Something amazing! Keysight UXR 110GHz, 256GS/s, 10-bit Scope Teardown & Exp.

[Berni]

Yeah this sort of "2 phase" power is common in American residential buildings. It does give the benefit of reducing neutral return current since the 110V loads on each phase balance out somewhat. And it makes it easy to get 220V for the heavy loads like aircon and stoves.

Here in Europe its quite common for a residential building to have 3 phase and then simply distribute all the loads over the phases. Heavy loads such as ovens tend to have 3 phase input, but most loads not since even just a regular outlet can provide 3kW. Areas like mine with a lot of farms also find 3 phase useful as 3 phase motors are very common for driving various machinery (Some of the motors are really big beasts too)

[maxwell3e10]

Can someone get this oscilloscope and measure the power line phase shifts? 

[mastaplanna]

Ok, so I watch this sweet video, and I'm new to electronics, but getting the feel for it...the way I understand it, Platinum is the best conductor of electricity, if I'm correct the acquisition board looks to be milled from gold, why wouldn't they use platinum?  Whats the difference if it adds another 200K in cost?  I mean if I'm dropping 1.3 mil, I'll come up with another 200K or so for it...I'm just sayin

[Berni]

That is NOT a solid block of gold, its just a thin layer of gold that was electrochemicaly plated on. Such a big chunk of gold would cost a fortune and there is no point in it being solid. Nor is gold the best conductor of electricity. The most conductive metal is silver flowed really closely by copper(hence why we use copper for wires). Gold is the 3rd most conductive element (About 50% worse then copper) and is just slightly more conductive than aluminum.

Aluminum is easily machined into shape, is cheap and light so it ends up the material of choice for solid shielding blocks. While on a circuit board copper is the metal of choice because of its excellent conductivity and the ability to be easily soldered.

So why is it gold then? One reason, oxidation resistance.
Gold unlike most other metals really really does not want to oxidize even in very aggressive conditions. If you plate a thin layer of gold onto a metal object means you use very little of the expensive gold while it makes the metal object immune to oxidation. This is very helpful in anything that has to make a good electrical contact (oxides are not electrical conductive). Additionally gold loves being soldered as much as copper does so gold plated parts solder really well (Even more so due to no oxides). The oxidation resistance also makes gold easy to cold weld, just smoosh two objects of gold together hard enough under the right conditions and they spontanusly stick together without any heat required(This is why chip bond wires are gold)

So why not platinum? It has even better oxidation resistance and can also be soldered to copper. Well.. the electrical resistance of platinum is 6 times higher than copper and it costs a lot more than gold. So you would be paying a lot more for a coating that's just as good as gold or possibly even worse than gold.

[mastaplanna]

I don't disagree it could or is gold plating only...but here's my thing...an ounce of gold is $1200...that gold module would only cost, at most 20K...that thing wouldn't be more than 15-20 ounces and thats being generous...Platinum is only $860 an ounce...it seems that in a $1.3 million machine, this is the one to try it on...I still don't see a good argument against it...I'd build a machine out of each metal and see how it performs, how do you know unless you do?

When Sharier picked it up, how much did it weigh?

And, we use copper because its cheap

[Berni]

Yes exactly we use copper because its cheap while being almost as good as silver. The extra cost of silver is not worth the small performance advantage.

Just because it is expensive does not mean it is good. To make a good shielding box all you need is something that is highly electrically conductive, but since you need to somehow get your electronics in there means you have to have a lid on your box, so you want to have no oxides between the lid and the box so that you maintain good electrical conductivity all the way around the lid. Gold is perfect for keeping oxides away and its convenient to be able to solder your electronics on to it. Additionally high speed electronics generate heat so you likely also want the box to be thermally conductive to be able to easily get the heat out. The most thermally conductive metal is also silver so the ideal box would be a solid silver box with gold plating. But again copper is almost the same performance so might as well use the cheaper copper.

It is possible the box is made out of solid copper inside as i have seen it used here or there in RF boxes.

Oh and the scope does not cost anywhere near 1.3 Milion USD to produce. It probably costs them somewhere in the range of $50 000 to produce a unit. The rest of the cost is tooling (ASIC masks, injection molding, assembly setup, test systems) and an even larger part is R&D cost. Many milions of dollars went into paying a large team highly skilled engineers and some more milions went into building prototypes. Since they are sure as heck not going to sell 10 000 of these scopes means they have to sell for a high price to pay back the development process. Also they need to turn a profit in the end, they are a company after all.

[Fungus]

I don't disagree it could or is gold plating only...but here's my thing...an ounce of gold is $1200...that gold module would only cost, at most 20K...that thing wouldn't be more than 15-20 ounces and thats being generous...Platinum is only $860 an ounce...

And silver is only $15 an ounce.

Gold-plated silver would be the way to go.

[TheUnnamedNewbie]

Oh and the scope does not cost anywhere near 1.3 Milion USD to produce. It probably costs them somewhere in the range of $50 000 to produce a unit. The rest of the cost is tooling (ASIC masks, injection molding, assembly setup, test systems) and an even larger part is R&D cost. Many milions of dollars went into paying a large team highly skilled engineers and some more milions went into building prototypes.

All of those things are what it costs to produce this scope. There is a difference in 'BOM cost' and 'cost produce a device'. The first is just components, the latter includes every cost you made to get it to market. (how exactly you distribute fixed/one-time costs is part of what makes good financial and bookkeeping staff so valuable)

And, we use copper because its cheap

Copper is also stronger and iirc behaves better when alloyed (in terms of conductivity). Copper is also a better conductor. Coating things with gold has nothing to do with electrical performance and everything to do with chemical - gold does not form an oxide, it does not tarnish, ... You might find thick layers of gold though because you can't just coat copper with gold, gold with diffuse into copper. You instead need to first plate it with nickel or nickel and paladium (ENIG or ENEPIG). Using thicker layers ensures you really use the conductivity of gold.

[mastaplanna]

Anyone have one of these yet?

[Keysight DanielBogdanoff]

Anyone have one of these yet?

I do! 

[nctnico]

Anyone have one of these yet?

I do! 

Don't put it on the BBQ!

[Keysight DanielBogdanoff]

Haha, no. Those days are over, I think. Even with a scope that was already broken, I took so much flack for that video

[TheSteve]

Haha, no. Those days are over, I think. Even with a scope that was already broken, I took so much flack for that video

I could have fixed it...

[bitseeker]

Yeah, if Keysight would like to pass out broken scopes or other test equipment, we're all for it. I'll PM Daniel my address.

[mastaplanna]