Laptop fan not working

[H0R53]

I have a very old Dell Latitude D620. The fan itself works, I tested this by stripping the wires and plugging it in to the USB. However it doesn't work when plugged into the fan header on the D620 motherboard. Should I do a reflow or just deal with it? It doesn't get noticeably hot unless running some OpenGL games.

[RadiatingLight]

it might be designed to turn off at idle. does it turn on when playing games?

[manikyath]

48 minutes ago, H0R53 said:

Should I do a reflow

under any scenario, you should *NEVER* do a reflow... it's like hittig the side of your CRT to get the image to display properly. it may fix the symptoms, but the actual problem remains.

[H0R53]

2 hours ago, manikyath said:

under any scenario, you should *NEVER* do a reflow... it's like hittig the side of your CRT to get the image to display properly. it may fix the symptoms, but the actual problem remains.

I had a D830 that I reflowed the nVidia chip in and it's been my mobile workhorse for years now. Your analogy is inadequate and partly incorrect.

3 hours ago, RadiatingLight said:

it might be designed to turn off at idle. does it turn on when playing games?

It never turns on.

[manikyath]

1 minute ago, H0R53 said:

I had a D830 that I reflowed the nVidia chip in and it's been my mobile workhorse for years now. Your analogy is inadequate and partly incorrect.

or you have been stupidly lucky that the problem never came back.

[H0R53]

3 minutes ago, manikyath said:

or you have been stupidly lucky that the problem never came back.

There was an HP with a flickering display problem I traced back to the GPU, reflowed that thing and customer hasn't complained.

[H0R53]

I installed Speedfan and it works. I don't know why but Windows 10 doesn't use the fan at all.

[manikyath]

1 hour ago, H0R53 said:

There was an HP with a flickering display problem I traced back to the GPU, reflowed that thing and customer hasn't complained.

i fixed a mouse by strapping it in front of a space heater for half an hour, does that mean the problem has been solved, or does that mean that the heat just made the symptom go away?

[H0R53]

13 hours ago, manikyath said:

i fixed a mouse by strapping it in front of a space heater for half an hour, does that mean the problem has been solved, or does that mean that the heat just made the symptom go away?

You won't know until you take it apart and diagnose it properly. Old laptops like this have very old solder balls that crack and separate from their components. A reflow often fixes this, alleviating the problem. However, reflows don't solve everything. I asked because if you knew your shit, which you obviously don't, the D620 is a 10 year old laptop and at this point in time the components may have already started to separate from the motherboard.

 

If you've got a headache I can knock you out. That may stop your complaining, but when you wake up you'll be bitching again.

 

Reflows are not "the ultimate solution", but Linus, the God of Computers himself, reflowed a graphics card in a video he put on YouTube.

 

Maybe you should do some research before blindly answering a question on a topic you know absolutely nothing about.

[manikyath]

2 hours ago, H0R53 said:

You won't know until you take it apart and diagnose it properly. Old laptops like this have very old solder balls that crack and separate from their components. A reflow often fixes this, alleviating the problem. However, reflows don't solve everything. I asked because if you knew your shit, which you obviously don't, the D620 is a 10 year old laptop and at this point in time the components may have already started to separate from the motherboard.

 

If you've got a headache I can knock you out. That may stop your complaining, but when you wake up you'll be bitching again.

 

Reflows are not "the ultimate solution", but Linus, the God of Computers himself, reflowed a graphics card in a video he put on YouTube.

 

Maybe you should do some research before blindly answering a question on a topic you know absolutely nothing about.

you know those macbooks with bad GPU's they "reflow" to get working?

 

yeah... that's not solderballs, that's a capacitor next to the GPU that goes bad, and heat "fixes it".

 

in fact, linus's GPU "reflow" most likely didnt even get hot enough to melt any solder, so how do you then explain that it magically fixes solder balls if it didnt even melt them?

[H0R53]

3 hours ago, manikyath said:

you know those macbooks with bad GPU's they "reflow" to get working?

 

yeah... that's not solderballs, that's a capacitor next to the GPU that goes bad, and heat "fixes it".

 

in fact, linus's GPU "reflow" most likely didnt even get hot enough to melt any solder, so how do you then explain that it magically fixes solder balls if it didnt even melt them?

