Starlogic HD20W

[hoadle]

Re: Starlogic HD20W

I work some old 20 inch Dell & Nec + gateway LCD monitor, inverter board had 3 set quickdisconnect for 6 CCFl backlight, on normal condition 2 CCFL outside on top & bottom will on & one on mid is backup (that mean one CCFL backlight out both mid backlight on top & bottom will turn on)
I think you problem is inverter board had:
1/ bad cap (most time small cap will blow at bottom) replace all cap
2/ make sure check fuse F101 & F102 if one blowed will not turn on top or bottom backlight & will auto shut off
3/ bad transformer take close look see some burned
If you want quick fix google inverter # or LCd monitor model# & parts or ebay or
http://www.shopjimmy.com/
good luck

[retiredcaps]

Re: Starlogic HD20W

[QUOTE=bsweasel;118900throw it in the dumpster.[/QUOTE]

May I suggest you post it on your local kijiji or craigslist. Someone else might be able to use your parts.

[bsweasel]

Re: Starlogic HD20W

I'm about ready to pull all the new lamps, and throw it in the dumpster. Its been weeks since I looked at it and its bumming be out. I'm lost.... logic says its the lamps, but I obviously I'm wrong......

[bsweasel]

Re: Starlogic HD20W

Well.....2nd lamp soldered in and still not lighting up. What to do now......I'm thinking 12 gauge.....shotgun!!!

[PlainBill]

Re: Starlogic HD20W

My soldering is not all the great, that is for sure....but I'm thinking I didn't break it....

You are right...the one transformer drives 2 lamps that have white wires and are in circuit with the D121... Since it has different connectors I wasn't able to easily swap lamps around to see if the voltage changed on that part of the board, but it was similar to the other voltage so I ignored.

So I'm back to my original idea. There are 2 lamp powered by D122 and related inverter. I replaced the one lamp with the blue wire because it appeared to not light so well during the 2seconds to black. Now when it lights it looks MUCH brighter than its brother that is also powered by the same inverter. Swapping these 2 lamps, the one new and the one old, to the other side of the board yeilds similar results when measuring D120. I think I either had 2 bad lamps. (now down to one bad lamp since I replaced one) or both sides of the board have the same bad solder joint problem. But that doesn't seem logical when swapping sides put the voltage readings similar to the 3rd set of lamps. Make sense???

Sort of. But my brain isn't working well today.

PlainBill

[bsweasel]

Re: Starlogic HD20W

There are several possibilities.

1. Either the new lamp is defective, or a mistake was made while installing it.

2. There is a problem on the board. You have three transformers driving 6 lamps. At least one of them must drive lamps at both ends of the board.

I'd rather suggest that there is a bad solder joint on the board than that you somehow failed to connect the new lamp properly. With all the hard work you've put in on this saying "Hey, you must not know how to solder, or are so clumsy you broke the lamp" sounds very harsh. I'd rather reserve my rudeness for people who deserve it.

PlainBill

My soldering is not all the great, that is for sure....but I'm thinking I didn't break it....

You are right...the one transformer drives 2 lamps that have white wires and are in circuit with the D121... Since it has different connectors I wasn't able to easily swap lamps around to see if the voltage changed on that part of the board, but it was similar to the other voltage so I ignored.

So I'm back to my original idea. There are 2 lamp powered by D122 and related inverter. I replaced the one lamp with the blue wire because it appeared to not light so well during the 2seconds to black. Now when it lights it looks MUCH brighter than its brother that is also powered by the same inverter. Swapping these 2 lamps, the one new and the one old, to the other side of the board yeilds similar results when measuring D120. I think I either had 2 bad lamps. (now down to one bad lamp since I replaced one) or both sides of the board have the same bad solder joint problem. But that doesn't seem logical when swapping sides put the voltage readings similar to the 3rd set of lamps. Make sense???

[PlainBill]

Re: Starlogic HD20W

But what is the chance of that open being on both sides of the board? when I move those lamps to the other side of the inverter to the D120 section, the higher voltage problem stays with the lamps. Wouldn't that tell us its both the lamps? Or is it possible we are barking up the entire wrong tree?

There are several possibilities.

1. Either the new lamp is defective, or a mistake was made while installing it.

2. There is a problem on the board. You have three transformers driving 6 lamps. At least one of them must drive lamps at both ends of the board.

