Toshiba 19AV51U power regulation problem

[iKitsune]

I've adopted a Toshiba 19AV51U for $5 today at a yard sale, which was suspected of having bad caps based on the behavior (pulsing power light). Upon inspecting the power supply, it appears to have intact caps, of the Chemi-Con KZE variety. Output voltage "wanders"; it slowly fluctuates down from 12v and 5v to roughly 9v and 2v, and then back up to around a peak of about 13v and 6v before dropping again. Unsure if the "wandering" is due to lag in my $12 DMM taking a while to register a faster change. With the power supply disconnected from the mainboard and CCFLs, the only output I find powered appears to be nominally +5v, and is "wandering" from about 4.7v to 5.4v. Visually, everything appears OK to me. I'm hoping someone with more experience than SMPSes than me can offer a hint as I'm lost, and I'd really adore getting this unit working.

[PlainBill]

Re: Toshiba 19AV51U power regulation problem

I've adopted a Toshiba 19AV51U for $5 today at a yard sale, which was suspected of having bad caps based on the behavior (pulsing power light). Upon inspecting the power supply, it appears to have intact caps, of the Chemi-Con KZE variety. Output voltage "wanders"; it slowly fluctuates down from 12v and 5v to roughly 9v and 2v, and then back up to around a peak of about 13v and 6v before dropping again. Unsure if the "wandering" is due to lag in my $12 DMM taking a while to register a faster change. With the power supply disconnected from the mainboard and CCFLs, the only output I find powered appears to be nominally +5v, and is "wandering" from about 4.7v to 5.4v. Visually, everything appears OK to me. I'm hoping someone with more experience than SMPSes than me can offer a hint as I'm lost, and I'd really adore getting this unit working.

First, a very minor point, this is a TV, not a monitor. Not a big deal, but....

In the future, when you get one of these please do everyone a favor and take one over-all shot of each side of the board. Trying to navigate around 4 pictures (which don't quite cover everything) for the top made me so tired I didn't even try the bottom.

This is a fairly complex supply. It's probably got PFC (power factor correction), but I didn't bother looking for the third controller. See above for the reason. It has a main and a standby supply. The first step would be to try switching it into standby (AKA OFF). Then check the output of the standby supply. It's the section in the lower left corner of the first picture. It should probably have 5 volts out and be stable. Don't worry if the output voltage isn't 5 volts as long as it is stable.

If it isn't stable, check the voltage across the big 400volt cap. That should be stable.

PlainBill

[iKitsune]

Re: Toshiba 19AV51U power regulation problem

First, a very minor point, this is a TV, not a monitor. Not a big deal, but....

Sorry, in my mind it looked more like a monitor than a TV inside. I guess with LCDs the distinction is pretty blurry.

In the future, when you get one of these please do everyone a favor and take one over-all shot of each side of the board. Trying to navigate around 4 pictures (which don't quite cover everything) for the top made me so tired I didn't even try the bottom.

Again, I apologize, my poor tired 2001 Sony Mavica's autofocus wouldn't lock when I tried to get an overall shot of the board, and I doubt a cameraphone picture would help anyone. I'm hoping to get a new camera soon.

This is a fairly complex supply. It's probably got PFC (power factor correction), but I didn't bother looking for the third controller. See above for the reason. It has a main and a standby supply. The first step would be to try switching it into standby (AKA OFF). Then check the output of the standby supply. It's the section in the lower left corner of the first picture. It should probably have 5 volts out and be stable. Don't worry if the output voltage isn't 5 volts as long as it is stable. If it isn't stable, check the voltage across the big 400volt cap. That should be stable.

PlainBill

5vsb is not stable. It's fluctuating between 4.7 and 5.3v. 400v cap is slowly fluctuating between 167 and 169. My DMM does not show fractions on the 600v scale.

I'm uploading a new attempt at front and back photos of this board.

[iKitsune]

Re: Toshiba 19AV51U power regulation problem

So, hrm. After reading up on SMPS theory, I've an idea, but I wanted to run it by everyone before I go dropping money on barking up the wrong tree: I'm theorizing the 400v 150uF filter cap has dried up and died. Any easy way to verify this, without an oscilloscope or ESR meter?

[PlainBill]

Re: Toshiba 19AV51U power regulation problem

Sorry, in my mind it looked more like a monitor than a TV inside. I guess with LCDs the distinction is pretty blurry.


Again, I apologize, my poor tired 2001 Sony Mavica's autofocus wouldn't lock when I tried to get an overall shot of the board, and I doubt a cameraphone picture would help anyone. I'm hoping to get a new camera soon.


5vsb is not stable. It's fluctuating between 4.7 and 5.3v. 400v cap is slowly fluctuating between 167 and 169. My DMM does not show fractions on the 600v scale.

I'm uploading a new attempt at front and back photos of this board.

Did you have enough control that you were able to put the TV in standby? That fluctuating voltage really bothers me. The standby SMPS should be able to regulate the output voltage even if the input voltage is varying by 20 volts or more. If on the other hand there was a problem with the load on the main SMPS, it could cause the voltage across the big cap to vary, but the standby SMPS should be stable.