450F is plenty to melt solderballs. Electrolytic capacitors cannot be fixed with heat, they use a unique electrolytic solution that actually gets ruined the hotter it gets. There's something called 'research', and you should do it. The melting point of lead-free solder is 188C, or 370F. I should also add that the toaster oven in my kitchen can reach 500F and it cost me $35. Keep digging and you'll pop out in China. You see the smug face in my avatar? That's there for a reason

[manikyath]

1 minute ago, H0R53 said:

There's something called 'research', and you should do it.

i'd rather say you're the one that should do said research.

2 minutes ago, H0R53 said:

Electrolytic capacitors

did i ever speak about electrolytics?

2 minutes ago, H0R53 said:

The melting point of lead-free solder is 188C, or 370F.

sure it is, but that doesnt mean an oven that can reach that temperature can magically "fix" broken joints. if i put my soldering iron to that temperature i'm burning up chips before i even get solder to melt.

 

there's two issues here:

- you need to heat *FAST* in order to not kill the chip while soldering, most ovens i know of arent capable of this, and in fact the video you referred to earlier is a prime example of "the solder didnt even melt", if this created a "fix", this is the good old "heating up a cap made it sorta work again" thing.

- if you so magically manage to melt said solder, what you just did is called a "cold joint", where essentially there's not enough flux for the solder to flow properly, and as a result, the broken ball may be fixed, but it's barely connected at all, essentially waiting to break again in the exact same spot.

 

in either of these cases, all you do is buy time, that time may be a few months, a few years, or a few days. what you do is make the symptom go away, but the underlying problem is still there (either a bad cap, or a bad solder joint) waiting to return, in the exact same symptops as before. there's plenty of folks on the forum who bake their old cards to keep them alive, and plenty of them will tell you that it is recurring maintenance rather than a permanent fix.

 

with that, you should understand the amount of disrespect i gain towards you when you so boldly claim you do this to the machines of (paying?) customers:

20 hours ago, H0R53 said:

and customer hasn't complained.

 

[H0R53]

57 minutes ago, manikyath said:

i'd rather say you're the one that should do said research.

did i ever speak about electrolytics?

sure it is, but that doesnt mean an oven that can reach that temperature can magically "fix" broken joints. if i put my soldering iron to that temperature i'm burning up chips before i even get solder to melt.

 

there's two issues here:

- you need to heat *FAST* in order to not kill the chip while soldering, most ovens i know of arent capable of this, and in fact the video you referred to earlier is a prime example of "the solder didnt even melt", if this created a "fix", this is the good old "heating up a cap made it sorta work again" thing.

- if you so magically manage to melt said solder, what you just did is called a "cold joint", where essentially there's not enough flux for the solder to flow properly, and as a result, the broken ball may be fixed, but it's barely connected at all, essentially waiting to break again in the exact same spot.

 

in either of these cases, all you do is buy time, that time may be a few months, a few years, or a few days. what you do is make the symptom go away, but the underlying problem is still there (either a bad cap, or a bad solder joint) waiting to return, in the exact same symptops as before. there's plenty of folks on the forum who bake their old cards to keep them alive, and plenty of them will tell you that it is recurring maintenance rather than a permanent fix.

 

with that, you should understand the amount of disrespect i gain towards you when you so boldly claim you do this to the machines of (paying?) customers:

 

Yes, but I didn't use the oven method for that, I used an expensive heat gun I borrowed from my old boss's shop, where I usually get most of my tools. I had a D820 that had a Quadro NVS 120m, caused all sorts of problems, and I smacked it with a heat gun and it hasn't malfunctioned, as a matter of fact I sold it and I keep regular contact with the woman I sold it to and she says it runs great. There are many methods of reflowing components (you'd know if you did the little 'research' thing I mentioned earlier).

 

You didn't mention electrolytic capacitors but the majority of small electronics capacitors are electrolytic, such as those in graphics card power circuits because the PSU usually handles the actual delivery, especially on laptops like MacBooks. A quick Google search on capacitor types would reveal this information. Tiny capacitors are usually ceramic and reflowing does absolutely nothing to them. The capacitors in power supplies are usually film caps, and used because of the comparatively low heat overhead a power supply generates and are responsible for the efficiency coefficient in regards to operation temperatures, such as the variance of output efficiency in this power supply. Again, please do research before writing your next reply.

 

You must be in China by now, that hole you're digging yourself keeps getting deeper!