I'd rather suggest that there is a bad solder joint on the board than that you somehow failed to connect the new lamp properly. With all the hard work you've put in on this saying "Hey, you must not know how to solder, or are so clumsy you broke the lamp" sounds very harsh. I'd rather reserve my rudeness for people who deserve it.

PlainBill

[bsweasel]

Re: Starlogic HD20W

But what is the chance of that open being on both sides of the board? when I move those lamps to the other side of the inverter to the D120 section, the higher voltage problem stays with the lamps. Wouldn't that tell us its both the lamps? Or is it possible we are barking up the entire wrong tree?

[PlainBill]

Re: Starlogic HD20W

Got my new CCFL today. Replaced the center CCFL, and its the same symtoms....2 seconds to black....and after a few times of off and on with the power supply now its about 1/2 second to black. Maybe both lamps on that section of the inverter?

D122 which is the section where the replacement lamp is, is still measuring 1.3-1.6 volts during the brief flash of the lamps.....

Plainbill, you have any more insights? This is driving me nuts!! If I get this fixed I'm going to throw a celebration party to regrow some of my hair!!!

The only insights I can come up with is an open somewhere between D122 and the point the return line is connected to ground.

PlainBill

[bsweasel]

Re: Starlogic HD20W

Got my new CCFL today. Replaced the center CCFL, and its the same symtoms....2 seconds to black....and after a few times of off and on with the power supply now its about 1/2 second to black. Maybe both lamps on that section of the inverter?

D122 which is the section where the replacement lamp is, is still measuring 1.3-1.6 volts during the brief flash of the lamps.....

Plainbill, you have any more insights? This is driving me nuts!! If I get this fixed I'm going to throw a celebration party to regrow some of my hair!!!

[bsweasel]

Re: Starlogic HD20W

I'm thinking that either the lamp on the pink or blue wire, or both are bad. Since they are on the double connector and this is the one that is giving me troubles. Has anyone seen a 3 lamp setup like this, and can I take out one/two, or do I have to buy all three in a "kit" like this one?

I'm just going to keep answering my own questions. It appears that I will be disassembling this, ordering new ccfl and soldering it to the wires. Upon removing them and inspecting during turn on. The middle ccfl with the blue wire is the one that is not lighting up fully as compared to the other five ccfls.

[bsweasel]

Re: Starlogic HD20W

I'm thinking that either the lamp on the pink or blue wire, or both are bad. Since they are on the double connector and this is the one that is giving me troubles. Has anyone seen a 3 lamp setup like this, and can I take out one/two, or do I have to buy all three in a "kit" like this one?

[bsweasel]

Re: Starlogic HD20W

could I switch the CCFL's from one side to the other and repeat the test on D120 and D122? If its the lamps won't D122 react the same as D120?

EDIT: I did switch CCFL and D122 now is 1.7v.....so I'm looking at bad wires or bad lamps it appears.

[bsweasel]

Re: Starlogic HD20W

T102 has continuity to the top 2 pins of CN103
T101 has continuity to top pin of CN104 and bottom pin of CN102.
T100 has continuity to bottom 2 pins of CN101

So next I guess we test C124 and C146?

C146 I can't see any markings on it......

C124 is:
C
18J
3kv
jnc

This is going to be tricky I'm ill equiped to test these I think ...no CCFL's with the right connectors.....and no ESR meter to test caps....I'm in trouble!!

[PlainBill]

Re: Starlogic HD20W

Sorry for the delay, been busy.... Here are the results.

D122 = 0.6v
D121 = 0.5v
D120 = 1.7v

I used a surge bar on the floor for a foot activated power switch. Put my leads on the pins and turned it on to get these reading. They only read for a second then go to zero. Also during this test I could hear a little buz sound at startup. I also did NOT have a video signal going to the monitor...not sure if that makes a difference?

Good job. Here's a very quick rundown on what is going on. The output voltage from T100 goes through L100 and then to two of the CCFLs. C124 and C146 are in series and form a capacitive voltage divider. While it is physically smaller, C146 has a much greater capacitance than C124. That means the majority of the voltage is dropped across C124. Under normal operation there should be a very small voltage across the two bottom pins of D120. If a lamp is broken, a wire disconnected or broken, etc the voltage across the output of T100 will rise, the voltage across the bottom pins of D120 will rise, shutting down the inverter. This is referred to as either Open Lamp Detection, Over Voltage Protection, or some variation of that. Of course, the same arrangement exists on the other transformers.

