LC-15SH7U Button Question

I have a Sharp LC-15SH7U that took a power surge & went dead. After fixing a shorted zener diode in the 33V power supply & a shorted dual fet in the backlight inverter the tv powers up just fine & has a normal picture but now the top six buttons don't work properly. Power works properly Menu, Input, & Volume Down don't work at all. Volume Up works as Volume Down. Channel Down works as Volume Up. Channel Up works as Channel Down. Button order is Menu, Vol down, Vol up, Chn down, Chn up, Input, Power.

It has a four wire interface to the main board. I haven't traced the circuit out yet, but a quick glance shows all switches are connected to pin 1. I imagine the other three pins work a set of buttons each, but I'm not sure. Is this a common configuration for Sharp TV's? I haven't been able to find any service info for this TV, so I'm sure I'll have to start tracing.

Thanks for any help, Steve

Reply to
Steve
Loading thread data ...

On Thu, 26 Feb 2009 21:27:58 -0600, Steve put finger to keyboard and composed:

I have seen the following technique used in corded VCR remote controls:

SW1 SW2 SW3 SW4 _|_ _|_ _|_ _|_ | | | | | | | | GND o-- R1 --o-- R2 --o-- R3 ---o-- R4 ---| | SW o-------------------------------------|

The resistors could be binary weighted, ie R, 2R, 4R, 8R.

I can't see how this scheme fits in with your setup, though.

- Franc Zabkar

--
Please remove one \'i\' from my address when replying by email.
Reply to
Franc Zabkar

It's a pretty clever little circuit. It's a four pin connection. Pin

4 is ground, 3 is 3.3V. Pin 1 goes to the menu button with a basic pull up resistor, switched ground configuration. Pin 2 goes to all the other buttons in a voltage divider configuration. Power button pulls pin 2 to ground, the other buttons switch in different legs of the divider. The voltages measured are CH+ .50V CH- .88V V+ 1.30V V- 1.68V Input 2.51V Power 0V This corresponds to the circuit tracing. By that, I mean the CH+ button swtiches in 1.5k, CH- switches in another 1.5k for 1.5k + 1.5k, V+ switches in 1.5 + 1.5 + 2.2k, V- 1.5 + 1.5 + 2.2 + 2.7k, and input 1.5k + 1.5k + 2.2k + 2.7k + 10.5k. All are divided into themselves plus the pullup of 10k (for V- it would be (1.5 + 1.5 + 2.2 + 2.7)/(1.5 + 1.5 + 2.2 + 2.7 + 10). I'm awful w/ ascii schematics, I can give it a shot if anyone wants. It looks like the circuit's fine, so it's somewhere else downstream that either isn't measuring properly, or is doing the wrong function for the wrong voltage. Not suprising for a lightning strike.

Steve

Reply to
Steve

Looks like I'm SOL. The traces each go through a 33 ohm resistor into pins 3 & 4 of the uC. I imagine the prom got corrupted in the surge. It has an EEPROM port, I bet if someone could flash it it'd work fine. Oh well, I haven't tried the remote yet. This one was kind of fun, the board was a PITA though, not meant for rework.

Steve

Reply to
Steve

Well, after some probing, it looks like it might not be wiped out program. The divider input is 3.3V, but the output is 3.5V when open circuit. The only thing else connected to the circuit is the uC. It looks like the uC pin may be pulling the divider a little high causing all the buttons to shift up, causing the higher ones to not work, and the lower ones to be shifted one over. What is curious, though, is that the menu button has its own pin, and it doesn't seem to work. The menu button is only high or low, no divider output. When high, it is pulled up to 3.5v as well, instead of 3.3v.

Reply to
Steve

On Fri, 27 Feb 2009 21:30:06 -0600, Steve put finger to keyboard and composed:

Why not install a buffer between pin 2 and the switch commons? You could use a 3.3V rail-to-rail op-amp wired as a voltage follower. I've "repaired" faulty I/O pins on uCs using similar methods. It appears that the Menu button is unrecoverable, though.

- Franc Zabkar

--
Please remove one \'i\' from my address when replying by email.
Reply to
Franc Zabkar

Good solution, shouldn't take more than a few parts & I could probably just glue them to an unpopulated section of the board. This was kind of a botched repair anyway, I had to guess at equivalent parts and the board was damaged during the rework. Oh well, can't win them all.

Thanks, Steve

Reply to
Steve

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.