Viewsonic VA2431wm bad cap replaces, new problem

[Tom-Servo]

Hello. I inspected a power board for a VA2431wm that blue led was constant but wouldn't turn on. On further inspection the cap near PT801 was bulging, so I replaced that cap with a NICHICON - HD Series UHD1E102MHD and 2 other caps that were also 1000 uF 25 Volts.

When I put the monitor back together and plugged it in the blue turned on, but when I pushed the power button the blue led would turn off and on every 2 seconds or so. I removed the board and tried pluggin it in again and I was able to hear a short whistle sound every time the LED went out.

I tested the transistor nearest to the bad cap, PT801. (part number 80GL24T-23-DN?, attached closeup picture with ohm readings) The side with only 2 leads had a resistance near 500 Ohm, while the other side corresponding leads were close to zero. The other Transistor, T901 (part number 80GL22T-3-H?, still in circuit, no close up) had all readings of close to 0.

I used the service manual pdf acer_x233h_lcd_monitor.pdf for the power board reference. It appears the same board (715G2824-4-5) is used in both monitors. Here's the endpoints for all the windings:

PT801
1 and 2 winds to 5 and 6
7 winds to 8 (Only one with resistance, 500 Ohm)

T901
1 winds to 3
4 winds to 6 (5 is pictured but missing?)
7 and 8 winds to 9 (bridged to 10)
10 winds to 11 and 12

Thinking about what I have read so far I'd think there is an isue with Transistor T901, but I'm not confident on my understanding of testing transistors yet. I would think PT801's Secondary windings would be 1,2 to 5,6, but that would mean they should have the 500 Ohm resistance, not 7 and 8.

Please let me know if I'm on the right track or not, and if I am how do I find all the compatible replacement transistors. Thanks!

[budm]

Re: Viewsonic VA2431wm bad cap replaces, new problem

"Thinking about what I have read so far I'd think there is an isue with Transistor T901," You are talking about TRANSFORMER NOT TRANSISTOR, right?
Pictures of the bottom side of the board?
The board looks over heated, you may have bad power MOSFETS which drive the inverter transformer.
PT801 is the inverter transformer.
T901 is the power transformer.

"When I put the monitor back together and plugged it in the blue turned on, but when I pushed the power button the blue led would turn off and on every 2 seconds or so. I removed the board and tried pluggin it in again and I was able to hear a short whistle sound every time the LED went out."
If the power transformer is the problem then you will not have your power LED.
The power supply section (T901 is the power supply transformer) is an ALWAYS ON power supply, meaning it runs the minute you plug the monitor into the outlets.
When you hit the power switch, the main board will send the command ON signal to turn on the inverter circuit, this circuit uses the most power so if it is bad it will draw lots of current which will send the power supply into shutdown-restart cycle so you will the see the power LED goes on and off.

[Tom-Servo]

Re: Viewsonic VA2431wm bad cap replaces, new problem

Oh yeah I mean transformers, whoops.

Attached is a picture of the bottom of the board. It does appear that at least one of the ICs near the inverter transformer look like they have blots.

[Agent24]

Re: Viewsonic VA2431wm bad cap replaces, new problem

They look like ICs in the 8-Pin package but actually they are MOSFETs. I think Q805 and Q806?
It's hard to tell from the photo if they are blown or just have flux residue on them from the factory. Can you get a better photo?

You will need to test them with your multimeter for shorts.

Can you upload that PDF you talked about earlier so we can see the PSU schematic?

[fzabkar]

Re: Viewsonic VA2431wm bad cap replaces, new problem

Can you upload that PDF you talked about earlier so we can see the PSU schematic?

I haven't checked them, but these look like they might be the ones:

[Tom-Servo]

Re: Viewsonic VA2431wm bad cap replaces, new problem

They look like ICs in the 8-Pin package but actually they are MOSFETs. I think Q805 and Q806?
It's hard to tell from the photo if they are blown or just have flux residue on them from the factory. Can you get a better photo?

You will need to test them with your multimeter for shorts.

Can you upload that PDF you talked about earlier so we can see the PSU schematic?

