TR Forums

notfred wrote:

just brew it! wrote:

mikewinddale wrote:

Wow, Superspy, I didn't know that manual impact screwdrivers were a thing. I mean, I had literally never heard of them. Awesome, thanks!

Ditto. Might need to get myself one of those.

You hit the end with a hammer which drives the bit down in the screw head and also torques the screw to undo it. Mainly used on automotive stuff, I wouldn't be using something like that on anything like computers!

I agree. They do make hammer-less versions of those that work more like an automatic punch (where you just press on the screwdriver and it will suddenly slip and torque) that I presume are much less violent, but I couldn't find an example of one to link.

Hmm, with an impact screwdriver, can I get a little bit of torque out of it by tapping it lightly? Or is it more of an all-or-nothing thing? Like, hit it hard and you dislodge a rusted automotive screw, but if you hit it lightly, nothing happens?

Anyway, I'll first try the ratcheting wrench screwdriver, the soldering iron, and the WD-40 before I try an impact driver.

Well, the better precision bits didn't help. I almost stripped the screw trying, so I'd better stop before I completely strip it.

WD-40 didn't help either.

So I guess now I'll try the soldering iron.

Although one of the troublesome screws is what attaches the internal battery to the motherboard, preventing me from flipping the motherboard over to add extra RAM. I'm pretty sure I shouldn't try heating up that screw to 250C. So I don't know what I'll do about that.

Maybe try turning it a bit clockwise (tighter) to break the Locktite? As you seem to be approaching the point where you will need to drill the head off the screw. Although little bits of screw head doesn't sound like a good thing inside a laptop.

Assuming the battery housing is plastic, a very small point soldering iron tip should concentrate the heat in the screw and plastic is not good heat conductor.

mikewinddale wrote:

Hmm, with an impact screwdriver, can I get a little bit of torque out of it by tapping it lightly? Or is it more of an all-or-nothing thing? Like, hit it hard and you dislodge a rusted automotive screw, but if you hit it lightly, nothing happens?

Anyway, I'll first try the ratcheting wrench screwdriver, the soldering iron, and the WD-40 before I try an impact driver.

I'm not 100% sure on their internals, but I think there's a minimum force you have to strike before the rotating mechanism happens (they have to 'fall' past a certain point before they touch the rotational section).

mikewinddale wrote:

Although one of the troublesome screws is what attaches the internal battery to the motherboard, preventing me from flipping the motherboard over to add extra RAM. I'm pretty sure I shouldn't try heating up that screw to 250C. So I don't know what I'll do about that.

You really don't want to apply the iron very long anyway. The idea isn't to melt the thing out, it's to heat cycle it one or more times so that thermal expansion and contraction breaks up any corrosion or adhesive and allows the threads to turn more freely. For something the size of a laptop screw, a couple seconds on and 10-20 seconds off should do the trick. Repeat if necessary.

farmpuma wrote:

Assuming the battery housing is plastic, a very small point soldering iron tip should concentrate the heat in the screw and plastic is not good heat conductor.

Good to know, thanks.

ludi wrote:

You really don't want to apply the iron very long anyway. The idea isn't to melt the thing out, it's to heat cycle it one or more times so that thermal expansion and contraction breaks up any corrosion or adhesive and allows the threads to turn more freely. For something the size of a laptop screw, a couple seconds on and 10-20 seconds off should do the trick. Repeat if necessary.

True. However, I checked the specs for blue loctite, and the manufacturer says it breaks at 250C. So I figured that regardless of how long I hold the soldering iron to the screw, if my goal is to heat the screw up to 250C, then I threaten to heat up part of the lithium battery to 250C as well.

So I'll look closer and see how thick the plastic housing around the lithium battery is.

farmpuma wrote:

Assuming the battery housing is plastic, a very small point soldering iron tip should concentrate the heat in the screw and plastic is not good heat conductor.

Good to know, thanks.

ludi wrote:

You really don't want to apply the iron very long anyway. The idea isn't to melt the thing out, it's to heat cycle it one or more times so that thermal expansion and contraction breaks up any corrosion or adhesive and allows the threads to turn more freely. For something the size of a laptop screw, a couple seconds on and 10-20 seconds off should do the trick. Repeat if necessary.

True. However, I checked the specs for blue loctite, and the manufacturer says it breaks at 250C. So I figured that regardless of how long I hold the soldering iron to the screw, if my goal is to heat the screw up to 250C, then I threaten to heat up part of the lithium battery to 250C as well.

