Laptop Backlight broken

[Tekno Venus]

Hi guys,

One for the hardware experts here. Our family laptop here is a Dell Studio 1537 that we've had for quite a few years now. It's not used massively, but we do use it. The other month, we switched it on to find the screen wasn't working. Shining a torch at the screen showed that the display was in fact working, but the backlight was broken. This was confirmed when we used an old external monitor.

So, the normal cause of this is a broken inverter board. So I ordered one of these: NEW DELL STUDIO 1535 1536 1537 LED SCREEN INVERTER BOARD M465J 0M465J | eBay and replaced it. This was much harder than it needed to be - Dell really didn't make that laptop easy to repair... I think you can guess where this is going. Put it back together and turned it on - no backlight.

The inverter is definitely the right part, the model number of the old one matches the new one. Does anyone have any ideas on what else it could be? The screen is an LED backlit one, not CCFL so replacing the CCFL is obviously not possible!

The local repair shop wanted to charge £50 for a diagnosis, and then charge for the repair afterwards. Considering the inverter was cheap, I decided to try replacing it. Other than a new screen (which is about £90 and out of our budget), I don't know what to do.

Thanks in advance!

Stephen

 

[Wrench97]

Way back when, Dell would always send both a inverter and a screen to be replaced in pairs under warranty.
Once in a great while it would turn out to be the cable that attaches the lid to the motherboard.
Unfortunately there is not a good way to find out which without swapping parts in.

 

[Digerati]

Dell really didn't make that laptop easy to repair...

It's the entire industry, not just Dell. Because there are no ATX Form Factor Standards for notebooks, and because us consumers demand thinner and lighter notebooks, notebooks are (1) extremely proprietary, and (2) difficult, at best, to open up for repair, or even essential cleaning of heat-trapping dust drawn in by the cooling fan(s).

But I have to admit I am a bit confused here. By definition, in electronics (and for 100+ years), a device that converts DC voltage to AC voltage is called an "inverter". CCFLs require high AC voltages so they use inverters to convert DC to that high voltage AC. LED backlit monitors are low DC voltage devices and don't use "inverters". So I don't know what or why Dell is calling an inverter because from an electronics standpoint, LED backlit monitors don't use, or need one. LED monitors do, however, use a "driver board" and I suspect that is what Dell in referring to and sadly, "marketing weenies" were allowed to stick their nasty noses in where they don't belong and hosed up proper electronics terminologies - again! :mad7: :banghead:

For a good read, or the truly adventurous, there are CCFL to LED conversion kits to swap out CCFL backlit panels with LED backlit panels that include a LED "driver" board to replace the CCFL "inverter" boards.

According to the link you provided, your notebook came in several different versions, some with LED backlighting, some with CCFL backlighting. I would verify to make sure of what you have.

Otherwise, because it is getting along in years and throwing good money after bad is never a good idea, I would just connect an external monitor, keyboard and mouse to it, and call it desktop.

 

[Tekno Venus]

Hey Bill,

You will have to excuse my terminology here - I'm no hardware guy!

The thing I have against Dell is their use of plastic tabs and clips. We have two Dell's here - mine and the family one. My one was the worst - to get to the HDD you had to take the entire palm rest off which is held on by 20+ plastic tabs around the edge. There are a fair few scratches around the edge of the palm rest where I struggled to get it off. It's not exactly like Dell didn't have any space - it's no thin and light! Anyway, I digress.

As I said, I may have my terminology wrong - I'm not sure.

I originally ordered this: NEW GENUINE DELL STUDIO 1535 1536 1537 1735 1736 1737 CCFL LCD INVERTER 0P927C | eBay but upon taking the laptop I found it doesn't fit because it is the wrong inverter. So I returned it and ordered this one: NEW DELL STUDIO 1535 1536 1537 LED SCREEN INVERTER BOARD M465J 0M465J | eBay after looking at the model on the back of the circuit board. So I definitely have the LED backlit screen. So I would assume this is a driver board as you call it.

