Soyo MT-NI-DYLM1996 - Two Seconds to Black

[browork]

I have a Soyo Model MT-NI-DYLM1996 with the common Two Seconds to Black issue.

It has been said in the past by PB "The problem you describe is known as the 'two seconds to black' problem, and can have many causes. Bad caps in the power supply are one of the possible causes..., as are bad CCFLs, bad wiring to the CCFLs, a shorted or open inverter transformer, a bad driver to an inverter transformer, or a bad component in the circuitry that monitors the CCFL status."

All the caps on the power supply look good. They are made by fuhjyyu.

The CCFLs and wiring is good, as I can plug them into a different monitor (inverter), and they light up and stay lit.

The ccfl transformers read open. This may be the cause, but the screen does light up for a few seconds before going to black, so I am not sure here.

As for the circuitry I know not what to check.

On the logic board...

Voltages are good going to the logic board. 5V where there should be 5V and 12V where there should be 12V. See the pictures.

Also both of the voltage regulators read 3.3V which is what they are labeled.

Any ideas where to go from here?

[PlainBill]

Re: Soyo MT-NI-DYLM1996 - Two Seconds to Black

I have a Soyo Model MT-NI-DYLM1996 with the common Two Seconds to Black issue.

It has been said in the past by PB "The problem you describe is known as the 'two seconds to black' problem, and can have many causes. Bad caps in the power supply are one of the possible causes..., as are bad CCFLs, bad wiring to the CCFLs, a shorted or open inverter transformer, a bad driver to an inverter transformer, or a bad component in the circuitry that monitors the CCFL status."

All the caps on the power supply look good. They are made by fuhjyyu.

The CCFLs and wiring is good, as I can plug them into a different monitor (inverter), and they light up and stay lit.

The ccfl transformers read open. This may be the cause, but the screen does light up for a few seconds before going to black, so I am not sure here.

As for the circuitry I know not what to check.

On the logic board...

Voltages are good going to the logic board. 5V where there should be 5V and 12V where there should be 12V. See the pictures.

Also both of the voltage regulators read 3.3V which is what they are labeled.

Any ideas where to go from here?

Toasty has a slightly different name for 'fuhjyyu'. It's not one used in polite company. VERY bad reputation (the caps, not Toasty).

There are several different approaches.

The easiest is if I tell you to replace the caps, then let me know if that doesn't work.

The best is to hook up a CCFL from the other monitor to replace one in this monitor, see if the other monitor lights up for two seconds. If it does, move it to the next connector on the Soyo inverter, until you have verified all outputs are working. If this doesn't work, take a picture of the back of the Soyo inverter from the reverse angle - with the inverter components closest to the camera.

PlainBill

[browork]

Re: Soyo MT-NI-DYLM1996 - Two Seconds to Black

Alright, caps replaced and still the same issue.

As I stated above, I plugged the CCFLs into another monitor and they light up and stay lit. So I do not think it is the CCFLs.

Your suggestion of taking testing different CCFLs on this board, caused the same issue, two seconds lit then dark.

Attached the another (flipped) pic of the power/inverter board.

BTW your Twins just beat my beloved Royals...

[PlainBill]

Re: Soyo MT-NI-DYLM1996 - Two Seconds to Black

Alright, caps replaced and still the same issue.

As I stated above, I plugged the CCFLs into another monitor and they light up and stay lit. So I do not think it is the CCFLs.

Your suggestion of taking testing different CCFLs on this board, caused the same issue, two seconds lit then dark.

Attached the another (flipped) pic of the power/inverter board.

BTW your Twins just beat my beloved Royals...

I may not have communicated clearly. Or maybe you figured it out. When you have 4 CCFLs in a monitor it's pretty hard to figure out if one of them is not lit. By hooking up 1 CCFL from another monitor and 'walking it' through all four connectors you can check if all inverter outputs are 'live'.

In any case, I've marked 9 points on the back of the inverter. Gnd is at the left end. V1 thru V4 are used of monitoring the output voltage to each CCFL. R1 thru R4 are the return lined from the CCFLs. In each case, the voltage should be less than 10 volts.

Your mission, should you decide to accept it, is to measure the voltage on each of the 8 points WHILE THE CCFLS are ON. All precautions related to working on high voltage circuits apply here. I usually recommend having one person work with the meter while someone else (preferably someone who does not have a financial interest in your demise) 'cycles' the power button on the monitor. We are interested in one point which is significantly different from the three other related points.

Yes, the Twins have been doing that consistently to some pretty good teams.

PlainBill

[EGuevarae]

Re: Soyo MT-NI-DYLM1996 - Two Seconds to Black

Alright, caps replaced and still the same issue.

Replaced with what?

[browork]

Re: Soyo MT-NI-DYLM1996 - Two Seconds to Black

All the caps were replaced with Panasonic, except the 220uF 25V which I didn't have in Panasonic so I had to use a Xicon 220uF 35V. Or rather I didn't have 220uF with at least 25V in Panasonic. I have some in Panasonic 220uF 6.3V I have been using repairing some HP motherboards.