[manikyath]

2 minutes ago, H0R53 said:

You didn't mention electrolytic capacitors but the majority of small electronics capacitors are electrolytic,

 

You must be in China by now, that hole you're digging yourself keeps getting deeper!

i never said it's the electrolytics that are failing  there's plenty of other types under the hood of a computer, you just keep pushing on trying to validate a method that's no better than percussive maintenance.

 

also, i'm trying hard to keep away from the personal insults, but you seemingly seem to only be worried about attacking my person, rather than my knowledge, seeing you've yet to give any sort of explanation about what makes me wrong, instead you seeminly have no problem with telling me what an idiot i must be. which in case you havent noticed yet, is not working.

[H0R53]

38 minutes ago, manikyath said:

i never said it's the electrolytics that are failing  there's plenty of other types under the hood of a computer, you just keep pushing on trying to validate a method that's no better than percussive maintenance.

 

also, i'm trying hard to keep away from the personal insults, but you seemingly seem to only be worried about attacking my person, rather than my knowledge, seeing you've yet to give any sort of explanation about what makes me wrong, instead you seeminly have no problem with telling me what an idiot i must be. which in case you havent noticed yet, is not working.

"trying hard to keep away from personal insults"

I think what I'm doing is working if you have to try to not insult me.

 

Reflowing is infinitely better than 'percussive maintenance', with the latter you have a much higher chance of actually damaging your product. Reflowing is generally targeted at specific components and if it doesn't work doesn't hurt anything if done properly.

[manikyath]

20 minutes ago, H0R53 said:

"trying hard to keep away from personal insults"

I think what I'm doing is working if you have to try to not insult me.

or rather, it's human nature to want to drop to that level if somoene is preaching things so stupid you're not even sure where to start on proving them wrong.

 

21 minutes ago, H0R53 said:

Reflowing is infinitely better than 'percussive maintenance', with the latter you have a much higher chance of actually damaging your product. Reflowing is generally targeted at specific components and if it doesn't work doesn't hurt anything if done properly.

reflowing is the 21st century equivalent of hitting the side of a CRT to make it display properly again, you are doing something that gets rid of the symptom (for example, the display crapping out when you move it), instead of actually fixing what is causing it (for example, a loose bolt)

 

and you are seriously in denial if you're gonna say "targeted at specific components" in a discution where you just used linus's oven baked GPU as an example to prove your right.

 

besides, this point still stands:

2 hours ago, manikyath said:

there's two issues here:

- you need to heat *FAST* in order to not kill the chip while soldering, most ovens i know of arent capable of this, and in fact the video you referred to earlier is a prime example of "the solder didnt even melt", if this created a "fix", this is the good old "heating up a cap made it sorta work again" thing.

- if you so magically manage to melt said solder, what you just did is called a "cold joint", where essentially there's not enough flux for the solder to flow properly, and as a result, the broken ball may be fixed, but it's barely connected at all, essentially waiting to break again in the exact same spot.

and until you can give me a valid explanation about how you're somehow so "skilled" (at least, that's what you're claiming) you can avoid these two issues reliable enough to turn it into a buisiness, i'm not taking any of your "oh you should do research" crap anymore, because in all honesty, the only one who's digging a hole here is you, and you're either too self centered or close minded to admit it.

 

the truth of the matter is i just came in to tell you reflowing isnt a good idea (and for goodness sake who in their right mind would reflow a laptop where only the fan inst working properly, without doing any additional troubleshooting first) and you could have easily just ignored that honest warning if you were so set on your case, but instead you ended up trying to prove your right to someone who may just know his shit slightly better than you do, and maybe had a field of education that teaches this crap.

[H0R53]

1. To reflow a laptop GPU, find some heat shields (quarters work well, and even better if you wrap them with kapton tape) and cover the area surrounding the component you want to reflow.

2. Start your heat gun and let it heat up all the way before reflowing.

3. You want to get the heat gun as close to the component as possible without actually burning it by getting too close. You want to have the heat gun at your targeted area for no longer than 5 minutes.

4. If you preheat an oven or toaster oven to 450F, you can pop the entire board in for 5-7 minutes, and components should be fine. Keep in mind if it's in too long you risk permanently damaging your board and other components that are not designed to deal with this level of heat.

5. Let the entire board cool down, and reassemble to test. Do not reflow a board more than twice as repeated reflows can lead to catastrophic failures.

 

Properly done reflows are done for specific reasons. Hitting things is done because you don't know what's wrong and you're too lazy (or too stupid) to bother to fix it.