The first step would be to replace (or at least temporarily bridge) the 150uF, 400 volt cap and see if that stabilizes the supplies.

PlainBill

[iKitsune]

Re: Toshiba 19AV51U power regulation problem

Did you have enough control that you were able to put the TV in standby? That fluctuating voltage really bothers me. The standby SMPS should be able to regulate the output voltage even if the input voltage is varying by 20 volts or more. If on the other hand there was a problem with the load on the main SMPS, it could cause the voltage across the big cap to vary, but the standby SMPS should be stable.

The first step would be to replace (or at least temporarily bridge) the 150uF, 400 volt cap and see if that stabilizes the supplies.

PlainBill

I must have posted my theory above while you were writing your post, PlainBill. I do not have enough control to leave the set in standby, but disconnecting the mainboard appears to make it drop all but the 5vsb anyhow. I am digging through my boxes of donor parts to see if a quickie temp replacement 150uF 400v cap is available. I will post back once I've tried that.

[iKitsune]

Re: Toshiba 19AV51U power regulation problem

The symptoms did not change after replacing the 400v 150uf cap with a known-decent (good is a stretch, it's a CapXon) 400v 120uf.

[PlainBill]

Re: Toshiba 19AV51U power regulation problem

The symptoms did not change after replacing the 400v 150uf cap with a known-decent (good is a stretch, it's a CapXon) 400v 120uf.

I knew that would be too easy.

Here's the next thing to try. Since you have an infinite source of spare parts, there are two electrolytic caps next to the heat sink for the power FET for the standby power supply. They are just above and to the right of the bridge rectifier. Try replacing those. If bad they could have an effect on the output of the standby SMPS. This is still a long shot.

PlainBill

[iKitsune]

Re: Toshiba 19AV51U power regulation problem

The caps that are there are 50v 10uf, I can't find any like them in my arsenal. Will 50v/20uf work, or is there another step we could try?

[iKitsune]

Re: Toshiba 19AV51U power regulation problem

More information: I decided to note some voltages around the board just in case it's helpful.

At C810, one of the pair of small lytics by the bridge rectifier, I measured 65-68vdc across it.

At W17, a wire running across the top of the board, immediately to the left of the insulated wire near the main transformer, I measured 81-83 vdc.

At W48, a wire running across the top of the board in the inverter section, near and perpendicular to the barcoded label, I measured between 12.3 and 17 vdc.

On the power connector going to the mainboard, I measured between 4.9 and 5.1 (nominally 5v?) on the first two pins, 0 volts on the next two, a voltage with a large swing between 0.4 and 2.4 volts on the next pin, two more with 0 volts, two moving between 5.5 and 9 volts, 2 more at 0, and one showing a steady 0.05 volts.

I tracked down a pinout of one of the control chips on the mainboard; on a pin that is intended to be 5 vdc, I measured between 4.7 and 4.9 vdc.

[iKitsune]

Re: Toshiba 19AV51U power regulation problem

I'm out of ideas. Within my limited experience, everything -looks- good. This board doesn't seem to have any common faults I can find (it's a VTV-P1901 power supply, looks like a -lot- of Sony and Toshiba models use it.) Before I drop eighty bucks on a new one, any other ideas I might try?

[iKitsune]

Re: Toshiba 19AV51U power regulation problem

This is turning into a journal of myself learning about SMPSes. I found the datasheet for the SMPS controller - those caps PlainBill pointed out are connected to Vcc of the standby PWM controller, a Leadtrend LD7576PS. There are two of them, one apparently for the standby power, and one for the main power. [edit] I'm retarded and I can't read caps, so I didn't notice those two caps are different values. The standby controller has a 10uF and a 22 uF 50v cap, and the other controller has just a 22 uF. I'm going to switch those between the two positions and see what changes. I believe the 10 uF to be in the voltage feedback loop; the datasheet for the controller suggests one.

The next thing I'm going to check after that is the optoisolators located between the two transformers - the Leadtrend controller seems to be picky about its voltage feedback loop.

Plainbill: If you're still reading this, am I on the right track?

God, I've learned more about switched-mode power supplies today than I thought I ever would.

I'm now determined - I'm going to fix this board, regardless of effort. This seems more a learning experience than "pay dude in Korea 80 clams" would be.

[PlainBill]

Re: Toshiba 19AV51U power regulation problem

This is turning into a journal of myself learning about SMPSes. I found the datasheet for the SMPS controller - those caps PlainBill pointed out are connected to Vcc of the standby PWM controller, a Leadtrend LD7576PS. There are two of them, one apparently for the standby power, and one for the main power. [edit] I'm retarded and I can't read caps, so I didn't notice those two caps are different values. The standby controller has a 10uF and a 22 uF 50v cap, and the other controller has just a 22 uF. I'm going to switch those between the two positions and see what changes. I believe the 10 uF to be in the voltage feedback loop; the datasheet for the controller suggests one.

The next thing I'm going to check after that is the optoisolators located between the two transformers - the Leadtrend controller seems to be picky about its voltage feedback loop.