[browork]

Re: Soyo MT-NI-DYLM1996 - Two Seconds to Black

In any case, I've marked 9 points on the back of the inverter. Gnd is at the left end. V1 thru V4 are used of monitoring the output voltage to each CCFL. R1 thru R4 are the return lined from the CCFLs. In each case, the voltage should be less than 10 volts.

Your mission, should you decide to accept it, is to measure the voltage on each of the 8 points WHILE THE CCFLS are ON. All precautions related to working on high voltage circuits apply here. I usually recommend having one person work with the meter while someone else (preferably someone who does not have a financial interest in your demise) 'cycles' the power button on the monitor. We are interested in one point which is significantly different from the three other related points.

Voltage at all eight points...

V1-V4 all over 600V (my multimeter maxs at 600V) see picture.
R1 - 19V
R2 - 17V
R3 - 14V
R4 - 14V

When measuring the R readings that was the peek voltage at the time the light shut off. It may have been going higher. Either way this does not bode well I am thinking....

[PlainBill]

Re: Soyo MT-NI-DYLM1996 - Two Seconds to Black

Voltage at all eight points...

V1-V4 all over 600V (my multimeter maxs at 600V) see picture.
R1 - 19V
R2 - 17V
R3 - 14V
R4 - 14V

When measuring the R readings that was the peek voltage at the time the light shut off. It may have been going higher. Either way this does not bode well I am thinking....

Something is very strange on those readings. You DID use the DC volts setting, didn't you? In particular, the V readings should not have been that high - this design uses a capacitive voltage divider on the voltage sense lines.

I've circled 6 jumpers that provide ground to the monitoring circuitry. Verify those are soldered properly. Also, could you provide a higher resolution picture of the area outlined in blue, specifically the components between the two rows of transformer pins and below the inverter controller. By the way, what is the part number for U2?

PlainBill

[browork]

Re: Soyo MT-NI-DYLM1996 - Two Seconds to Black

I thought I used DC voltage, I meant to use DC voltage, however upon retesting I guess I did not... The switch between AC and DC voltage on my multimeter is the most hated thing I have that is electronic.

All the R Voltages reads between 1.4V and 1.9V.
All the V Voltages still read over 600V.

All the circled solder joints look fine.

Attached is a closer picture of that area. Natural lighting used since you can't really get a good flash picture that close with so many reflective objects.

[PlainBill]

Re: Soyo MT-NI-DYLM1996 - Two Seconds to Black

I thought I used DC voltage, I meant to use DC voltage, however upon retesting I guess I did not... The switch between AC and DC voltage on my multimeter is the most hated thing I have that is electronic.

All the R Voltages reads between 1.4V and 1.9V.
All the V Voltages still read over 600V.

All the circled solder joints look fine.

Attached is a closer picture of that area. Natural lighting used since you can't really get a good flash picture that close with so many reflective objects.

The picture is excellent; I'm not having any success following the traces. I've attached a picture of an inverter. Note that this is SIMILAR, but not identical to your inverter. One big difference is that this schematic has one transformer. Your inverter has two transformers, each with two secondaries.

The design in the schematic uses the same circuit for monitoring the 'health' of the backlight system. It monitors the output voltage of the transformer (V) and the current on the return line from the CCFL (R). The major difference is this has only one of each circuit; your inverter has 4 of each.

C8 / C13 form a capacitive voltage divider. Keeping units the same, C8 is 22pf, C13 is 33000pf. Therefor the voltage across C13 should be 1/1500 the voltage across C8. Again, the component values (and the voltage ratio) may be different on your inverter.

CR2 is a dual diode, the common terminal is connected to the junction of C8 / C13. When the waveform at the common terminal goes negative, it is clamped by the diode connected to ground. When it goes positive it charges C16.

Cr1, R16, R15, R13 and C15 perform a similar function, measuring the current on the return line.

I've marked up the section of the inverter. The dual diodes associated with the voltage sense are circled in yellow; those associated with current sense in orange. Note that all the diodes circled in yellow have two leads marked with a red dot. Again, with the CCFLs on, what are the voltages on those pins? (One pin of each pair seems to be connected to ground.)

PlainBill

[browork]

Re: Soyo MT-NI-DYLM1996 - Two Seconds to Black

The voltage readings you asked for. And yes one pin on each leads to ground if you were asking that.

From left to right...

Labeled D12 1.33V (Yellow)
Labeled D21 2.10V (Orange)
Labeled D11 1.34V (Yellow)
Labeled D23 0.75V (Orange)
Labeled D19 1.33V (Yellow)

The other voltages on the pins you didn't label with a red dot...

Labeled D13 0.84V (Orange)
Labeled D20 1.34V (Yellow)
Labeled D22 0.68V (Orange)

[PlainBill]

Re: Soyo MT-NI-DYLM1996 - Two Seconds to Black

The voltage readings you asked for. And yes one pin on each leads to ground if you were asking that.