Now we have to figure out WHY this is happening. One possibility is an open lamp. Another is the output from T100 is not reaching one (or both) of the lamps connected to it. A third possibility is C124 has increased in value, or is leaky. A fourth is that C146 has decreased in value.

Let's take this a step at a time. Start with the output of T102 (it's the trace covered in white, also the one going to C126). It should go to 1 pin on a connector on each end of the board. Do the same for T101 and T100. If you can establish continuity to two connector pins for each transformer, we will pursue the other possibilities.

PlainBill

[bsweasel]

Re: Starlogic HD20W

Starting at the left end of the inverter, one end of C126 is tied to T102. The other end is tied to D122 and C148. C126 and C148 form a voltage divider. D122 rectifies that; measure the DC voltage between the two 'bottom' legs of D122.

Moving to T101, C125, C147, and D121 perform the same function. This time, measure the voltage between the two legs on the right side of D121.

And last, T100, C124, C146, and the bottom legs of D120.

PlainBill

Sorry for the delay, been busy.... Here are the results.

D122 = 0.6v
D121 = 0.5v
D120 = 1.7v

I used a surge bar on the floor for a foot activated power switch. Put my leads on the pins and turned it on to get these reading. They only read for a second then go to zero. Also during this test I could hear a little buz sound at startup. I also did NOT have a video signal going to the monitor...not sure if that makes a difference?

[PlainBill]

Re: Starlogic HD20W

I don't have any ccfls with 4 wires in a dual plug...it appears that the 2 thicker wires are on one side of the connector and the 2 returns are on the other....is that common to split the hot and returns into a 4 wire connector? I did substitute a ccfl with a single connector 2 wire config into the 2 connects that had such and it didn't make a difference.

Will the transformers give the correct outputs without the CCFL's attached after it goes black? or do I have to power it on while holding probes to the CCFL outputs to read the output voltage.

I personally wouldn't try to measure the voltage out of an inverter transformer with a standard DMM. I know people have tried it, but the voltage is at the limit of or above the maximum voltage specified for most DMMs.

There is a safe way to do this, however. Make these readings while the CCFLS are on.

Starting at the left end of the inverter, one end of C126 is tied to T102. The other end is tied to D122 and C148. C126 and C148 form a voltage divider. D122 rectifies that; measure the DC voltage between the two 'bottom' legs of D122.

Moving to T101, C125, C147, and D121 perform the same function. This time, measure the voltage between the two legs on the right side of D121.

And last, T100, C124, C146, and the bottom legs of D120.

PlainBill

[bsweasel]

Re: Starlogic HD20W

I don't have any ccfls with 4 wires in a dual plug...it appears that the 2 thicker wires are on one side of the connector and the 2 returns are on the other....is that common to split the hot and returns into a 4 wire connector? I did substitute a ccfl with a single connector 2 wire config into the 2 connects that had such and it didn't make a difference.

Will the transformers give the correct outputs without the CCFL's attached after it goes black? or do I have to power it on while holding probes to the CCFL outputs to read the output voltage.

[PlainBill]

Re: Starlogic HD20W

Plainbill, U101 was above....I was right, I knew someone would ask. I just don't know how to test it after reading that datasheet.

This is going to be a real PITA to test using the usual techniques - actually, using ANY techniques. U101 is a general purpose SMPS controller, not a dedicated backlight controller. I'm pretty sure the IC itself is good, it's a question of why it's shutting the backlights off.

Possible causes are:

1. Low or no output from one transformer.

2. One of more CCFL (or the wiring to it) is bad.

3. A component in the sense circuitry is bad.

The logical way to tackle this is to use a known good CCFL as a substitute for each of the ones in the panel, in turn. If the inverter stays on only when the test CCFL is used to replace one particular tube in the monitor, we know the problem is that CCFL or the wiring.

If the test CCFL does not work when it replaces one particular tube, we know the problem is that output.

If nothing can change the 'two seconds to black', we know it's the sense circuitry.

PlainBill

[retiredcaps]

Re: Starlogic HD20W

I just don't know how to test it after reading that datasheet.

I believe the proper way to test is to look at page 2 for the pinouts.

Pin 7 is GND so your black lead/probe goes here. Then your red lead/probe goes to the other pins and you measure the DC voltage and compare them within the specified ranges on page 3-4.

Of course a schematic with the "known good" voltages would be more helpful, but at least you get an idea of the range. For example, if you measure 2V DC and the value on Vref is supposed to be 5V DC, then you can search what is causing this low voltage.