Oh yeah I just finished testing the MOSFETs and I was typing up the following reply:

Ok I did some readings with my multimeter on Q805 and Q806 (AO4620). Both chips are wired more or less the same with pin 5-8 bridged. Between Q806 Pin 5, 1, 3 and 4 show no resistance, while Q805 does not. So it appears Q806 is toast (and it looks like it is toast from it's appearance too.)

Is there a substitute for AO4620 that mouser or some other US vendor stocks? I did see SI4542DY listed as a replacement but I'm not sure how suitable it is.

Here's the best picture I can get of the MOSFETs now. I can get better shots tomorrow when it's daylight if needed. Also attached is the pdf with the schematic for the power board, starting at page 46:

[Agent24]

Re: Viewsonic VA2431wm bad cap replaces, new problem

Don't worry about the photo, if the measurements differ then likely it's blown. Replace both at once though because even if only one has blown the other was probably overstressed during the event. Also, if you use a different model or manufacturer for the replacement you should not mix them.

The SI4542DY is very similar and would likely work, but it is a bit slower going by the datasheet. I don't know if that would be a problem in an inverter driver or not. If you want to be sure, get one the same or faster. Where did you find the SI4542DY listed as a replacement?

[Tom-Servo]

Re: Viewsonic VA2431wm bad cap replaces, new problem

Don't worry about the photo, if the measurements differ then likely it's blown. Replace both at once though because even if only one has blown the other was probably overstressed during the event. Also, if you use a different model or manufacturer for the replacement you should not mix them.

The SI4542DY is very similar and would likely work, but it is a bit slower going by the datasheet. I don't know if that would be a problem in an inverter driver or not. If you want to be sure, get one the same or faster. Where did you find the SI4542DY listed as a replacement?

The page that lists SI4542DY and SI4542DY-T1-E3 as a "Replacements & Cross References" is:

http://parts.arrow.com/item/detail/a...nductor/ao4620

I then tried searching for it at:

http://www.mouser.com/Semiconductors...4542DY&FS=True

Which attributes listed on that page are important when it comes to matching the speed and compatibility of the AO4620? I wouldn't mind getting something slightly better if that's all that's readily available.

[Tom-Servo]

Re: Viewsonic VA2431wm bad cap replaces, new problem

Took too much time to edit my last post so posting another reply.

Using the following filters on mouser:


Category Semiconductors > Discrete Semiconductors > Transistors > MOSFET

Pd - Power Dissipation 2 W
Mounting Style SMD/SMT
Package/Case SOIC-8
Transistor Polarity N and P-Channel | N-Channel, P-Channel
Id - Continuous Drain Current 7A, - 5A

I come up with FDS8958A_F085:

Is that any better. Is there anything compatible enough or better to use, or is it best to stick with AO4620?

[Tom-Servo]

Re: Viewsonic VA2431wm bad cap replaces, new problem

It seems that I made a mistake. Before I posted here I tried plugging in the power board with the inverter transformer removed to try to see if the blinking blue LED would stop. When I plugged it in the blue led came on for a second and then turned off. When I resoldered the inverter transformer back on the blue led still doesn't come back on. When I measure ground to +5 volts on the cable that goes to the control board I get 2.8 volts or so.

So is it likely I damaged the power transformer? How can I check it out?

Pretty embarrassing mistake. Bad things happen when I'm in a hurry.

[Agent24]

Re: Viewsonic VA2431wm bad cap replaces, new problem

It's very unlikely you blew the power transformer, you would have to do something really stupid like put the primary winding directly across the mains without a fuse, or throw it at the wall or something.

Sounds to me like something else has shorted since the inverter transformer was not installed (likely the other MOSFET) and is now pulling the output of the power supply down, since all rails are powered from the same transformer, if one has a short or heavy load on the output, the whole lot will be affected.

Check the resistance between the +16v rail (for the inverter) and ground. You could measure between the output of L904 and GND.

[Tom-Servo]

Re: Viewsonic VA2431wm bad cap replaces, new problem

It's very unlikely you blew the power transformer, you would have to do something really stupid like put the primary winding directly across the mains without a fuse, or throw it at the wall or something.

Sounds to me like something else has shorted since the inverter transformer was not installed (likely the other MOSFET) and is now pulling the output of the power supply down, since all rails are powered from the same transformer, if one has a short or heavy load on the output, the whole lot will be affected.

Check the resistance between the +16v rail (for the inverter) and ground. You could measure between the output of L904 and GND.

