Element ELEFW40C won't turn on.

[Spork Schivago]

Re: Element ELEFW40C won't turn on.

Yeah, I'll run it by them once I pull the flash chip and see if I can dump it, just in case it's bad, although I doubt it is.

I've learned a lot while working on this particular board and that's good. With this board, it's a problem I haven't ran into before and I think that's when we learn the most, when we do something we're not comfortable at.

Anyway, I thought this might be a good time to try out my newly acquired slightly used Agilent 16702B Logic Analyzer. I want to see if I can see the commands going to and from that 25Q16BSIG. I'm reading up on how to use it. It could be a fun experiment, while I'm waiting for stuff to arrive in the mail. I can't contact the customer until after 5:00PM, so I have a little bit of time to experiment with it. I don't know if my logic analyzer is too old though, it might not be able to read it. It might still be worth trying, because then I'll learn something about it, I'm sure.

[vinceroger69]

Re: Element ELEFW40C won't turn on.

would be good too i guess if customer goes for a replacement board as if eeprom/spi flash is bad you can get a good dump of the working board and maybe also repair faulty board if thats what is at fault.i It does not help that we cant find a schematic for this psu i have been looking too but no luck as yet.

[Spork Schivago]

Re: Element ELEFW40C won't turn on.

For my Agilent 16702B logic analyzer, I have two state and timing modules. One is the 16740A and the other is the 16717A

16740A: http://www.keysight.com/en/pd-100000...e?cc=US&lc=eng

16717A: http://www.keysight.com/en/pd-100000...e?cc=US&lc=eng

I wonder if either of these would be good enough to see the information on the bus there. It's definitely a new learning experience trying to do this.

I'm working on acquiring a 16760A State Timing Analyzer Module. It'd be nice if I could snag on them fairly cheap like.

[Spork Schivago]

Re: Element ELEFW40C won't turn on.

Should there be a clock signal going to that 8-pin chip that contains the firmware? If so, there's not. I would think that should be receiving a clock input as soon as the TVs powered on. Isn't that 8-pin chip the thing that controls the television?

[Agent24]

Re: Element ELEFW40C won't turn on.

Yes, it will have a clock pin. Look up the EEPROM datasheet.

No, the EEPROM Chip doesn't control the TV. It just contains some data which gets loaded. It's like the BIOS EEPROM on a motherboard, basically.

It will receive a clock signal to read the contents when the microcontroller requests it, somewhere early on in the startup sequence I expect. If you get no clock signal (be sure you didn't just miss it, or something) the micro is probably not outputting one. Either it's bad, or it's halting somewhere in the startup sequence before it even tries to read the EEPROM.

If that is the case, then the microcontroller could be being held in reset by a faulty reset circuit, bad voltage supply, maybe broken main crystal oscillator...

If you can find the reset pin of the micro, check that that's working normally. Check that the main crystal is running. Check the voltages for the micro are clean and in spec.

[Spork Schivago]

Re: Element ELEFW40C won't turn on.

Yes, it will have a clock pin. Look up the EEPROM datasheet.

I think you misunderstood. I know it has a clock pin. I'm saying there's no clock pulse on that pin. It's 0.000VDC.

No, the EEPROM Chip doesn't control the TV. It just contains some data which gets loaded. It's like the BIOS EEPROM on a motherboard, basically.

It will receive a clock signal to read the contents when the microcontroller requests it, somewhere early on in the startup sequence I expect. If you get no clock signal (be sure you didn't just miss it, or something) the micro is probably not outputting one. Either it's bad, or it's halting somewhere in the startup sequence before it even tries to read the EEPROM.

I will throw my oscilloscope on it tomorrow, but I setup my logic analyzer and believe I did everything correctly. This was my first attempt at using it, but I followed directions and read a lot today. I'm pretty sure I did it right. There was nothing there. Could you give an example of what content would generally be on the EEPROM? Is it stuff like the character chart for various text that appear in the menus, perhaps the menus themselves? I was looking at it like a PC BIOS, which is responsible for initializing the various hardware, the various BIOSes on that hardware, and executing the code at sector 0 of the hard drive to start the OS.

