Gateway 2100

[freelander]

Re: Gateway 2100

The CCFLs are fine and I can confirm all other components are fine. I can swap this board with a known functioning board and everything is perfect. Therefore, I know that the problem is on this board but not entirely sure where.

I have included pics of the underside of this board with close ups of the primary and secondary windings as well as the 6 small transformers.

By the way, so that I know what to look for in the future, what should I be looking for in the continuity and diode tests on the transistors?

[PlainBill]

Re: Gateway 2100

The CCFLs are fine and I can confirm all other components are fine. I can swap this board with a known functioning board and everything is perfect. Therefore, I know that the problem is on this board but not entirely sure where.

Thanks for reminding me of that. What can I say, my brain cells are dying. Fortunately there are a lot of younger, smarter people to back me up.

I have included pics of the underside of this board with close ups of the primary and secondary windings as well as the 6 small transformers.

Thanks. I've marked the picture up with a number of things for you to do. All of these can be done with the components on the board. The yellow circles are the outputs of the two transformers. Measure the resistance of each transformer.

The red and black circles are the current sense transformers for each CCFL.
Check the resistance of each transformer primary and secondary (I forgot to mark the secondaries).

Measure the voltage from red to black on each transformer while the backlights are on. (The terminals circled in black seem to be connected together, so if you prefer, choose just one of them.)

If you are having problems getting readings in the brief time the backlights are on, give me the part number of the IC circled in dark blue. This is the inverter controller and it may be possible to fool it to believe everything is good long enough to take your readings.

By the way, so that I know what to look for in the future, what should I be looking for in the continuity and diode tests on the transistors?

I was operating on the assumption that they are bipolar transistors. This may not be true, more on that later.

A bipolar transistor can be modeled as two diodes with either the anode or cathode connected together. The resistance function of most DMMs are designed to use a very low voltage so as to not cause any diodes to conduct. By measuring the resistance between each terminal of a transistor you can determine if it is shorted. Typically a short will be under 10 ohms, a 'good' transistor should show over 100 ohms. (There are exceptions when making measurements in circuit, chiefly if the transistor is directly coupled to a transformer. This circuit uses 3 capacitors so that is not an issue.)

By switching to a diode test function and measuring the voltage drop across the leads of a transistor in all directions b-c, c-b, b-e, e-b, c-e, e-c you should get a voltage approximating a diode drop for only two combinations. Again, this is not a perfect test, but the combination of the two tests should identify any shorted transistors and any transistor that has an open junction.

After some reflection, I realized I am not sure how the transistors are connected, nor if they really are bipolar transistors. To that end, I have circled them in baby blue. What are the part numbers?

PlainBill

[freelander]

Re: Gateway 2100

Please be patient with me...how am I measuring the resistance of the transformer with the yellow circles? I am assuming that I am measuring each pair of yellow circles. Doing that, I get 156 Ohm for the top pair and 156 Ohm for the bottom pair.

I'm not sure what you mean by measure the primary and secondary of each transformer?!?

I will try the current sense transformers soon. In the meantime, the transistor codes are:
U301/U304 = D405 BE5211
U302/U305 = D408 BE6A1D
U303 = B113193G 1MP7 G0542

[freelander]

Re: Gateway 2100

ok...I am using DC which I expect is right measuring with my leads matching your colors.

T401 ~ 0.25VDC before shutting off
T402 ~ 0.25VDC before shutting off
T403 ~ 0.25VDC before shutting off
T405 --> this kind of does nothing and then makes my DMM go crazy...what does this mean?
T406 --> similar to T405
T404 --> similar to T405

I am assuming that this means that something is up with the transformer at the bottom of the picture that feeds this side?!?

[PlainBill]

Re: Gateway 2100

Please be patient with me...how am I measuring the resistance of the transformer with the yellow circles? I am assuming that I am measuring each pair of yellow circles. Doing that, I get 156 Ohm for the top pair and 156 Ohm for the bottom pair.

I'm not sure what you mean by measure the primary and secondary of each transformer?!?

I will try the current sense transformers soon. In the meantime, the transistor codes are:
U301/U304 = D405 BE5211
U302/U305 = D408 BE6A1D
U303 = B113193G 1MP7 G0542

The readings match for both. That's a good sign - sorta. At least the secondaries read the same, which eliminates an open secondary winding.

I'll leave the transistors for later, but the next one is a NASTY!!! The actual part number is BIT3193G. Pins 14, 15 and 5 are the pins of interest. If pin 14 goes over 2.0Vdc an error signal is set. If pin 15 is below 1.3Vdc an error signal is set. If pin 5 is greater than 2.5 volts the outputs are shut down until the chip is powered off, then back on.