From left to right...

Labeled D12 1.33V (Yellow)
Labeled D21 2.10V (Orange)
Labeled D11 1.34V (Yellow)
Labeled D23 0.75V (Orange)
Labeled D19 1.33V (Yellow)

The other voltages on the pins you didn't label with a red dot...

Labeled D13 0.84V (Orange)
Labeled D20 1.34V (Yellow)
Labeled D22 0.68V (Orange)

Excellent!!! As an exercise for the student, did you notice anything unusual?

PlainBill

[browork]

Re: Soyo MT-NI-DYLM1996 - Two Seconds to Black

D21 does not seem right. I checked it twice.

My first inclination is that it should be close to D23 reading, but I may be wrong.

Beyond that I am not sure.

[PlainBill]

Re: Soyo MT-NI-DYLM1996 - Two Seconds to Black

D21 does not seem right. I checked it twice.

My first inclination is that it should be close to D23 reading, but I may be wrong.

Beyond that I am not sure.

Well done. If you group the orange readings you notice that three of them vary from 0.68 to 0.84. The other reads nearly three times that.

Now the next step is to determine if this is the CCFL (or it's wiring), or the monitoring circuitry. To do that, reverse the connections for the two CCFLs at the R1,R2 end. IOW, unplug the CCFLs plugged into left end. Plug the one that was in the 'upper' connector into the lower; plug the one that was in the 'lower' connector into the upper. The repeat the measurements on D21 and D13. If the high reading stays on the same diode, the problem is on the board. If the reading moves to the other diode, the problem is the CCFL or it's wiring.

PlainBill

[browork]

Re: Soyo MT-NI-DYLM1996 - Two Seconds to Black

Readings with the CCFLs switched...

D12 1.22V
D13 1.54V
D21 1.76V
D11 1.23V
D20 1.23V
D23 0.76V
D22 1.55V
D19 1.22V

Just to confirm I switched the CCFL wires back and got almost the same readings listed in my previous post...

D12 1.33V
D13 0.74V
D21 2.34V
D11 1.35V
D20 1.35V
D23 0.74V
D22 0.56V
D19 1.33V

It looks like it moved, but others changed too so I am unsure... Even though these CCFLs can be plugged into another monitor and stay lit for 15-20 minutes, they still may be bad???

All readings were double checked as well as the DC setting, and confirmed by my helper who probably does have a financial gain with my demise but she loves me too much to try anything of the sort.

[PlainBill]

Re: Soyo MT-NI-DYLM1996 - Two Seconds to Black

Readings with the CCFLs switched...

D12 1.22V
D13 1.54V
D21 1.76V
D11 1.23V
D20 1.23V
D23 0.76V
D22 1.55V
D19 1.22V

Just to confirm I switched the CCFL wires back and got almost the same readings listed in my previous post...

D12 1.33V
D13 0.74V
D21 2.34V
D11 1.35V
D20 1.35V
D23 0.74V
D22 0.56V
D19 1.33V

It looks like it moved, but others changed too so I am unsure... Even though these CCFLs can be plugged into another monitor and stay lit for 15-20 minutes, they still may be bad???

All readings were double checked as well as the DC setting, and confirmed by my helper who probably does have a financial gain with my demise but she loves me too much to try anything of the sort.

I 'rationalized' this to better see what is happening.

`First Swap. Back

Voltage

D11 1.34V 1.23V 1.35V
D12 1.33V 1.22V 1.33V
D19 1.33V 1.22V 1.33V
D20 1.34V 1.23V 1.35V

Current

D13 0.84V 1.54V 0.74V
D21 2.10V 1.76V 2.34V
D22 0.68V 1.55V 0.56V
D23 0.75V 0.76V 0.74V

Swapping the leads changed both the voltage and current EXCEPT for D23. The only idea that comes to mind is a leaky diode. This next series of tests will be tedious, but safe. Unplug the power cord, and give it a little time for all caps to discharge. Here is the datasheet for the diodes. Note how the pins are numbered.

First step: With the DMM on it's resistance range, measure the resistance between all three pins of the diode in both directions. This is a total of 6 readings per diode. R=Red; B=Black Do this for all 10 diodes - include the two I didn't circle. Many of the readings will show open, even on the highest range. Indicate those by a -, please don't say 'none'. If you find one of them shorted you have found the problem.

R-1 B-2
R-1 B-3
R-2 B-1
R-2 B-3
R-3 B-1
R-3 B-2

Now switch to the diode test scale and repeat the tests. In this case, note the voltage drop across each junction. Again, we are looking for a reading which is significantly different than the other diodes.

Note that there are three groups of diodes. I would expect similar readings from all diodes in a group.
Group 1: D13, D21, D23, D22
Group 2: D12, D20, D11, D19
Group 3: D17, D25

PlainBill