Is that not my goal? Can I sufficiently weaken the loctite at a temperature below 250C even if the manufacturer says it breaks at 250C?

Anyway, I'll look closer and see how thick the plastic housing around the lithium battery is.

I did use a stripped screw remover on a desktop machine once. The screw was not actually stripped I just couldn't get it to turn.
Dad gave me his screw remover and the electric drill and said "This will work" IT SURE DID!

mikewinddale wrote:

Is that not my goal? Can I sufficiently weaken the loctite at a temperature below 250C even if the manufacturer says it breaks at 250C?

It's likely going to soften significantly long before it reaches its 250C 'breaking' (probably melting) point. So even getting it around 100C will likely help a lot.

I think there's a kind of penetrating oil that has an effect on Loctite; at least, I recall somebody telling me that once. You'd definitely want to make sure it's a non-conductive penetrating oil (that stuff tends to find its way everywhere).

Well, I'm trying WD-40 for electronics. But I'm not sure if it's actually getting into the screw hole or not. That screw is in there damned tight.

mikewinddale wrote:

Well, I'm trying WD-40 for electronics. But I'm not sure if it's actually getting into the screw hole or not. That screw is in there damned tight.

Drill off the head of the screw, then take vice grips to the remaining shaft and twist it out. Probably a little harder to do on a tiny screw, but its a last resort kind of action anyway.

--SS

If you have access to a dremel with a cutting wheel (or just a file) you might be better off trying to cut the 'sides' off the head to create a flat spot (so the screw now looks like a T from the side, with the former Philips head on top) the vice grips can grab. With a screw that small you're going to need as much material as possible to grab.

Thanks everyone.

So I just got the soldering iron in the mail today, but it hasn't helped. The mechanical hard drive is still stuck to the caddy. I think someone at the factory screwed it too tight. I don't think the problem is the fastener fluid.

On the plus side, the new screwdrivers (with better leverage) were enough - by themselves - to loosen all the screws attaching the motherboard to the battery and to the chassis. So I was able to undo all those screws, flip the motherboard over, and upgrade the RAM. I just replaced 4 GB of PC3L-12800 CAS 11, with 8 GB of CAS 9. I screwed everything back together, booted up the computer, and everything is recognized.

So the SSD inside the laptop is still just sitting there, attached to the SATA port but not screwed to anything. I'm using scotch tape to hold it still. (I taped the SSD to the bare metal where the screws go.) I just ordered a new caddy on Ebay, so I'll just use that to attach the SSD.

I've still got this 400 GB mechanical hard drive stuck to its caddy. I was hoping to put it inside a USB chassis and use it as external storage, but maybe I'll instead just use one of those USB-SATA cables and let the drive sit out in the elements, without a chassis. Or maybe I'll take it up to Home Depot and let someone with some real tool skills try to get those screws out. At this point, it has passed my own competence. Someone is going to have to use a drill or something.

I had a bit of a scare when I tried to boot up my VeraCrypt-encrypted Windows partition, and it said it couldn't find the image. Then I tried to boot up my ecryptfs encrypted Linux partition, and it did fine. So that told me the drive wasn't damaged. So I rebooted, and this time, instead of selecting VeraCrypt from GRUB as I usually do, I selected VeraCrypt from the BIOS "esc" boot menu. This time it gave me a more detailed error message, telling me that it couldn't boot because it wasn't a secure image. So I realized that when I replaced the RAM, the system must have re-enabled secure boot. I guess any change to the hardware makes it re-enable secure boot? Anyway, I went into the BIOS, disabled secure boot, and then VeraCrypt Windows was able to boot again. So everything works!

Anyway, thanks everyone. Even though the soldering iron didn't help, it wasn't a bad purchase. For $30, I got a kit that came with not only a soldering iron, but also a pocket screwdriver with precision bits, a digital multimeter, tweezers, and a wire stripper. Not a bad purchase. So now I have several sets of precision screwdrivers.

I wish I had run more benchmarks before upgrading the RAM. But it seems that 500 MB of VeraCrypt encryption has sped up from 250 MB/s to 300 MB/s!! So just reducing the RAM's CAS from 11 to 9 has made a huge difference!

(This is an old processor, so it doesn't support AES-NI. This is a Celeron N2930, and I compared it to my other laptop, an AMD A10-5750M. The AMD processor is about 1.333 times faster than the Celeron in most benchmarks, but it gets 1.6 GB/s in VeraCrypt because it supports AES-NI.)