Otherwise, because it is getting along in years and throwing good money after bad is never a good idea, I would just connect an external monitor, keyboard and mouse to it, and call it desktop.

That's what we've been doing for about 4 months now (it's taken me a while to get round to doing it because of school/exams). The only monitor we have is old and small, and it's really not practical to keep doing that. Plus we have no space to set it up permanently, which means we have to set it up each time which is hassle.

Thanks for your help though Bill, I really do appreciate it!

Stephen

 

[Digerati]

You will have to excuse my terminology here

What I am saying is it is poor terminology by marketing weenies, not you, causing problems as they should, or have people that should know the correct terms.

Remember, most TVs in recent years have PC (standard D-Sub or DVI/HDMI) inputs too.

 

[Tekno Venus]

Do you know a way you could test the driver board/inverter without having to actually put the system back together again?

Since the laptop has to be fully disassembled to get to the board (unlike some where you can just pop the bezel of the screen), it takes a while to do and I'm scared about breaking the tabs if I keep opening it up. The more times I open it, the higher the chance of it not going back together again. I'm planning to take it apart once more to check I actually hooked the board up correctly and got all the connectors connected properly. Would there be a way to test the connections without having to reassemble it and then turn it on, i.e. with a multimeter? If you don't know of a way, don't worry - I just wondered.

Do you think a replacement screen would solve it? I can pick one up for ~£60, which is about the same price the nearest repair shop will charge for a diagnosis.

Thanks!

Stephen

 

[Digerati]

Do you know a way you could test the driver board/inverter without having to actually put the system back together again?

The problem is the makers do not publish detailed schematics and circuit diagrams. If they did, then you could see where to inject voltages to "light the circuit" and where to measure outputs, and what you can expect to see at those test points.

About all you can do with a multimeter with the board out of circuit is check for continuity through some of the devices and trace routes, if you can identify the device. But some devices may be read "open", until the proper voltage is applied so I would not hold out much hope there either.

If I had to guess, and if certain the board is the correct one, then my guess is a new screen would fix it. But I personally would not invest that much on a guess with a notebook that old.

You might check with that shop again and see if they credit that 60 quid diagnostics fee back to you if you allow them to do the labor. That can take some of the sting out of it, and then they are responsible for any collateral damage during disassembly/reassembly.

I know what you mean about breaking tabs. To shave a few pennies off assembly costs, these makers assume once "snapped" together, you will never need to open up the case again. At least I would like to think that is the reason and not to force users to give up and buy another notebook.

Sorry I am of no better help.

 

[Tekno Venus]

Do you know a way you could test the driver board/inverter without having to actually put the system back together again?

The problem is the makers do not publish detailed schematics and circuit diagrams. If they did, then you could see where to inject voltages to "light the circuit" and where to measure outputs, and what you can expect to see at those test points.

About all you can do with a multimeter with the board out of circuit is check for continuity through some of the devices and trace routes, if you can identify the device. But some devices may be read "open", until the proper voltage is applied so I would not hold out much hope there either.

That was my thought too, but nice to have it confirmed. Shame you can't get the schematics for products nowadays! Not like old days where motherboards looked like this: http://www.tomdalby.com/retro/images/48k_inside_i2.jpg and you could get a full circuit diagram for it: http://8bit.yarek.pl/computer/zx.48/2_sch.gif.

If I had to guess, and if certain the board is the correct one, then my guess is a new screen would fix it. But I personally would not invest that much on a guess with a notebook that old.

The model number of the board matches the one I took out and the seller stated it was taken out of a working laptop. I am willing to put the money for a screen if I knew it would work. That's the problem, because it's actually a decent machine otherwise.

You might check with that shop again and see if they credit that 60 quid diagnostics fee back to you if you allow them to do the labor. That can take some of the sting out of it, and then they are responsible for any collateral damage during disassembly/reassembly.

That's worth a shot, and perhaps speak to them about what I've tested and tried to perhaps save them some diagnostic time.

