NEC Accusync LCDV9 - no power totally dead

[retiredcaps]

NEC Accusync LCD9v made in May 2003. There is no power at all. NO power LED indicator light and no noise/sound or smoke coming from unit. It is completely dead.

Sorry the lighting on the pics make it hard to see the top of the caps due to some window/sun/reflection, but there are no visible bulging caps (most are Capxon and some are Su'scon).

I have checked all the fuses for continuity and they are fine. I also touched up some of the more questionable solder joints.

/student mode ON/

Since I have my friend's multimeter, I want to LEARN what else I can check for with power OFF.

Can I check for open resistors using the continuity? I know that a resistor can be bad if it is out of spec and that is the more likely scenario than an open resistor, but is it worthwhile to check an open resistor in circuit (with power OFF)?

I'm planning to change the caps, but I want to see if there is anything else I can check in circuit with power OFF?

/student mode OFF/

PS. Note cap C214 (near upper left hand corner) is actually straddling the IC below. It was held onto the IC with white glue, but the glue is not holding it down anymore.

[retiredcaps]

Re: NEC Accusync LCDV9 - no power totally dead

Okay, I made some progress here.

After reading some previous threads, I figured out to measure the voltage across the main filter cap and I get 166V DC.

Then I moved onto the secondary side and could not find 12V or 5V. I then focused on the I801 chip as a starting point and it is SG3842GD. According to the datasheet

http://www.alldatasheet.com/datashee...C/SG3842G.html

pin 5 is GND and pin 8 is VREF. The description for VREF is "5V reference voltage output".

So I put my multimeter on DC (autoranging) and put the black lead on pin 5 (GND) and the red lead on pin 8 (VREF) and I get 0V. I think I should be getting 5V? So can I assume I801 is malfunctioning because I'm not seeing 5V or am I selecting the wrong GND?

Pin 7 is VCC and it is "Supply voltage input". This fluctuates from 10V to about 16V and oscillates between 10V and 16V.

Measuring the voltage across C807 (su'scon 22uf 25V) oscillates from 10V to 16V. I replaced it with a 68uf 25V spare cap just for fun and I get the same results (oscillating between 10V to 16V).

[PlainBill]

Re: NEC Accusync LCDV9 - no power totally dead

Okay, I made some progress here.

After reading some previous threads, I figured out to measure the voltage across the main filter cap and I get 166V DC.

Then I moved onto the secondary side and could not find 12V or 5V. I then focused on the I801 chip as a starting point and it is SG3842GD. According to the datasheet

http://www.alldatasheet.com/datashee...C/SG3842G.html

pin 5 is GND and pin 8 is VREF. The description for VREF is "5V reference voltage output".

So I put my multimeter on DC (autoranging) and put the black lead on pin 5 (GND) and the red lead on pin 8 (VREF) and I get 0V. I think I should be getting 5V? So can I assume I801 is malfunctioning because I'm not seeing 5V or am I selecting the wrong GND?

Pin 7 is VCC and it is "Supply voltage input". This fluctuates from 10V to about 16V and oscillates between 10V and 16V.

Measuring the voltage across C807 (su'scon 22uf 25V) oscillates from 10V to 16V. I replaced it with a 68uf 25V spare cap just for fun and I get the same results (oscillating between 10V to 16V).

Pretty good job of troubleshooting thus far. It appears the SMPS is starting, but the 'run' power never kicks in. If you look at the 'Typical Application' drawing on page 1 of the datasheet you will see it shows a 10 uF capacitor and a diode coming from the 'auxiliary winding' on the transformer. A bad power FET, a shorted diode on the transformer output, a reversed cap, or an open 'run' diode would give the symptoms you are seeing. Use your DMM on the 5volt output. If it pulses at the same rate as Vcc, the power FET is good. Personally, I'd bet on SG3842G. You can cheat - do a search on this site for SG3842G (or SG3842) and see if others have reported problems with it.

PlainBill

[retiredcaps]

Re: NEC Accusync LCDV9 - no power totally dead

Thanks PlainBill. I couldn't find anything searching around on the SMPS.

Just an update on this. Some caps came in today so I changed out the small 1uF 50V caps and soldered them in correctly (checked polarity 3 times).

I plugged in the monitor and started to measure DC voltage on the secondary side when I smelled something. I quickly disconnected power and saw that I804 (MP 1410) had some burning around pin 6 & 7. I verified it was temperature hot so it was under duress for some reason.

I found the datasheet at



and it says it is a 2A Step Down DC/DC converter.

Right now, the plan is to re-examine all the components in circuit (with NO power) to see if there is any more damage.

[Regenpak]

Re: NEC Accusync LCDV9 - no power totally dead

The SG3842 (AKA UC3842) pin 7 fluctuating between 10 and 16 volts means that it hovers between its startup voltage (16 V) and undervoltage lockout (UVLO - 10 V). This could be caused by the capacitor C807 being bad (use of low ESR unit is mandatory). It could also be that there's an overload condition causing it to shutdown and then startup again. A test I like to do is to power the circuit from an external power supply while monitoring the feedback optocoupler I802. If the TL431 zener (next to R825) starts to draw current you've hit the target output voltage. Then see if the current drawn by the circuitry is as expected. In case of an overload situation troubleshooting should be easy (smoke!) so use a power supply with current limiting capabilities. I804 could very well turn out to be the culprit.

Also check the sustain power circuit (D801 I guess) and C814.

[retiredcaps]

Re: NEC Accusync LCDV9 - no power totally dead

Thanks for the UC3842 reference. I would have never equated SG for UC.

I don't have an ESR tester and don't plan to get one. I can't do any testing today because it is thunder and lightning outside and it wouldn't be safe to probe around.

I also don't have a PS with current limiting capabilities.

Thanks for your explanation. I think I follow what you are saying. It certainly helps fill in some gaps. I certainly can't match your 30 years of experience, but I'm eager to learn.

The SG3842 (AKA UC3842) pin 7 fluctuating between 10 and 16 volts means that it hovers between its startup voltage (16 V) and undervoltage lockout (UVLO - 10 V). This could be caused by the capacitor C807 being bad (use of low ESR unit is mandatory). It could also be that there's an overload condition causing it to shutdown and then startup again. A test I like to do is to power the circuit from an external power supply while monitoring the feedback optocoupler I802. If the TL431 zener (next to R825) starts to draw current you've hit the target output voltage. Then see if the current drawn by the circuitry is as expected. In case of an overload situation troubleshooting should be easy (smoke!) so use a power supply with current limiting capabilities. I804 could very well turn out to be the culprit.

Also check the sustain power circuit (D801 I guess) and C814.

[oPossum]

Re: NEC Accusync LCDV9 - no power totally dead

Just spent a few hours trying to fix one of these.

Traced the problem to a short of the 3.3 V rail inside the Genesis GM2120 chip. It causes the 3.3 V switching regulator (MP1410) to fail. When I replaced the MP1410 with something that can handle a bit more current it still failed to power up and the 3.3 V rail measured about 1/2 a volt. Used a bench supply to put out 3 amps in into the 3.3 V rail and the GM2120 got hot.

Not sure if I am going to try to replace the GM2120.