PlainBill

[PlainBill]

Re: Gateway 2100

ok...I am using DC which I expect is right measuring with my leads matching your colors.

T401 ~ 0.25VDC before shutting off
T402 ~ 0.25VDC before shutting off
T403 ~ 0.25VDC before shutting off
T405 --> this kind of does nothing and then makes my DMM go crazy...what does this mean?
T406 --> similar to T405
T404 --> similar to T405

I am assuming that this means that something is up with the transformer at the bottom of the picture that feeds this side?!?

Sorry, I wasn't explicit. I want this measurement done using the AC ranges. However, this has me very puzzled. Obviously, something different is happening on the two halves. Your idea has merit, but let's make sure we know which side is bad.

I've circled a point above the upper transformer in yellow, and a spot just below it in white. Another spot below the lower transformer has also been circled in yellow. Using the AC volts scale, measure the voltage from the upper yellow spot to the white spot while the backlights are working. Repeat for the lower yellow spot to the white spot.

More later; I'm also playing plumber today AND fighting with both the post office and a doctor's office. It would be nice to succeed at least one of the above.

PlainBill

[freelander]

Re: Gateway 2100

The readings match for both. That's a good sign - sorta. At least the secondaries read the same, which eliminates an open secondary winding.

I'll leave the transistors for later, but the next one is a NASTY!!! The actual part number is BIT3193G. Pins 14, 15 and 5 are the pins of interest. If pin 14 goes over 2.0Vdc an error signal is set. If pin 15 is below 1.3Vdc an error signal is set. If pin 5 is greater than 2.5 volts the outputs are shut down until the chip is powered off, then back on.

PlainBill

Pin 14 gets to ~0.25VDC before the backlight goes off
Pin 15 gets to ~0.056VDC before the backlight is off ******
Pin 5 gets to ~1.5VDC until the backlight shuts off but then climbs to 3.89VDC...it drops back to zero when the panel switches in to standby.

I am a chemical engineer that dabbles in electrical projects so bear with me if I get some of the fundamentals backwards. I read that the pin out is labelled from the dent and around but I have included a picture so that you know for sure which pins I tested.

[freelander]

Re: Gateway 2100

Sorry, I wasn't explicit. I want this measurement done using the AC ranges. However, this has me very puzzled. Obviously, something different is happening on the two halves. Your idea has merit, but let's make sure we know which side is bad.

I've circled a point above the upper transformer in yellow, and a spot just below it in white. Another spot below the lower transformer has also been circled in yellow. Using the AC volts scale, measure the voltage from the upper yellow spot to the white spot while the backlights are working. Repeat for the lower yellow spot to the white spot.

More later; I'm also playing plumber today AND fighting with both the post office and a doctor's office. It would be nice to succeed at least one of the above.

PlainBill

For the transformers, with my common lead on yellow and my other lead on white with the DMM on AC, I get a reaction but the voltage is still 0.00xx VAC as it moves so I'm not sure what that tells us.

[freelander]

Re: Gateway 2100

ok...I am using DC which I expect is right measuring with my leads matching your colors.

T401 ~ 0.25VDC before shutting off
T402 ~ 0.25VDC before shutting off
T403 ~ 0.25VDC before shutting off
T405 --> this kind of does nothing and then makes my DMM go crazy...what does this mean?
T406 --> similar to T405
T404 --> similar to T405

I am assuming that this means that something is up with the transformer at the bottom of the picture that feeds this side?!?

I have re-done these measurements with AC scale again with my common lead on black.

For all of these, it messes up my display on the DMM until the backlight shuts off and then scales down. I repeated multiple times for each transformer and sometimes the scale would hit 2.5VAC, sometimes OL but most often jibberish and other random stuff like changing units etc.

[PlainBill]

Re: Gateway 2100

Your numbering is correct. The dot is pin 1, then counterclockwise around the chip. Next, I'm guessing here, but it may be that your DMM is slow to respond to changing signals. When the BIT3193G starts up the voltage at pin 5 rises slowly. The protection logic isn't enabled until the voltage at pin 5 gets over 2.5 volts.

The fact that the voltage at pin 15 is so low indicates the controller is shutting down because the current isn't going high enough. I'll try to figure out how to determine which set of outputs is causing the problem.

By any chance do you have a spare CCFL around (even one in a panel) that you can hook up to this controller. It would appear that one transformer may not be outputting anything, but I would like to verify that; the voltage readings aren't telling me.

PlainBill

[freelander]

Re: Gateway 2100

Yep, I have a spare CCFL and I have complete units that are functioning that I can compare with if necessary. My son has an analog tester from an electronics kit that I will try to dig up for better response. I'll let you know any change shortly.

