Apple Multiple Scan 15AV Display.Help identifying a resistor

[tony359]

Hi all,

I am struggling to identify a couple of resistors which are connected in series between the source of a Mosfet and the negative of the bridge rectifier of the SMPS section of an old CRT monitor.

The resistors read 0.6 Ohm out of the circuit.

Pictures attached. The black seems to be black so my options are 18Mohm or... 56.8Ohm but there is no black as tolerance.

So either it's an 18 MOhm and the resistor has shorted or something's is fishy about that colour code.

The SMPS tries to power up but keeps switching off. Hard to say why but my scope seems to confirm that the Current pin of the controller is getting a voltage higher than 1V which means "overcurrent" so the controller shuts it down.

Can I have your opinion please? Would 18Mohm large resistors make any sense to connect the source of a Mosfet to the negative of the bridge rectifier?

Thank you!

[momaka]

Re: Help identifying a resistor

Resistors don't go short-circuit when they fail. They always go open-circuit or high-resistance.

0.6 Ohms per resistor is correct. These are 4-band resistors. The 5th (black) band might be reliability or something else. Otherwise, what you have is green (5), blue (6), silver (0.01 multiplier), and gold (5% tolerance)... so 56 x 0.01 = 0.56 Ohms.

If these were blown open, you would have found the MOSFET to be shorted as well.

But since that's not the case and your PSU does seem to try to start, check the output rectifiers for short-circuit or if something else is causing an overload. The issue could also be a bad "startup" capacitor - typically a small electrolytic in the range of 10 to 47 uF and rated for 25-50V. When this startup cap goes bad, the PSU will "hickup" and not start. Bad capacitors on the output can also cause this.

The quickest way to test is to take a hair dryer or heat gun and heat the entire PSU board a little (just where it starts to feel very hot, but not burning your hand). Then plug it in and see if it starts up. If it does, the issue is likely bad capacitors (be it output or startup cap or both.) If not, then look for shorted components on the output.

[PeteS in CA]

Re: Help identifying a resistor

What momaka said. Those resistors are being used for current sense, the common "failure" being the MOSFET shorting D-S and flash-frying the current sense resistors. The good news, then, is that the associated MOSFET is not shorted.

[tony359]

Re: Help identifying a resistor

Thank you all.

I checked the PWM controller and what I see is that the Current sense pin goes over 1V - according to the datasheet that is enough for the switching to shutdown. So the controller seems to be sensing a current overload.

I have checked everything I could and I did not find anything obvious.
Is the startup capacitor the one which powers the controller chip while the switching builds up, before the PSU is able to power itself up? I do see VCC of the controller chip raising to around 15V, then the PSU "clicks", the voltage drops and everything stops. I'll check that too.

But as I see Current sense going over 1V I'd think there is a short/overload somewhere, would you agree?

controller datasheet and Current sense screenshot attached!

Thanks for your help so far!

[PeteS in CA]

Re: Help identifying a resistor

Seems possible. I'd look at the rectifiers for the various outputs.

[R_J]

Re: Help identifying a resistor

Start by replacing the cap on the vcc line, or at least scope that line. Post the model number or the service manual.
Check that the horz. output transistor is not shorted

[momaka]

Re: Help identifying a resistor

Check that the horz. output transistor is not shorted

^ What he said.

I see you mentioned this PSU is for a CRT set. When the HOT (Horizontal Output Transistor) shorts, that will make the PSU overload / not start.
In particular, check the HOT between C-E junctions. B-E might show very low resistor, since it typically has a low-value resistor between B-E (but not always.)

[tony359]

Re: Help identifying a resistor

Thanks.

The monitor is an old Apple Multiple Scan 15AV Display. The guts are actually made by LG and all I can see on the board is PN=6870T024A 11

The "service manual" can be found here: https://1drv.ms/b/s!AukePwK7XyEPhOZL...X5v0Q?e=kcJhdZ

Some pics attached.

I will definitely check the horizontal transistor (I did check them all but will do again) but I have a feeling this is a PSU issue as the PSU is barely starting up, it feels as there is not much coming out of that supply before it shuts down. But I could be wrong.

In fact, when I power up from cold, it stays on for say a second, then it shuts down and the cycle speeds up over 2-3 seconds until it only stays up for say 300ms.

Thanks for your help so far!

[tony359]

Re: Help identifying a resistor

Update.

Horizontal transitor (C5129) showed a short between base and emitter (short in diode mode and 0.4Ohm in resistance mode). I took it out but the transistor is fine.