Plainbill: If you're still reading this, am I on the right track?

God, I've learned more about switched-mode power supplies today than I thought I ever would.

I'm now determined - I'm going to fix this board, regardless of effort. This seems more a learning experience than "pay dude in Korea 80 clams" would be.

Yes, you are on the right track. The most likely point for the problem to occur is in the feedback circuit. The problem that has me baffled is how a single problem can affect BOTH supplies.

PlainBill

[iKitsune]

Re: Toshiba 19AV51U power regulation problem

"When the going get weird, the weird turn pro", a great man once said.

I have more information, after receiving roughly 165 vdc across the tip of my pinky!

I borrowed a friend's oscilloscope for about half an hour this afternoon; that's all he'd trust me with it. Both PWM controllers appear to be on, they're emitting a roughly-square-wave on their OUT pins, at about 65 khz. I have confirmed that the ripple on the DC output, and across the 400v cap is the same frequency, but I'm unsure as to what frequency, as I'm not 100% as skilled with a scope as I'd like to be.

Would the chopper transistors let 60 or 120 hz ripple all the way through the system? I know how to set the volts per division on the scope, the ripple was about 2 volts p/p on the 5V lines, and about 6 volts on the 12v line. The enable voltage to the PWM controllers is rippling too, but staying within the "enable" range of the controller.

[PlainBill]

Re: Toshiba 19AV51U power regulation problem

"When the going get weird, the weird turn pro", a great man once said.

I have more information, after receiving roughly 165 vdc across the tip of my pinky!

I borrowed a friend's oscilloscope for about half an hour this afternoon; that's all he'd trust me with it. Both PWM controllers appear to be on, they're emitting a roughly-square-wave on their OUT pins, at about 65 khz. I have confirmed that the ripple on the DC output, and across the 400v cap is the same frequency, but I'm unsure as to what frequency, as I'm not 100% as skilled with a scope as I'd like to be.

Would the chopper transistors let 60 or 120 hz ripple all the way through the system? I know how to set the volts per division on the scope, the ripple was about 2 volts p/p on the 5V lines, and about 6 volts on the 12v line. The enable voltage to the PWM controllers is rippling too, but staying within the "enable" range of the controller.

165 volts? "That which does not kill you only makes you stronger."

The large filter cap should show perhaps a maximum of 5-10 volts of ripple at 120 Hz. The output of the SMPS filters should be clean - less than 50 mV. Where did you hook the scope ground when you measured the outputs?

There are two 8 pin ICs on the hot side of the power supply. What are the part numbers?

PlainBill

[iKitsune]

Re: Toshiba 19AV51U power regulation problem

Scope ground was connected first to the chassis, and then I read somewhere "Don't do that!", so I connected it to the output ground going to the mainboard.

They're Leadtrend LD7576PS PWM controllers.

I found the datasheet at http://www.datasheetdir.com/LD7576PS+download ; I've been referring to it while testing this circuit. It's funny, I'm really learning how SMPSes work just fooling with this one. Maybe some day I'll be able to fix one on my own.

[iKitsune]

Re: Toshiba 19AV51U power regulation problem

So I was staring at this power supply today, and I realized something - what if it's not ripple; what if it's the controller noticing a fault condition, shutting down, auto-resetting, etc, ad infinitum? The PWM controller can be set to auto-recover in a really BS short time. Would anyone have any suggestions on checking that?

[PlainBill]

Re: Toshiba 19AV51U power regulation problem

So I was staring at this power supply today, and I realized something - what if it's not ripple; what if it's the controller noticing a fault condition, shutting down, auto-resetting, etc, ad infinitum? The PWM controller can be set to auto-recover in a really BS short time. Would anyone have any suggestions on checking that?

I haven't discussed this, but you must have the power supply plugged into an isolation transformer if you are using a scope. If you are using a DMM, be cautious - pin 4 - Gnd - is at about 80 VAC relative to earth ground.

The pertinent pins are CS - current sense - pin 3 and Comp - feedback - pin 2.

The SMPS controller will start limiting the output if pin 3 rises above about .85 volts. The Comp input is at 5.6 volts if the output is below the selected voltage and normally regulates at about 2.3 volts.

PlainBill

[iKitsune]

Re: Toshiba 19AV51U power regulation problem

OK. Had one too many tonight to check it out, but I will tomorrow when I can see more clearly. Thanks for all your help.

Also, is the double-conversion UPS I already have my desk on a sufficient isolation transformer? It's a Liebert, not a little Belkin or some such.

[PlainBill]

Re: Toshiba 19AV51U power regulation problem

OK. Had one too many tonight to check it out, but I will tomorrow when I can see more clearly. Thanks for all your help.

Also, is the double-conversion UPS I already have my desk on a sufficient isolation transformer? It's a Liebert, not a little Belkin or some such.

Darn if I know. One thing to try is to hook about a 5 watt 125 volt light bulb from the - lead of the big cap to ground. Measure the AC voltage across it. If it is higher than .5 volt, the UPS does not provide sufficient isolation.

PlainBill