• Announcements

    • Robin

      Welcome!   03/05/2016

      Welcome, everyone, to the new 910CMX Community Forums. I'm still working on getting them running, so things may change.  If you're a 910 Comic creator and need your forum recreated, let me know and I'll get on it right away.  I'll do my best to make this new place as fun as the last one!
Sign in to follow this  
Scotty

Story, Friday June 30, 2017

Recommended Posts

On 7/3/2017 at 5:43 PM, Pharaoh RutinTutin said:

Hasn't this sentiment been applied to many subjects over the millennia including, but not limited to, wedding rings, dental crowns, coins, capital reserves, and enemies of the Green Lantern Corps?

I looked around a bit and it turns out that you are correct. Gold was the material of choice a few years back, but for a lot of circuits they are not using copper and for the cheapest possible even aluminium. Gold remains the choice for high power and high reliability solutions. Copper is actually a slightly better conductor and the bond wires can be made even thinner than with gold, but it's also more likely to break due to metal fatigue.

Gold has a weakness in that if bonded to an aluminium surface it may develop something called "purple plague". This is a intermetallic compound of aluminium and gold that's a bad conductor and mechanically weak. This occurs when the joint is heated to more than 650°C. At lower temperatures in the range of 400-450°C other compounds can form.

Share this post


Link to post
Share on other sites
On 7/4/2017 at 0:44 AM, hkmaly said:

I can assure you I have chips around - not CPUs, smaller chips of course - which have contacts totally not looking gold. So, unless they put additional effort in recoloring it ... (on the other hand, I suspect that several those who DO look gold are actually copper).

I wasn't talking about the visible parts of the chips. It used to be that all integrated circuits used gold wire to connect the chip to the external connections on the packaging. Here is a picture showing what I mean.

aciSjj0.jpg

I have however looked around a bit and apparently copper is often used now, and for low power and cheap circuits they can even use aluminium.

Old processors using a Pin Grid Array package had pins made of gold. These were soft and prone to bend when inserted into the sockets, but didn't usually break unless you wiggled them around. Later versions moved to Ball Grid Array, and on these the contact surfaces are gold plated. The "pins" in the socket are made of a bronze alloy which is also gold plated.

Gold is used because it doesn't corrode, it's soft and malleable and a decent conductor.

Share this post


Link to post
Share on other sites
34 minutes ago, Cpt. Obvious said:

Copper is actually a slightly better conductor and the bond wires can be made even thinner than with gold, but it's also more likely to break due to metal fatigue.

You can make thinner wires with gold, but it must be pure gold, not just gold-coated. Unless you talk about diamond coated copper.

Share this post


Link to post
Share on other sites
4 minutes ago, hkmaly said:

You can make thinner wires with gold, but it must be pure gold, not just gold-coated. Unless you talk about diamond coated copper.

You can make thinner wires out of gold, but as gold isn't as good a conductor they have to be thicker than the equivalent copper wire. And thicker means more expensive, so unless corrosion resistance is very important, or it needs to withstand a lot of vibrations copper is often the preferred material. For common components vibration is not a problem, thermal expansion can be a problem, but in low power electronics it's not a big problem and copper is usually good enough.

Share this post


Link to post
Share on other sites
1 minute ago, Cpt. Obvious said:
14 minutes ago, hkmaly said:

You can make thinner wires with gold, but it must be pure gold, not just gold-coated. Unless you talk about diamond coated copper.

You can make thinner wires out of gold, but as gold isn't as good a conductor they have to be thicker than the equivalent copper wire. And thicker means more expensive, so unless corrosion resistance is very important, or it needs to withstand a lot of vibrations copper is often the preferred material. For common components vibration is not a problem, thermal expansion can be a problem, but in low power electronics it's not a big problem and copper is usually good enough.

Yeah ... didn't though about if those thin gold wires would conduct good enough.

Speaking about it, that diamond coated copper wire might not conduct good enough either.