I inspected the PCB and I see the Base is going back to ground via a small transformer (labelled T701). This small transformer reads indeed 0.4 ohms on my meter.

Based on what Momaka says, I assume this is normal?

I have two small capacitors close to the SMPS controller chip. They are 100uF 50V. One of them is indeed going to VCC. The capacitor tests fine out of circuit.

If I scope VCC, it jumps on and off - between 10 and 15V. Pic attached (roll mode).


I took out the main switching transformer to understand how many voltages are generated. I counted 5. I can follow them on the back of the PCB, they all go through diodes and capacitors (obviously!) and then disappear around the board. I tested all I could think of but I did not find a short.

I've removed a connector which seems to take filament voltage and 12V to the neckboard but nothing.

The outputs sometimes go through transistors and voltage regulator but, again, I don't see a short there.

Any other ideas?

I might want to inject some voltage on the various rails to see if by any chance my thermal camera sees something?

I have a twin board which - surprisingly - is showing the same fault, I could use that one.

Thank you!

Good news: using voltage injection I identified a small shorted transistor in the middle of the board - not sure of what it does. It's a 2SA1275-Y. I can buy one on Ebay for an arm and a leg, would you help me find a modern equivalent please?

It's a PNP one, In stock I have 1015, 1020, bd682G (slightly bigger) and*bc638. Happy to purchase some but hopefully not £5.99 for one!

[momaka]

Re: Help identifying a resistor

Hmmm. Unfortunately, I don't see a schematic diagram in the SM.

T701 is likely the Base drive transformer for the HOT. 0.4 Ohms resistance on its windings doesn't seem troubling at this point, since it only has a few turns on there, meant to drive the HOT with relatively low voltage but plenty of current. By the way, what's the part number on the HOT? Does it have a built-in damper diode? If not (usually not on CRT monitors), there should be a separate damper diode across C-E junctions. Check that diode... though I suspect you will find it OK, otherwise you would have seen a short-circuit on the HOT while in circuit between C-E.

That said, also check all of the other components on the base of the HOT transistor. There should a be a low-value resistor to ground (4.7 to 47 Ohms?) See if that's good.

Lastly, you can try running the monitor with the HOT removed. If the PSU doesn't start up, then there is something else wrong. If it does, then you might have a shorted/bad FBT.

And just another sanity check... this being an LG-made monitor, I don't expect they'd have a separate HV circuit or separate G2 supply, like Sony does. On Sony monitors, those two also have high-power transistors that could fail and cause the PSU to tank / not start. But here, I don't think that's the case. Either way, check all large transistors on the board, if you can.

[tony359]

Re: Help identifying a resistor

Thanks Momaka. The HOT is C5129.
I've checked everything around it and could not find issues.

Yes, no schematics - that's why I quoted "service manual", it's more a "service centre" manual to show them how to swap parts

For the little I know about CRT, there are three HV cables coming out from the flyback. One goes to the CRT, two (focus and G2?) go to the neckboard.

I edited my previous message after you posted yours, I found a shorted small transistor - I see it's on the 70V line going to the neckboard, but it powers something else on the main PCB, it just uses the 70V rail I think.

Can you help me finding a modern equivalent of that please?

Would this https://cdn.badcaps-static.com/pdfs/...9f2572864c.pdf work? Making sure it's a P version for Hfe 160-320 as the original.

Thank you!

[momaka]

Re: Help identifying a resistor

I edited my previous message after you posted yours, I found a shorted small transistor - I see it's on the 70V line going to the neckboard, but it powers something else on the main PCB, it just uses the 70V rail I think.

Can you help me finding a modern equivalent of that please?

Would this https://cdn.badcaps-static.com/pdfs/...9f2572864c.pdf work? Making sure it's a P version for Hfe 160-320 as the original.

Yup, I was just looking at datasheets while you were posting your edit and came up with the same thing as you: 2SA1013. Digikey (US) has them for $0.46 a piece.

I also found a Toshiba TTA004B, which comes in a slightly bigger TO-126N case and can handle a bit more power. Other specs seem comparable. More or less same price too. So maybe worth getting both?

For the little I know about CRT, there are three HV cables coming out from the flyback. One goes to the CRT, two (focus and G2?) go to the neckboard.

Cool, standard CRT affair then (HV anode, focus/G1, and screen/G2 all generated by the flyback.)

Thanks Momaka. The HOT is C5129.
I've checked everything around it and could not find issues.

