1. This forum section is a read-only archive which contains old newsgroup posts. If you wish to post a query, please do so in one of our main forum sections (here). This way you will get a faster, better response from the members on Motherboard Point.

STN LCD AC Bias frequency problem

Discussion in 'Embedded' started by Not Really Me, Nov 20, 2006.

  1. We are using a Sharp LH75401 with a monochrome STN LCD in 320 line x 240 pix
    QVGA portrait mode.

    In our application, our STN LCD manufacturer (a company named Truly) doesn't
    bother giving a real spec for the AC Bias (signal called LP on the Truly
    docs), but tells us it should be half the frame rate.

    In our case, the frame rate is 168 Hz so they want an AC Bias of 84 Hz.

    On page 13-21 of the LH75401 User Guide, Timing2 register, field name ACB,
    AC Bias Pin Frequency for STN displays says the the value should be (line
    clocks) -1. The field is only 5 bits wide and our display is 320 lines, so
    the description does not make sense. This seems to mean that the AC Bias
    rate can not be any slower than 1/10th of our frame rate (320/32)

    Does anyone have any information on what the typical relationship between
    the AC Bias and the Frame Rate should be on STN LCDs?

    Not Really Me, Nov 20, 2006
    1. Advertisements

  2. Not Really Me

    larwe Guest

    Yes it does make sense. If you want the AC bias pin to clock at (hsync
    rate) you put 0 in that register. If you want it to clock at (hsync
    rate / 2) you put 1 in there.

    LP usually means "line pulse", or rather horizontal sync; are you sure
    you're not misreading the datasheet for the LCD?
    larwe, Nov 21, 2006
    1. Advertisements

  3. Not Really Me

    Didi Guest

    In our application, our STN LCD manufacturer (a company named Truly) doesn't
    Last time I used an STN LCD was 12 or so years ago, but I am pretty
    sure I remember the same signal coming out of the Hitachi chip I
    used then (along with plenty of external logic). It would just toggle
    every V-sync, IIRC it was used by the LCD module to keep reversing
    its internal bias (DC would damage it). To my memory, all LCD
    modules were pretty much the same regarding this signal, so
    my guess is you can just use the one supplied by the chip, if this
    is what it is mean for and if it looks like what I described above,
    more or less. Nowadays, I use TFTs, which keep this sort of thing
    internal, but I would be surprised if the mono LCDs have changed
    a lot since those days.

    Didi, Nov 21, 2006
  4. Scott,
    * The 401 is the one with the colour controller rather than the mono but I
    believe that that's OK.
    * I think that the LP pin is not the AC bias. Typically it is called M. Some
    displays generate it internally.
    * The Sharp docs are as confusing as ever. I think the "ACB =
    (lineclocks)-1" ought to be frame clocks. Normally the M signal will toggle
    every few frames - in your case every frame, so I'd expect ACB = 0. However,
    the text in chapter 14 also suggests line clocks using different words, so
    it might be possible thats what they mean but I doubt it.
    * 184 Hz frame rate seems a bit high to me. I'd expect 75-100 Hz.

    I have a LH75410 based system here with mono 320x240 (landscape) display but
    changing ACB makes no visible difference. Perhaps the display I have isn't
    using M, some do some don't. I'd have to take the machine apart and prod
    with a scope to find out what frequency it's running at. I wonder if I can
    see the change on the port pin from inside the chip...

    Peter Dickerson, Nov 21, 2006
  5. Thanks to the others that replied also.

    Peter, you are correct, it is the version with the color controller. The
    Sharp spec says it does mono also. The chip was chosen over the mono only
    version for some other feature.

    The Truly datasheet seems to be adjusted whenever we ask a question. I
    really think they make it up as they go. The display is a 4 bit wide I/O
    monochrome. The first version of the datasheet was so completely wrong that
    it listed the interface as a simple latched 8 bit port. I guess you put up
    with this type of problem to save 40-50% over competing units. The LP
    signal is the AC Bias. We have tried all values from 0 to 31 in the ACB
    field of the register and only 31 gives a stable display. I am also
    concerned that the frame rate is too high, as well as the clock frequency
    which may be exceeding the LCD spec. I getting someone to verify that now.

    As a bit of probably important information I only received yesterday, the
    real problem is temperature related. The display works fine at room temp,
    but is intended for an outdoor unit. When stored over night in a freezer
    at -20 C, the display still works fine. As the temperature is brought up,
    we start to get a very noticeably flicker (particularly on the upper 1/3 to
    1/2 of the LCD) at around 0 C and it doesn't go away until around 18 - 20 C.
    On some displays it never goes away and the seems to have damaged the
    display permanently. The display is rated for operation down to -20 C and
    storage to -30 C. In the production version we will use a heater to keep it
    at 0 C or above.

    Not Really Me, Nov 21, 2006
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.