I know what you mean about breaking tabs. To shave a few pennies off assembly costs, these makers assume once "snapped" together, you will never need to open up the case again. At least I would like to think that is the reason and not to force users to give up and buy another notebook.

Exactly, and it's a real pain! With the increase in metal construction in laptops, the plastic tabs seem to be reducing. It seems some manufacturers are starting to make their devices a bit more repair friendly, although some definitely aren't!

Sorry I am of no better help.

You've been really helpful, there's very little you can do over the internet anyway when it comes to hardware :)

Stephen

 

[Digerati]

Not like old days

Right. But that was when motherboards cost many $100s, or more and it was actually economically feasible to expend several manhours troubleshooting and repairing. But today, with labor costs of $60 - $75 per hour, or more, and the cost of most motherboards $100 or less, it is typically cheaper to just buy new.

Another problem is that most people don't understand that, in most cases, replacing the faulty part is actually the easy and fast part. It is the troubleshooting process needed to find the actual fault (diagnosing the problem) that takes the technician's expertise, time, training and expensive test equipment.

When I first started repairing computers, to update the BIOS you replaced the BIOS chip! Then they got hi-tech and you used a high-intensity UV light to erase the old BIOS so you could "burn" in the new programming. I even went to a 40 hour class for "micro-miniature soldering techniques" to learn how to work on multi-layered circuit boards without destroying the boards or burning up the components while mounting. Now I cannot even remember the last time I plugged in my iron.

and perhaps speak to them about what I've tested and tried to perhaps save them some diagnostic time.

Yes, some are more than willing to "talk shop" with you, others are not. Perhaps if you go in to buy some RAM, a flash drive or maybe a case fan from them, that may loosen some lips.

It seems some manufacturers are starting to make their devices a bit more repair friendly, although some definitely aren't!

It seems to me the thinner and lighter the notebook, the harder it is to perform maintenance on. And with the move to handhelds, self-repairs will become even harder.

Long live the PC! :grin1:

there's very little you can do over the internet anyway when it comes to hardware :)

I wish everyone understood that repairing electronics is really a "hands on" process and in most cases, troubleshooting electronics involves eliminating all the other possibilities first, until all that is left is the problem child.

The good news is, you can still use the computer with an external monitor, keyboard and mouse. But of course, showing airport security they work is a real pain.

 

[Tekno Venus]

Right. But that was when motherboards cost many $100s, or more and it was actually economically feasible to expend several manhours troubleshooting and repairing. But today, with labor costs of $60 - $75 per hour, or more, and the cost of most motherboards $100 or less, it is typically cheaper to just buy new.

Yeah, exactly, but still - they are useful! That ZX Spectrum motherboard I linked to is a really interesting board to work on, I have 3 of them. Two work and the other doesn't work at all and has been salvaged for parts. The reason it's so interesting is due to a cost cutting measure of Sinclair who:

used faulty 64K chips (internally 2 X 32K). All the chips in the 32K bank of RAM had to have the same half of the 64K chips working. A link was fitted on the pcb in order to choose the first half or the second half. It was possible with a few logic chips for an experimenter to have access to the faulty 32K bank with all sorts of odd results possible!

At least I've never had to work on the Issue 1 version of the board, yuck! http://k1.spdns.de/Vintage/Sinclair... (orthogonal)/ZX Spectrum PCB issue 1 16K.jpg

I know I'm going a tad off topic, but it's my thread and I'm happy to have an interesting chat about hardware

Another problem is that most people don't understand that, in most cases, replacing the faulty part is actually the easy and fast part. It is the troubleshooting process needed to find the actual fault (diagnosing the problem) that takes the technician's expertise, time, training and expensive test equipment.

Bang on - if it was just a case of swapping hardware it'd be much easier!

When I first started repairing computers, to update the BIOS you replaced the BIOS chip! Then they got hi-tech and you used a high-intensity UV light to erase the old BIOS so you could "burn" in the new programming. I even went to a 40 hour class for "micro-miniature soldering techniques" to learn how to work on multi-layered circuit boards without destroying the boards or burning up the components while mounting. Now I cannot even remember the last time I plugged in my iron.