Well, the resistor going from B to E has to be checked out of the circuit so that you're not reading across the transformer... but with the other shorted transistor you found, I think this one is likely going to be OK.
At least the 2SC5129 doesn't appear to be anything unobtanium, had it gone bad. I had to purchase a HOT for a 19" Iiyama monitor, and that one was $6 without the shipping. But that's how it is - specialty high-power transistors will always be expensive.

Yes, no schematics - that's why I quoted "service manual", it's more a "service centre" manual to show them how to swap parts

Lol, good ol' Apple assuming we are all monkey techs that can't figure out anything ourselves. I guess they had that attitude back in the day too and not just a "recent" thing.

[tony359]

Re: Help identifying a resistor

I am in the UK - Can order them from RS components or Farnell but the issue is shipping: by the time you get them it's many pounds!

Ebay sellers know that obviously. I might have found the 1013 from one seller but I am checking the Hfe as it's not mentioned on the listing. I'll find out.

The Toshiba's are even more expensive. Unfortunately it used to be easy to purchased those things online - I think CPC or Farnell used to have a £5 minimum purchase for free shipping. That's long gone, it now takes £30 or £35 to have free shipping and some companies add "handle fees" for small orders.

Just for fun: ordering 5 SA1013 from Farnell UK (RS sell you at least 50) it would cost me £14.58. Is it better in the US?

Bottom line: I don't want to spend £5 for ONE transistor on Ebay!

I'll check the resistor on the HOT but I did remove the transformer to check things!

I'm really curious to see what happens - this would be my first repaired CRT!

Thanks a lot for your time so far!

[R_J]

Re: Help identifying a resistor

Ksa1013ybu Digikey $0.74 each

[tony359]

Re: Help identifying a resistor

I am not in the US The attached is for 10 transistors.

I can find it from RS-Components UK but they still charge a similar amount of money for small orders.

[tony359]

Re: Help identifying a resistor

Update.

I swapped the transistor (Ebay) and checked things around.

The monitor powered up and stayed in standby happily (Yay!). When I applied VGA signal, the monitor fully powered up, I had HV, a weird image on screen and then The Magic Smoke!

The same transistor shorted (Base with Collector) and a resistor connected to it went on fire.

I've looked around and here are my notes:

- The transistor seems to be generating +70V for the neckboard. On the Neckboard there is a CVA2422TL CRT driver. I've noticed that I've got 90Ohm between GND and the 70V line on the neckboard - my other neckboard reads some Kiloohms and climbing.

CVA2422TL datasheet: https://pdf1.alldatasheet.com/datash...CVA2422TL.html

Shall I move my attention to the neckboard? Could it be the reason why the 70V rail fails every time?

Could a bad CRT cause this issue on the neckboard? Please note I read 90 Ohm with the neckboard disconnected from everything (but the Flyback)

I don't want to give up know. The area which went up in smoke seems pretty simple - I'm sure some of you can recognise the topology, I suspect it's some voltage regulation?

When the screen came up it was pretty weird, white with blotches. Maybe someone can recognise the symptom of something? See video below.

Also, I cannot assume that the trimmers around the PCBs haven't been touched. Still, I wouldn't expect components to go up in flame if something is misadjusted?

Here is the video of the accident (I have a small youtube channel where I publish my repairs):

https://www.youtube.com/watch?v=YqpgZc0HX2M

Attached some pictures - the main PCB is single layer so the circuit can be easily followed there!

Thanks so far!

[R_J]

Re: Help identifying a resistor

There are some diodes off that 70v line, D319, D320, D318 Also check the transistors, Q307~309, Q306
It should not be too hard to locate what is causing the 90Ω reading.

The crt screen is likely magnetized and need degaussing, the built in automatic degaussing circuit may not be enough to do the job since it only operates for a a few seconds when the monitor is plugged in.

[momaka]

Re: Help identifying a resistor

- The transistor seems to be generating +70V for the neckboard. On the Neckboard there is a CVA2422TL CRT driver. I've noticed that I've got 90Ohm between GND and the 70V line on the neckboard - my other neckboard reads some Kiloohms and climbing.

That would definitely be one of the first things I suggest to investigate, as that could well be why Q953 shorted again.

Check the components that R_J suggested.

Also check the following diodes on the neck board:
D312, D313, D314, D315, D316, and D317.
These are protection diodes on the outputs (pins 3, 4, and 10) of the video driver IC. The first three (D312-D314) connect between 70V rail and output of the IC (through 33-Ohm resistors, it looks like), and the latter three to ground(?) If any of these are bad, they can pull down the 70V rail.