If that is the case, then the microcontroller could be being held in reset by a faulty reset circuit, bad voltage supply, maybe broken main crystal oscillator...

If you can find the reset pin of the micro, check that that's working normally. Check that the main crystal is running. Check the voltages for the micro are clean and in spec.

Okay, I need you to back her up for a second. When you say micro, you're not talking about the flash chip anymore, right? You're talking about the central processing unit of the television (or the TV's equivalent of a CPU), is that correct?

To check that the x-tal is running, how would I do that? Just use my scope? Put a probe on one of the pins and see if anything shows? If I don't know which settings to put the scope on, is it safe to just play around with it a bit? I still have a bit of learning to do with it, but I've used it successfully now a few times. I even used it to help troubleshoot a car not too long ago.

There's more than one x-tal usually, if I remember correctly. I'll check both.

Could you or someone else give me a basic block diagram of a television startup? Something like TV gets power, CPU receives power and waits until the button's pressed. Maybe things will be easier if I fully understand how these things work. I'm going to read Budm's stickied post, I just found that a little bit ago, but it probably won't be for a couple more days. If that info is in there, I'll just read it from there.

[budm]

Re: Element ELEFW40C won't turn on.

http://www.electronicrepairguide.com...r-circuit.html

http://www.embedded.com/design/mcus-...tal-TV-design-

[Spork Schivago]

Re: Element ELEFW40C won't turn on.

There's a chance too that maybe the eeprom is fried. With the DMM (not the logic analyzer), these are the voltages I get, whether I pushed the on button or not.

Pin 1: 0.000VDC CS# -- Chip Select Input
Pin 2: 3.320VDC SO (IO1) -- Data Output (Data Input Output 1)
Pin 3: 3.320VDC WP# (IO2) -- Write Protect Input (Data Input Output 2)
Pin 4: 0.000VDC VSS -- Ground
Pin 5: 0.000VDC SI (IO0) -- Data Input (Data Input Output 0)
Pin 6: 0.000VDC SCLK -- Serial Clock Input
Pin 7: 3.321VDC HOLD# (IO3) -- Hold Input (Data Input Output 3)
Pin 8: 3.321VDC VCC -- Power Supply

So hold is high, write protect is high (unless it's being used as an IO pin), and SO is high. I've been studying the datasheet, earlier in the day, I believe a person sets what mode they want it in (dual, quad, single, etc) during burn time. I believe it's much like a PIC, where there's configuration bits that you set. So I don't really know for certain, for example, which pins are IO, because depending on the configuration bits, they can serve different purposes. Sometimes, ICs fail in such a way that they send voltage out pins that aren't supposed to have voltage. I'm wondering if that's what might be happening.

Do you think that's a possibility or should I be concentrating more on the possible reasons you gave?

Thanks!

[Spork Schivago]

Re: Element ELEFW40C won't turn on.

http://www.electronicrepairguide.com...r-circuit.html

So they're just like my programmable interrupt controllers. I gotcha. And it's the microcontroller that has the instructions that control the television then? The microcontroller is what tells the power supply to start up, once someone hits the power button, it tells the back lighting circuit to kick on the back lights, etc, is that right?

I read that link you sent Budm. That's very informative. I should be checking to see if the TV's microcontroller is getting that RESET signal. From what I just read, if it doesn't get that, the TV won't turn on. I'm starting to think maybe it's a bad microcontroller or a corrupt eeprom. I'm assuming the reason tv's have eeproms, even though the microcontrollers generally have them built in is because the microcontrollers don't have large enough eeproms and the developers need more room than what the microcontrollers provide. Am I right in this?

[budm]

Re: Element ELEFW40C won't turn on.

http://www.embedded.com/design/mcus-...tal-TV-design-

[vinceroger69]

Re: Element ELEFW40C won't turn on.

some very good information in this thread i have been reading the links budm has posted so learning what a eeprom/spi flash ic does also found on shopjimmy what issues they say a replacement eeprom ic may solve

IMPORTANT:
These main boards can have multiple failures, in addition to the EEPROM.
If EEPROM replacement does not repair your board, there are likely additional problems.

Symptoms EEPROM replacement MAY fix:
- On screen display is stuck on logo screen
- TV won’t power on with remote or buttons (LED indicator remains same color)
- TV won’t turn on after being off for 30+ minutes
- TV shuts off by itself during channel scan or when switching inputs
- TV turns on and is stuck in color test pattern mode forever
- TV turns on but there is no backlight (LED indicator shows TV is on but there is no image)

[Spork Schivago]

Re: Element ELEFW40C won't turn on.

some very good information in this thread i have been reading the links budm has posted so learning what a eeprom/spi flash ic does also found on shopjimmy what issues they say a replacement eeprom ic may solve

IMPORTANT:
These main boards can have multiple failures, in addition to the EEPROM.
If EEPROM replacement does not repair your board, there are likely additional problems.

Symptoms EEPROM replacement MAY fix:
- On screen display is stuck on logo screen
- TV won't power on with remote or buttons (LED indicator remains same color)
- TV won't turn on after being off for 30+ minutes
- TV shuts off by itself during channel scan or when switching inputs
- TV turns on and is stuck in color test pattern mode forever
- TV turns on but there is no backlight (LED indicator shows TV is on but there is no image)

In the past, when I had a television to fix, I would just come here for help getting it fixed, but now I want to work on them for a living and I really want to understand how they all work. So when someone says test this diode, instead of just testing it, I want to know why we're testing it and what it's purpose is. I feel I've learned a lot in this one thread, probably more than I've learned in all my other ones.

[Spork Schivago]

Re: Element ELEFW40C won't turn on.

Even with the scope, there's nothing on the SCLK. I even verified the scope was working by using the test probe / calibration part of it. I get a square wave on the test probe / calibration hook but nothing on the SCLK pin. I'll try hooking the scope up to x-tal and see if I get anything.

So how do those crystals work? It's a 24.000MHz crystal. Voltage is supposed to go through on pin (or terminal) of the crystal and then a sine wave is supposed to be produced on the other pin? The material inside resonates, producing the pulses...is there a chance the crystal just isn't activated yet if there's no pulse coming out of it? I guess that wouldn't be hard to test. I can just check to see if voltage is going to one of the pins. I'm going to have to remove the board from the TV because I can only access the pins on the bottom side of the board. This can get tricky, because the cables might not be long enough for me to set the board on it's side vertically or something. I wonder if all the cables are needed for this test. The two that go to the backlighting for instance, I wonder if I can remove them for this test....

[Spork Schivago]

Re: Element ELEFW40C won't turn on.

I hooked up the scope to the crystal. I think I messed up though. I put the probe on one pin and hooked the grounding clamp on the other. When I went to change the VOLTS per division, I'd get shocked.

Maybe I'm supposed to hook the grounding clamp to the board where it'd normally be hooked to the chassis. Without the board being grounded though (I'm holding it with my hand), maybe this is why it's not working? I did not get a sine wave though. I got a really messed up looking screen.

[Spork Schivago]

Re: Element ELEFW40C won't turn on.

I found where the pins for the crystal connect on the top side of the board to the two SMD capacitors and resistor. Maybe I can take my measurement from the oscilloscope somewhere around that area on the top side of the board, so I don't have to hold the board while trying to test it.

**EDIT: I bought a replacement board off e-bay. I talked the guy down from 60.00$ to 40.00$. He accepted my offer. I'm hoping once I get the new board, if it does work, I can use it to figure out what's wrong with the old board, but if not, at least I'll be able to fix the television for the customer. Thanks for all the help.

[Agent24]

Re: Element ELEFW40C won't turn on.

I hooked up the scope to the crystal. I think I messed up though. I put the probe on one pin and hooked the grounding clamp on the other. When I went to change the VOLTS per division, I'd get shocked.

Maybe I'm supposed to hook the grounding clamp to the board where it'd normally be hooked to the chassis. Without the board being grounded though (I'm holding it with my hand), maybe this is why it's not working? I did not get a sine wave though. I got a really messed up looking screen.

In most cases, you put the ground clip on, well, ground.

By connecting the crystal to the ground lead of the scope, you have in fact, connected it to mains earth. This is not what the circuit expects. It's probably why the display was messed up.
It's probably also why you're getting shocked.

You need the isolation transformer, really.

[Spork Schivago]

Re: Element ELEFW40C won't turn on.

I need the isolation transformer so when I make mistakes like this, I don't get hurt? Or I need the isolation transformer to properly test the crystal?

Also, can you recommend a good isolation transformer that wouldn't need modification that I could use with my scope? Thanks!

[Agent24]

Re: Element ELEFW40C won't turn on.

There's a chance too that maybe the eeprom is fried. With the DMM (not the logic analyzer), these are the voltages I get, whether I pushed the on button or not.

If the clock signal is fast enough and only present for a short time, a DMM may not show anything.

Also, I think we need to see if there's any activity from complete power off, not just button press.

I've been studying the datasheet, earlier in the day, I believe a person sets what mode they want it in (dual, quad, single, etc) during burn time. I believe it's much like a PIC, where there's configuration bits that you set. So I don't really know for certain, for example, which pins are IO, because depending on the configuration bits, they can serve different purposes. Sometimes, ICs fail in such a way that they send voltage out pins that aren't supposed to have voltage. I'm wondering if that's what might be happening.

What is the model number of this EEPROM? I haven't seen one before that has settable bits to change the operation behaviour. But I'm sure it's possible. It will still need a clock though, regardless of the mode it's in.

I suppose it could be internally shorted although that kind of failure is more likely with power devices, and would be strange for an EEPROM. I think usually the problem is that some data cells become worn out\stuck and it loses its data in some way or gets corrupted by a 'dirty' power supply. (too much ripple, usually the result of bad capacitors)

[Agent24]

Re: Element ELEFW40C won't turn on.

So they're just like my programmable interrupt controllers. I gotcha. And it's the microcontroller that has the instructions that control the television then? The microcontroller is what tells the power supply to start up, once someone hits the power button, it tells the back lighting circuit to kick on the back lights, etc, is that right?

Yes, the microcontroller will generally send an enable signal to the PSU to take it out of standby, send an enable signal to the inverter\LED driver etc.
If it's not running because it's got problems, or it's halting because it's receiving bad\no data from the EEPROM, it won't get to the point to tell the backlights or panel etc to turn on.

I'm assuming the reason tv's have eeproms, even though the microcontrollers generally have them built in is because the microcontrollers don't have large enough eeproms and the developers need more room than what the microcontrollers provide. Am I right in this?

Yes, size is one reason for using external EEPROMs with microcontrollers. I'm not 100% sure the exact reason in TVs in general though.

[dick_barton]

Re: Element ELEFW40C won't turn on.

If you're measuring a crystal on the main board which is cold side of the circuit, I cannot see how you would receive a shock. Use you multimeter and check what ac or dc voltage you have between those two points.

It is good practice to isolate any set you are working on from the mains using an isolation transformer. Cheaper than a funeral.

You are highly unlikely to measure the oscillation of a crystal by direct connection using your oscilloscope due to the impedance and capacitance the scope probe places on the crystal when you connect to it. This is not always the case but it is generally true for high frequency crystals.