Same with the Spectrum, the ROM chip was non upgradable. Can't remember if it was a EPROM or not, but even if it was it was never used that way.

I'm actually really lucky when it comes to soldering and repair work, my Grandad has experience working in electronics and electrical engineering and my Aunt actually teaches how to solder for a living (HOME).

Yes, some are more than willing to "talk shop" with you, others are not. Perhaps if you go in to buy some RAM, a flash drive or maybe a case fan from them, that may loosen some lips.

That may be a good idea, I will have to see what I can do.

 

[niemiro]

Since this discussion is already going slightly off-topic, I wonder whether I could quickly drag it further so

Does anyone know anything about soldering? I'm normally quite good at soldering, and can reliably produce many good quality small joints. The trouble is, I've got one iron on which I can't. It's a temperature controlled iron with a well tinned small wedge shaped tip (I've always kept it nicely tinned, it's never been damaged) for which the solder sticks more to the iron than the board. I end up having to physically scrape the solder off the iron and onto the board, which obviously gives very poor joints. Why is it that this iron does this? I've tried a whole range of temperatures, but it didn't seem to help much. Very high temperatures were a bit better, but not perfect - but I really shouldn't need such high temperatures. Does it need a different solder, and if so, what sort? I currently use a savbit alloy containing both lead and rosin. Should I be using something else? Any recommendations or ideas?

Thank you so much.

 

[Digerati]

I know I'm going a tad off topic, but it's my thread and I'm happy to have an interesting chat about hardware

Just had a very bad experience at Malwarebytes forum about that very thing. In a thread I started about a problem I was having with MBAM, a mod and I were actively having a good, and civil, but OT discussion about wifi, RF, and security, or rather the lack of security with RF when one of admins suddenly locked it without any notice. I was not a happy camper, as they soon discovered. :shame2:

the ROM chip was non upgradable. Can't remember if it was a EPROM or not

Well, being a "read only" module that makes sense. But an "Erasable Programmable ROM" could have been reprogrammed - with the right equipment, and of course, new code.

for which the solder sticks more to the iron than the board

That "suggests" the lead to the part you were soldering was not hot enough yet. Savbit alloy solder is (or was when I worked in a real repair shop) normally used with small copper wire joints and copper plated boards because it does to erode copper tips, or the joints.

Lead-free rosin-core solder works if the board is preheated. But most home enthusiasts don't have a pre-heater. I use 63 / 37 lead solder most of the time. And of course, flux is very important too.

But most important is the strategic placement of tongue.

 

[Tekno Venus]

Just had a very bad experience at Malwarebytes forum about that very thing. In a thread I started about a problem I was having with MBAM, a mod and I were actively having a good, and civil, but OT discussion about wifi, RF, and security, or rather the lack of security with RF when one of admins suddenly locked it without any notice. I was not a happy camper, as they soon discovered. :shame2:.

No worries here, I find off topic chats can actually lead to really interesting discussions. Depending where it goes, I may split it into its own thread, but we shall see :)

 

[Digerati]

That was my thing. If the guy had a problem because it was going off topic, then IMO, he should have split the OT portion out instead of closing it without any notice. Especially since it was an active discussion with one of the Mods on the site. But for that matter, it was my thread and it still pertained to security, and it, like this thread, contained information that others might have found useful. Oh well. Now I am running this way OT.

@Richard - have you by chance ensured that iron really is getting hot enough? Savbit alloy solder does have a pretty high melting point.

 

[Digerati]

BTW, one of the best pieces of test equipment I have is an IR Thermometer. It is great for checking temps inside your computer. But also great to make sure the grill or frying pan is searing hot, and the grandkid's forehead is cool (avoiding eyes, of course), and the freezer is freezing. You could sure check your soldering iron tips with this too.

 

[Wrench97]

Great for finding the miss in a Diesel engine also, just measure the exhaust ports at the head the cold one is the one not firing :)