If these are OK and the components that R_J mentioned above are OK, then it could also be the video driver IC. Easiest way to check if the video driver IC is not causing the short: lift one leg of ferrite bead L307. If 90-Ohm resistance disappears, then the video driver IC could be the issue... unless somehow capacitors C329 or C345 are shorted.

Could a bad CRT cause this issue on the neckboard? Please note I read 90 Ohm with the neckboard disconnected from everything (but the Flyback)

Extremely unlikely for the CRT tube to be bad, and impossible for it to cause the low resistance, even if it was bad (leaked air).
Also, the fact that you had a high voltage and raster appear is a good indication that the tube is healthy.

I don't want to give up know. The area which went up in smoke seems pretty simple - I'm sure some of you can recognise the topology, I suspect it's some voltage regulation?

Well, Q953 is controlled by Q954, so it may also be worthwhile to check Q954 now as well... though it should likely be OK due to R960.

It is a simple linear regulator circuit, it appears.
Q953 has high voltage coming from PSU and going into Emitter. R959 keeps Q953 Base at same potential as Emitter until Q954 is enabled (turned On), which then shorts one side of R960 to ground, thus making some current flow through R960 and lowering the voltage at the Base of Q953, which then outputs a regulated voltage. From your video, though, it appears as if R960 might actually be a Zener diode(?) If that's the case, please note so.

Also, could we see pictures of the entire solder-side of the main board? And top too, if you already have it out? I think it would be beneficial to have these for anyone trying to help and for anyone who might encounter the same issues too. You're fortunate enough to have two of these monitors, so you can compare values between them. But for anyone without, it can become a guessing game. The "service manual" clearly isn't as helpful as we'd like.

When the screen came up it was pretty weird, white with blotches. Maybe someone can recognise the symptom of something? See video below.

Looks like poorly degaussed screen to me as well, like R_J said.

Also, I cannot assume that the trimmers around the PCBs haven't been touched. Still, I wouldn't expect components to go up in flame if something is misadjusted?

Correct.
... unless a trimmer has gone completely open-circuit. But you can check for that fairly easily.

Here is the video of the accident (I have a small youtube channel where I publish my repairs):

https://www.youtube.com/watch?v=YqpgZc0HX2M

Cool!
Was really helpful to see the video too of exactly what happened.

I was repairing a 19" Iiyama CRT last summer and had a similar excitement moment after replacing the HOT... only to hear the HV try to come up and smell a burned transistor again. But in the case of that one, it was my fault - I used MX-4 thermal compound on the insulation pad of the HOT, and the HOT had a metal tab. The MX-4 went conductive enough to short out the B+. Miraculously, nothing else was taken out. The CRT still works, but I have it with a much wimpier HOT and I can't get it to do more than 1024x768 without the picture going really weak. Didn't have a new thermal pad, so it went back to the back of my list of projects.

[tony359]

Re: Help identifying a resistor

Thanks for all the super-helpful inputs.

I can confirm R959 and R960 are both resistors, 1K and 33K 5%. I've placed an order. Shall I be concerned by the blue coloured body of R960? Might that mean something?

I'll check on the neckboard, I think I'll try voltage injection again and see is something shows up on thermal camera.

Great idea about sharing the whole board for future reference, it's attached.

Q953 had died as I said but Q954 is still ok. Thankfully I bought a bunch of those 1013 transistors - it would have been sad to smoke my only £5 transistor!

I was thinking about why the more destroying outcome this time. If not mistaken the existing transistor shorted Collector with Emitter. This time Base and collector shorted.
Correct me if I am mistaken please (one reason I do these repairs is to learn new things): if Collector and Emitter short out, then I directly connect the PSU to whatever is shorted on the neckboard. Probably a larger component which could withstand the extra current without leaving visible clues while the PSU shutdown.

When Base and Collector shorted - and Q954 opened - the current from the PSU flowed through towards the short on the neckboard. That is a 1/8W resistor so it couldn't bear the extra current.

Would that make sense?

Also, agreed on the CRT which seems to be working as I had a reasonable raster for a moment.

This is pretty fun - Can't wait to try again.

Is the MX-4 conductive??

Thanks again!

[tony359]

Re: Apple Multiple Scan 15AV Display.Help identifying a resistor

Definitely a shorted CRT driver I'm afraid. I can see the physical short inside.

It would only draw 100mA with 5V but when I put 7V on the line it started snowballing and my bench PSU started limiting.

Those ICs are 9.99 on Ebay - I think I'll swap the neckboard for now, I have plenty of spares apparently as some of those monitors have smashed CRTs unfortunately.

I'll update when I try again!