Author Topic: Failure mode of LCD segments?  (Read 3637 times)

0 Members and 1 Guest are viewing this topic.

Offline BurningTantalumTopic starter

  • Frequent Contributor
  • **
  • Posts: 358
  • Country: au
Failure mode of LCD segments?
« on: June 01, 2016, 02:38:15 am »
I have a set of electronic scales for repair- they are for calibrating the output from a farm air-seeder, but are really little more than glorified bathroom scales.
The display is a 6 (or 5 1/2) digit LCD display with one small text of 'Zero' to indicate when the load cell is within a zero band.
All of the 7-segment characters have the 'e' and 'c' segments missing.
The LCD has 20 x 0.1" pitch lead-outs from 'clips' that contact the deposited conductive tracks. These clips are sealed under some type of clear compound.
I have very little experience with LCD displays other than the usual cleaning and re-seating of zebra strips.
Is this failure indicative of a display failure or a drive circuit failure? The driver is a uPD7225.
I dived in with a scope, with happy memories of multiplexed LED 7-segment displays, and was a bit baffled by the stepped waveforms that I found.
I got an informative document- EFM32 LCD Driver - AN0057 - Application Note - Silicon Labs - which was very enlightening, but before I go further I wonder if anyone has any guidance as to the best way to approach this.
I don't suppose that I will find a replacement LCD, but the drive IC is available.
Regards, BT
 

Offline Ian.M

  • Super Contributor
  • ***
  • Posts: 12807
Re: Failure mode of LCD segments?
« Reply #1 on: June 01, 2016, 04:55:53 pm »
Cut and patch tracks to swap a good segment for one of the bad ones.  The result would be scrambled numbers, but if the faulty segment doesn't move, its bad LCD glass (or an O/C connection to it).  If the fault moves to the new segment its a bad driver (or a bad connection to it).
 

Offline edavid

  • Super Contributor
  • ***
  • Posts: 3381
  • Country: us
Re: Failure mode of LCD segments?
« Reply #2 on: June 01, 2016, 05:23:45 pm »
Cut and patch tracks to swap a good segment for one of the bad ones.  The result would be scrambled numbers, but if the faulty segment doesn't move, its bad LCD glass (or an O/C connection to it).  If the fault moves to the new segment its a bad driver (or a bad connection to it).

It's not so easy with a multiphase (triplex?) display like this  :(   It would take some detective work to figure out if the 2 bad segments are on the same backplane phase or not.

Another possibility is that the contrast voltages are bad... I would try to check that first.
« Last Edit: June 01, 2016, 05:25:29 pm by edavid »
 

Offline BurningTantalumTopic starter

  • Frequent Contributor
  • **
  • Posts: 358
  • Country: au
Re: Failure mode of LCD segments?
« Reply #3 on: June 02, 2016, 12:45:18 am »
OK, thanks for the help- I will have a further read before I have a go at it.
The LCD is certainly much more complicated to drive than an LED 7-segment.
 It will be solely for my own education as the usual conditions apply- millionaire farmer sucking his teeth when I mentioned a few dollars, plus he doesn't need to calibrate his seeder until next March !
BT
 

Offline Ian.M

  • Super Contributor
  • ***
  • Posts: 12807
Re: Failure mode of LCD segments?
« Reply #4 on: June 02, 2016, 01:45:20 am »
When identifying segments to cut/patch to test as above,  I'd start from the uPD7225 datasheet.  Because it has the character decoders in it, the odds are extremely high that the segments and commons are wired as per datasheet, so they can be identified by tracing them from the chip pin to the LCD glass pin.
 

Offline edavid

  • Super Contributor
  • ***
  • Posts: 3381
  • Country: us
Re: Failure mode of LCD segments?
« Reply #5 on: June 02, 2016, 02:08:37 am »
Well, good point.  OP, are you sure that segment g is working?  For triplex, c, e, and g are connected to COM1.
 

Offline BurningTantalumTopic starter

  • Frequent Contributor
  • **
  • Posts: 358
  • Country: au
Re: Failure mode of LCD segments?
« Reply #6 on: June 02, 2016, 12:55:18 pm »
edavid-
Yes, it is definitely just the e & c segments that are absent.
It is evening here now, so I will have another perusal of the Silicon Labs application notes and try to get my head around the drive method, then have another look with the scope tomorrow.
Thanks again all,
BT
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf