HP w20 LCD Display Inverter, running out of options.

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

Both jumpers on, diplay don't turn off.
- tested 4 hours

One jumper on, one off, display don't turn off.
- tested with either for 4 hours.

After soldering last jumper off (display still hot) and
turning it back on after this, 2 secs to black.

So both sides are required without jumpers for failure

[PlainBill]

Re: HP w20 LCD Display Inverter, running out of options.

Both jumpers on, diplay don't turn off.
- tested 4 hours

One jumper on, one off, display don't turn off.
- tested with either for 4 hours.

After soldering last jumper off (display still hot) and
turning it back on after this, 2 secs to black.

So both sides are required without jumpers for failure

Not a problem. Things are starting to make sense, I think. Earlier you had posted a nice, sharp picture of some 'missing' components. I've marked up an area of the circuit board. I'd like the same quality picture of the area I've outlined in black.

By the way, I really appreciate this. Currently the temperature is 101°F, heading for 110°F. We are under an Ozone alert; it is suggested that anyone with respiratory problems stay indoors. If I wasn't helping you I'd be outside cleaning the pool.

PlainBill

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

Image seems to be missing. I'm glad to help by keeping you inside ;-). I would not consider going out at those temperatures either. Might catch skin cancer in the long run...

[PlainBill]

Re: HP w20 LCD Display Inverter, running out of options.

Image seems to be missing. I'm glad to help by keeping you inside ;-). I would not consider going out at those temperatures either. Might catch skin cancer in the long run...

Arghh!!!

PlainBill

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

Closeup

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

This might help us, I measured all connections with multimeter to see where each connection goes. Blue lines are directly in ground.

[PlainBill]

Re: HP w20 LCD Display Inverter, running out of options.

This might help us, I measured all connections with multimeter to see where each connection goes. Blue lines are directly in ground.

Thanks, that is a big help. Well, part of the mystery has been solved. Indeed, the Vsen voltage comes from the output of the two transformers. This is as the IC design specified. And the return lines from the CCFLs goes to the Isen input. However, it appears that the return signal also changes the voltage on the OVPT pin. This way they sense if a CCFL is drawing a reduced current directly.

In one way, this makes sense. In another way, it is a stupid design change. Instead of having 6 high voltage capacitors at about $.08 each they replace them with more expensive transformers.

ANYWAY, I am going to have to figure out how to figure out what exactly is going on with this circuit and how to locate the offending section. This could be a day or two.

PlainBill

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

Thanks PlainBill,

I studied that reference circuitry of OZ9939. It seems that on this rev 2 board resistor which is identified as R28 near C22 on reference circuitry is missing from ISEN line.

Is it possible that it is not needed at all?

I don't even dare to ask this again... or... I do: CP1 would connect ISEN with kind of "reference look a like" resistor to ground...
Wonder what it might do...
I'm not going to wreck my board trying it anyway so I'll wait your further judgement.

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

My picture was missing couple of wirings that may be hard to detect. He is more detailed one.

[PlainBill]

Re: HP w20 LCD Display Inverter, running out of options.

My picture was missing couple of wirings that may be hard to detect. He is more detailed one.

Oh, Wow!!! Dude. Look at all the pretty colors. Oh, man, teh colors, they're melting!!!

Wow, what a flashback!

Thanks for the updated picture. I've got a pretty good idea of what to look for now. Just trying to decide if I should try to draw up a schematic...

PlainBill

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

Oh, Wow!!! Dude. Look at all the pretty colors. Oh, man, teh colors, they're melting!!!

Wow, what a flashback!
PlainBill

lol

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

Thanks for the updated picture. I've got a pretty good idea of what to look for now. Just trying to decide if I should try to draw up a schematic...

Meanwhile I further investigated how CP1 would affect on ISEN. I have not done these calculations for several years so please correct if I'm wrong here.

ISEN is currently measured as 1.23 V when monitor is on as earlier stated in this thread. I measured that it's 242 ohms from ISEN to ground. That would give us the current of 0.0050826 A from ISEN to ground?

Should I add 4600 ohms (R45) in parallel with above resistance by soldering CP1 in, it would drop total resistance from ISEN to ground to approximately 230 ohms:

1 1 1
-= --- + ----- = 230 ohms
R 242 4600

Should that happen, total current from ISEN to ground would be increased to 0.0053478 Amps. I'd say that is safe to try on DC point of view?

Are there any AC powers to consider too? Or would that capasitor C30 let them pass directly to ground?

[PlainBill]

Re: HP w20 LCD Display Inverter, running out of options.

Meanwhile I further investigated how CP1 would affect on ISEN. I have not done these calculations for several years so please correct if I'm wrong here.

ISEN is currently measured as 1.23 V when monitor is on as earlier stated in this thread. I measured that it's 242 ohms from ISEN to ground. That would give us the current of 0.0050826 A from ISEN to ground?

Should I add 4600 ohms (R45) in parallel with above resistance by soldering CP1 in, it would drop total resistance from ISEN to ground to approximately 230 ohms:

1 1 1
-= --- + ----- = 230 ohms
R 242 4600

Should that happen, total current from ISEN to ground would be increased to 0.0053478 Amps. I'd say that is safe to try on DC point of view?

Are there any AC powers to consider too? Or would that capasitor C30 let them pass directly to ground?

That would be changing the wrong input.

Let's see if I can explain this properly. A CCFL inverter is a constant-current AC source. The usual design takes the sum of the currents through the CCFLs and uses that to adjust the output voltage. In this design, that is the Isen input.

In addition, there are two safety features that may be used. The first is an over-voltage detection. Simply put, the output voltage of the transformers is monitored, and if it exceeds some predetermined value, the inverter is shut down. This is often referred to as OLP (Open Lamp Protection). This design uses the Vsen input to monitor the output of each transformer; it cannot determine if a single CCFL is bad.

The second is a current sense feature. At the end of it's life, the current through the CCFL will increase. Also, if a CCFL is broken, there will be no current through it. This particular design seems to use the current through each CCFL to detect a bad CCFL. And it does this by changing the set point for the Over-voltage protection. As far as I can see, there is no 'end of life' detection for the CCFLs.

PlainBill

[PlainBill]

Re: HP w20 LCD Display Inverter, running out of options.

This is getting curiouser and curiouser. I'm waiting for the flash of inspiration telling me what is actually happening. So far it's been rather dim.

Would you measure the following resistors in circuit, please?

R21, R22, R25, R35

R32, R33, R36, R34

These appear to be across the windings of each transformer. Also, what are the numbers on each transformer?

PlainBill

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

That would be changing the wrong input.

Let's see if I can explain this properly. A CCFL inverter is a constant-current AC source. The usual design takes the sum of the currents through the CCFLs and uses that to adjust the output voltage. In this design, that is the Isen input.

In addition, there are two safety features that may be used. The first is an over-voltage detection. Simply put, the output voltage of the transformers is monitored, and if it exceeds some predetermined value, the inverter is shut down. This is often referred to as OLP (Open Lamp Protection). This design uses the Vsen input to monitor the output of each transformer; it cannot determine if a single CCFL is bad.

The second is a current sense feature. At the end of it's life, the current through the CCFL will increase. Also, if a CCFL is broken, there will be no current through it. This particular design seems to use the current through each CCFL to detect a bad CCFL. And it does this by changing the set point for the Over-voltage protection. As far as I can see, there is no 'end of life' detection for the CCFLs.

PlainBill

Thanks PlainBill.

I think I'm getting some of it, but I can honestly tell that I need to do some serious reading before I really get it.

This is getting curiouser and curiouser. I'm waiting
for the flash of inspiration telling me what is actually happening. So far it's been rather dim.

Would you measure the following resistors in circuit, please?

R21, R22, R25, R35

R32, R33, R36, R34

These appear to be across the windings of each transformer. Also, what are the numbers on each transformer?

PlainBill

R21 91.8 Ω
R22 91.9 Ω
R25 92.5 Ω
R35 92.5 Ω

R32 91.9 Ω
R33 92.5 Ω
R36 92.6 Ω
R34 92.6 Ω

(small differences seem to happen when measuring them several times, less than 1 ohm)

All 4 transformers are similar:
5212-0626-0
LSE M14I S

Curious as I am, also measured some resistances from those return lines to ISEN, VSEN and OVP (attached).

[PlainBill]

Re: HP w20 LCD Display Inverter, running out of options.

OK, I've figured out a little bit of what is happening with the transformers. Some of this is a guess because I don't have the board in front of me, but it makes a good working theory AND it explains some earlier measurements. Since I don't have a good picture of the top of the board, I'm going to be using the resistors attached to the transformers.

One of the CCFLs is attached to R34 and the associated transformer winding. The other side of the winding (and resistor) is grounded. Another is attached to R36 and the associated transformer winding. Again, the other end of the winding is grounded.

What had me baffled was the other windings on the two transformers. The third CCFL is attached to R32 and the associated transformer winding. That means the voltage at the other end of the winding is the sum of the voltage at the left end of R36 + the left end of R32. That is passed to R33, and the associated transformer winding. The result is the voltage at the right end of R32 goes through R51, Q17 (actually a dual diode where the negative transistions are clipped), and on to the Isen input of the inverter controller. That signal ALSO goes to R46 and Q22, which I believe is a true transistor. When the spikes get high enough to turn on Q22, it reduces the Vsens threshold and causes the inverter controller to shut down.

But that's only part of the story. The signal that goes to R51 also goes to D4, and then to R39/C32. I'm not sure what that is for.

And that's still only half of the story. The identical thing is happening with the other three CCFLs, two transformers, etc.

Now to figure out how to block part of the circuit.

PlainBill

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

That signal ALSO goes to R46 and Q22, which I believe is a true transistor. When the spikes get high enough to turn on Q22, it reduces the Vsens threshold and causes the inverter controller to shut down.

Thanks PlainBill.

I assume treshold reducement happens in circled area? Am I even close?

[PlainBill]

Re: HP w20 LCD Display Inverter, running out of options.

Thanks PlainBill.

I assume treshold reducement happens in circled area? Am I even close?

That looks like that would be the location for the other (right hand) side.

As an aside, nearly 30 years ago I interviewed a tech for a job. His knowledge of most areas of electronics was good, but he knew very little about transistor circuits. His explanation: The instructor said transistors were passé, in a few years everything would be done with op-amps and digital ICs. Seems to me the designers are still doing a heck of a lot with transistors.

PlainBill

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

Since I don't have a good picture of the top of the board, I'm going to be using the resistors attached to the transformers.

I took some pictures. There is nothing much at the other side, but I created "see through"....

[CapBlown]

Re: HP w20 LCD Display Inverter, running out of options.

What if it's not overvoltage that causes shutdown? Could it be undervoltage or undercurrent?

I was looking post #69 mine and noticed that AC drops over time. That's also what happened when I adjusted brightness down from 95% and it caused shutdown when brightness reached about 50% level.