

6
\$\begingroup\$

A short is suggested. It would affect all displays that use the shorted segments.

Your segment numbering is different from the linked schematic.

Between j & n on the schematic numbering or between i & m using your numbering.

Such a short could exist within a display, or a faulty etch or hairline short in the PCB. It seems unlikely to be soldering issue since the pins do not appear to be adjacent on any component, at least in the linked schematic, but that's another possibility.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.