Share this post


Link to post
Share on other sites
7 minutes ago, hkmaly said:

Yeah ... didn't though about if those thin gold wires would conduct good enough.

Speaking about it, that diamond coated copper wire might not conduct good enough either.

Hadn't heard about the diamond coated wire. That article talks about using it in processors so I think it will conduct, but it's not for bonding chips but rather for connecting individual components on the chip. Compared to the internal signals on a chip the external signals are quite "crude", using relatively high voltages and currents.

Share this post


Link to post
Share on other sites
1 hour ago, Cpt. Obvious said:
1 hour ago, hkmaly said:

Speaking about it, that diamond coated copper wire might not conduct good enough either.

Hadn't heard about the diamond coated wire. That article talks about using it in processors so I think it will conduct, but it's not for bonding chips but rather for connecting individual components on the chip. Compared to the internal signals on a chip the external signals are quite "crude", using relatively high voltages and currents.

I just found that article searching for something about the thinnest wire. But yes, it might be usable as logic/signal wire, but half of pins for modern CPUs are power supply and grounding and while partially it's to get current to all necessary places mostly I think it's because fewer wouldn't be able to transfer the current needed. After all, from physics point of view, CPU is device for turning electric energy into heat very similar to electric kettles.

Share this post


Link to post
Share on other sites
7 hours ago, hkmaly said:

I just found that article searching for something about the thinnest wire. But yes, it might be usable as logic/signal wire, but half of pins for modern CPUs are power supply and grounding and while partially it's to get current to all necessary places mostly I think it's because fewer wouldn't be able to transfer the current needed. After all, from physics point of view, CPU is device for turning electric energy into heat very similar to electric kettles.

In other words, we need bollocking high temperature superconductors that will work at room temperature.

Share this post


Link to post
Share on other sites
2 hours ago, The Old Hack said:

In other words, we need bollocking high temperature superconductors that will work at room temperature.

That wouldn't work.

Basically, there are places in CPU where you NEED to disappear the energy which brought the signal. NAND gate, for example. Obviously, there are non-deliberate energy loses as well, but I wouldn't be sure which are bigger.

Also, the temperature of CPU core is not "room", it's "boiling water" ... or "boiling oil" in case of GPU.

Although ... if we WOULD build CPU on room-temperature superconductors, it would likely require completely different construction which MIGHT have less thermal loses ...

54 minutes ago, ProfessorTomoe said:

I want my superconducting toaster!

Superconducting toaster will leave the bread cold.

Share this post


Link to post
Share on other sites
1 minute ago, hkmaly said:
56 minutes ago, ProfessorTomoe said:

I want my superconducting toaster!

Superconducting toaster will leave the bread cold.

Even a high-temperature one? A really, really high-temperature one?

Share this post


Link to post
Share on other sites
1 minute ago, ProfessorTomoe said:

Even a high-temperature one? A really, really high-temperature one?

Well ... actually, I'm not sure ... maybe if you would put enough current into the superconductor and it would be bent around the bread it would warm it with induction alone ... but the principle of superconductor is that it will NOT get hot when you put current into it, while principle of toaster and/or electric kettle is that it WILL get hot when you put current into it.

Share this post


Link to post
Share on other sites
6 hours ago, Pharaoh RutinTutin said:

I think what you want is actually a Super colliding toaster.  Something to make the marmalade impact the bread at .999c.

You really CAN'T make something impact something else at .999c unless you speak about particles. At that speed, the bonds holding the projectile together are completely irrelevant and it moves so fast it can fly several meters before it starts fusing. No way something as small and sparse as bread would stop it.

And while the idea of using relativistic speeds (just little smaller) to get marmalade INSIDE the bread is appealing, you need to remember that when the atoms DO hit they FUSE, meaning it won't be marmalade (or bread) anymore (probably will be mostly consisting of magnesium). Also, the energy will sublimate both the bread and marmalade.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this