[freelander]

Re: Gateway 2100

Using an analog meter, the yellow/white dot test on the transformer with the meter set to AC has no reaction in either direction.

For the current sense transformers, I get the following with my analog meter:
T401 = 100VAC
T402 = 220VAC
T403 = 125VAC
T405 = 125VAC
T406 = 125VAC
T404 = 150VAC

For the IC, I can't get the analog meter to show any DC voltage for pin 14, 15, 5.

[PlainBill]

Re: Gateway 2100

Using an analog meter, the yellow/white dot test on the transformer with the meter set to AC has no reaction in either direction.

For the current sense transformers, I get the following with my analog meter:
T401 = 100VAC
T402 = 220VAC
T403 = 125VAC
T405 = 125VAC
T406 = 125VAC
T404 = 150VAC

For the IC, I can't get the analog meter to show any DC voltage for pin 14, 15, 5.

That reading for T402 looks VERY high - nearly twice the others. Try substituting a known good CCFL for the one hooked to that transformer.

PlainBill

[freelander]

Re: Gateway 2100

I hooked up a good power supply and tested the current sense transformers with the monitor and functioning normally.

My DMM still goes crazy if I try to measure these values which doesn't make sense. However, the analog meter gives somewhat strange results. I get one transformer that is pegged at 250VAC, one at ~125VAC, and another between 150-175VAC. I get a similar set of readings for both sets of three transformers to drive the lights on the top or the bottom.

On the higher voltage (>250VAC) transformers, they turn off the display when I complete the test circuit while it is live but the others do not. Does that make sense?

My spare CCFL seems suspect right now since only two of three bulbs lights in either set of connections with the 'good' power supply.

[PlainBill]

Re: Gateway 2100

I hooked up a good power supply and tested the current sense transformers with the monitor and functioning normally.

My DMM still goes crazy if I try to measure these values which doesn't make sense. However, the analog meter gives somewhat strange results. I get one transformer that is pegged at 250VAC, one at ~125VAC, and another between 150-175VAC. I get a similar set of readings for both sets of three transformers to drive the lights on the top or the bottom.

On the higher voltage (>250VAC) transformers, they turn off the display when I complete the test circuit while it is live but the others do not. Does that make sense?

My spare CCFL seems suspect right now since only two of three bulbs lights in either set of connections with the 'good' power supply.

"Good power supply"? Do you have bad caps on this?

PlainBill

[freelander]

Re: Gateway 2100

I'm using quotes because I am confused about the readings. The good power supply has had all of the 470uF caps replaced but not the large ones and not the smallest ones. This board works great with multiple LCD panels so I trust that it is good.

[freelander]

Re: Gateway 2100

I have a common thread with three failed power boards for this monitor that I am not sure if it means something or not. The diodes look like little resistors and don't really look like solid state components. I have included a picture of both types for contrast. The ones that are black and look like solid state are on one of my functioning boards.

There is a follow up question about diodes like this. Can I test them in circuit and should that be with power on or off? If so, what should I be looking for to determine if it is good or bad?

[PlainBill]

Re: Gateway 2100

I have a common thread with three failed power boards for this monitor that I am not sure if it means something or not. The diodes look like little resistors and don't really look like solid state components. I have included a picture of both types for contrast. The ones that are black and look like solid state are on one of my functioning boards.

There is a follow up question about diodes like this. Can I test them in circuit and should that be with power on or off? If so, what should I be looking for to determine if it is good or bad?

You can do a rudimentary test in circuit. A more extensive test would require removing it from the circuit. A complete characterization would require expensive equipment. For a start, we will assume they are equivalent parts in different packages. This is quite common.

All tests will be done with power off.

Test 1. Set the DMM to the lowest (200 ohm) scale and measure the resistance across the diode. It should read more than 50 ohms; if it reads open (more than 200 ohms), that is fine. If it reads less than 10 ohms it is probably shorted. If it reads less than 1 ohm,it is definitely shorted.

Test 2. Set the DMM to the diode test range. Touch the leads to the ends of the diode. Note the reading. Reverse the leads and repeat. Note THIS reading. You should read about .7 volts in one direction and over range in the other direction.

PlainBill

[freelander]

Re: Gateway 2100

I tested all of the straight diodes on 4 boards and used the same lead orientation for each test. I have listed the results for the good board and 4 bad ones (note that two still need to have capacitors changed since I haven't changed them although they look ok).

[freelander]

Re: Gateway 2100

Any thoughts or direction on what I should check next to isolate the problem?