2.6 Ohm from Ground to L904 (both contacts are the same)

[budm]

Re: Viewsonic VA2431wm bad cap replaces, new problem

You should remove those two power MOSFETS.

[Tom-Servo]

Re: Viewsonic VA2431wm bad cap replaces, new problem

You should remove those two power MOSFETS.

OK, will remove the MOSFETS tomorrow, too tired to desolder well tonight. Should I then retest L904 and GND? If so what should the resistance be? Are there good substitutes for AO4620 from what I listed before or any others not listed, or is it best just to order the exact part number from China.

Thanks!

[budm]

Re: Viewsonic VA2431wm bad cap replaces, new problem

The resistance should be in the 1000's, you will see it shows as low resistance first due to the filter caps are being charged up at the beginning.
No power to the board of course.

[Agent24]

Re: Viewsonic VA2431wm bad cap replaces, new problem

OK, will remove the MOSFETS tomorrow, too tired to desolder well tonight. Should I then retest L904 and GND? If so what should the resistance be? Are there good substitutes for AO4620 from what I listed before or any others not listed, or is it best just to order the exact part number from China.

Thanks!

I think the DMC3032LSD (saw them at DigiKey) might be a bit closer, but likely either of the ones you found should work as well, that said I am not a design engineer and I do not understand most of the reasons behind why they chose the AO4620 to begin with.

As far as I know, as long as the replacement has a higher or same current\voltage capability, the gate drive voltage is the same (in this case, logic level), and it does not switch appreciably slower than the original, it should work OK. There are also other factors like gate capacitance but I have no idea how critical they are to this (or any) design, so I just try to find something that matches up as closely as possible.

If it were me I'd get the AO4620 from a reputable seller (kinda hard to do when talking about China), or if that wasn't possible, I'd probably use those DMC3032LSD ones myself. If you want a better answer, you'd need to wait for someone with more knowledge to chime in.

[budm]

Re: Viewsonic VA2431wm bad cap replaces, new problem

You also want the Rds(on) to be the same or lower (better) or really close to the original.
The Gate capacitance, if it is large then the Gate will stay charge longer so the MOSFET will stay on after the drive signal is removed and stay like that until the charge Voltage drops below the Gate turn on Voltage then it will be off.
The circuit is the full bridge drive with 4 MOSFETs forming the bridge, so you will want to make sure only one pair of the bridge is turn on while the other pair is OFF, most drive circuit will be designed with dead time to help preventing all the MOSFETs to be all on at the same time.



https://cdn.badcaps-static.com/pdfs/...b253f5562e.pdf

[Tom-Servo]

Re: Viewsonic VA2431wm bad cap replaces, new problem

I mixed the MOSFET that was shorted, it was Q805. I removed Q805 and tested the resistance between L904 (+16v rail) and ground. It counted up until 9.400 K ohm then continued to count up after that but rather slowly. I then removed Q806 and pretty much observed the same behavior. Though sometimes when I measured the resistance it started in the M Ohm range and counted down from there.

[Tom-Servo]

Re: Viewsonic VA2431wm bad cap replaces, new problem

Replaced both AO4620's and now it turns on though now it's doing 2 seconds to black. Now that I think about it my brother said that's what was happening before until it stopped turning on at all. I was able to get the monitor to turn back on for 2 more seconds when I changed resolution.

Two things that I know about the monitor which may or may not be enough to confirm that the issue is the back light rather than the inverter transformer.

1. The picture when on looks extremely drab.
2. My brother just let the monitor go to a black screensaver since he bought it (ie: never was shut off).

Is that enough info to isolate what's causing the issue, or should I test some additional points? Thanks.

[Agent24]

Re: Viewsonic VA2431wm bad cap replaces, new problem

Does it still have image (shine a strong light on the panel) or power LED when it goes black or does everything shutdown completely?

If you still have image but no back-light, it leans towards the possibility of a bad lamp. BUT still could be inverter. The only real way to test for bad lamps is to plug in some good ones. Try budm's guide "Inverter testing using old CFL" if you have no other LCD CCFLs.

If you have no ESR meter, then one thing I would try is replacing all the other capacitors in the PSU save the mains filter one (the largest) as you already had a bulged one and it's quite possible the rest could be bad but not showing visible signs.