View Single Post
  #8   Report Post  
Posted to sci.electronics.repair
[email protected] ohger1s@gmail.com is offline
external usenet poster
 
Posts: 191
Default Need *legible* datasheet

On Saturday, March 4, 2017 at 6:22:17 AM UTC-5, jeanyves wrote:
On 2017-03-02 14:53:54 +0000, said:

On Thursday, March 2, 2017 at 9:38:47 AM UTC-5, wrote:
On Wednesday, March 1, 2017 at 4:52:53 PM UTC-5, wrote:
I suggest you call Mouser or DigiKey, or another supplier that stocks
this device and ask them to scan you a high-res datasheet. Better yet,
the manufacturer if they have either e-mail or a US rep. I have
found Mouser very helpful to this end. Peter Wieck
Melrose Park, PA

Thanks guys, but neither Digi nor Mouser sell this IC. I can get it
through Alibaba but they don't seem to understand much English. I did
send an email to O2Micro with a datasheet request, but I have a feeling
I'll be hearing from Ed McMahon with a million dollar check first.

Hopefully O2Micro will come though.


After posting the above, I found an email from O2Micro in my spam
folder. This is the reply:

Thank you for your interest in O2Micro solution. In order for us to
provide you with the information requested, please kindly answer the
following qualifying questions. (An inquiry from an end-user will
generally be declined due to our limited resource.)

1. What's your company website?
2. What's purpose of this request?
3. How do you learned of this IC?
4. What type of system is this IC going into or in?
5. Are you an OEM or ODM? If you are an OEM, who is your customer?
6. What's the project's estimated annual usage?
7. When is the estimated time-frame of mass production of this project?



I have to go.. I think I heard Ed McMahon at my front door....


answer to their email. you're not asking for samples, you're asking for
a readable datasheet.
you have a contact there, go on with it until they clearly refuse to
answer you.

--

Jean-Yves.


I no longer need the datasheet as I was forced to draw one up from what I could determine on the board. The long story is that this is a 6 channel controller IC used in a Toshiba 58" LCD TV, one that (like most LED back lit LCD TVs), has a propensity to kill LEDs in 2 to 4 thousand hours. I do a lot of LED replacements but don't want them coming back in a few months for the same problem, particularly since the chances of something (like a bonded ribbon failure) going wrong increases every time the screen is taken apart, no matter how carefully. Setting the customer's menu control to a reasonable level doesn't work as any playing around in the menu almost invariably resets the back light level to 100%. The best thing to do is to remove the ability of the end user to set his back light to "retina burn", so I modify the circuit to reduce the back light wattage by about 30%, which still gives a respectable picture but should extend the life exponentially. Since I've been doing this about three years, none have returned.

So I sat down and drew out the six channel pin outputs, the six channel current sense inputs, the Vcc, the ground, the PWM input, the enable input, the dc boost gate output, the boost current sense input, etc. until I was left with a few pins that weren't so obvious. On most circuits I raise the value of the source resistors of the LED output mosfets to trick the controller to lower the output, but this version would shut down the back lights if the drive was reduced by 10%, so I had to play with the input side.

Most controller ICs are designed with both an analogue (dc) and a PWM input to control LED brightness, and the analogue is often not used (or used as a baseline output) but is terminated with a resistor to ground, or fed in a selected dc from a res/div network. This toshiba used a PWM input for active back light control, and that can only be modified by changing the firmware in the TV, so I turned to the analogue input. In this case, it was terminated to ground through a resistor and capacitor. Changing the value of that resistor got me to 30% reduction of output with no shutdown during the boot sequence.

As for O2 Micro, I did return the email and told them that we were an OEM but were no longer interested in their product as we didn't feel comfortable with them if the courtesy of an emailed datasheet was a strain on their "limited resources".