Craig Sawyers

But, but - they are upside down! The bases will come out with the wrong phase

Aged 66. Too damned young. RIP Frank Murray.

Have a great one!

I completely agree that a chip, or hybrid solution would be best. In fact the concept of using a laser trimmed metal foil R2R network, where the foil network was layed down by photolith ought to be superb, hermetically sealed and dry nitrogen filled. Cheap it would not be...

Unless my use of Google is in meltdown, the PCM1704 is obsolete. Is this wrong? For discrete R2R, assuming that matching is done correctly, the whole resistor chain needs to be kept isothermal. There is little evidence that this is the case, but could be arranged to be the case.

There is absolutely nothing to prevent laser trimming, or selection to any tolerance you like provided that temperature and voltage coefficients are low enough. 0.01% tolerance is 10ppm or 1 ohm error in 10k. Temperature coefficient is around 3ppm/C so to match tolerances at a 10ppm level the measurement temperature has to be constant to 1C or thereabout. Foil resistors are superb, but are punishingly expensive. And any SM component has to be soldered using precisely the soldering profile in the datasheet for the parts - because component stress and thermal shock will shift the resistance too. However you look at it, it isn't trivial, is all I am saying.

It is all down to how the tolerance is statistically distributed. You can get two different scenarios. (i) All resistors in any production batch are clustered in value or (ii) resistance values in a batch are randomly distributed. This could only be established by measurement. However, if (i) holds then the batch will have a constant error and if (ii) holds then the values will be normally distributed. Typically the nominal tolerance will be 3-sigma, which would mean that less than 0.27% of resistors were out of tolerance. So the majority of resistors will be significantly better than nominal tolerance. The difficulty in getting accuracy is that you need to test a large number of resistors. If 100 resistors are tested the nominal error is 10%. To get 1% confidence you would need to test 10,000 resistors - er, no thanks at that level!

You always crack me up

I see your fat German and up you a skinny geriatric Brit - the impeccable Ginger Baker, superbly recorded in 2005 in an 8 minute solo:

Now up to at least 31 dead RIP, with many survivors with up to 90% burns. Agree with Dusty - the youtube coverage is horrendous.

No - I've never been an academic. I did my first degree in electronics and then a doctorate in laser physics. But my career was first in professional consultancies and then in industry (including 7 years at Oxford Instruments - which is why we live here). And for the last 20 years as a self employed technology and business/management consultant. In fairness in my mid 20's I applied for a "new blood" lectureship at St Andrews University in Scotland. That was the strangest interview I have ever had, and decided academia was not a good fit for me (from my perspective didn't pay nearly enough, no prospects etc), and we parted company. Plus the clowns approached the CEO of the tiny 20 person company I worked for, for a reference without asking me first. So had I not left anyway to join PA Technology my career there would have been blighted. I did have a five year foray back into an academic environment in recent years, but as a self-employed large project manager. What particularly surprised and dismayed me is what a political and backbiting environment it is, particularly at a senior professorial level. Everyone watches their backs pretty much continuously, jockeying for position. I'd have hated that aspect of university life.

Erdogan and Putin have issued similar statements to the effect that the murder of the Russian ambassador is "nothing but a provocation". So, although essentially dictators in their own countries, they are allies and are talking and agreeing a form of words. So there will be no Russian reaction.

And that is absolutely correct. After 20 years I'd forgotten how a darned dilution fridge worked. I said " The hotter atoms of helium 4 diffuse upwards into the helium 3 (no viscosity, remember), and a truly impressive pump attached to the mixing chamber sucks these hotter helium 4 atoms out. That shifts the equilibrium point, and the next hottest helium 4 diffuses upwards, etc etc " Which is the wrong way round. Of course, the He3 sets up a concentration in the He4, and you pump the on the He4 side to reduce the equilibrium 6% of He3. The He3 moves very easily through the He4 because of its superfluidity. The downward evaporation of more thermally excited He3 into He4 is the cooling mechanism. The guy in the video describes it very well - thanks for posting it.

Have you got any test gear - DVM (preferably more than one), oscilloscope etc? The T2 is such a complex beast I wouldn't like to trouble shoot without test gear. Offhand it could be anything from a fault in the left hand power umbilical to breakdown in a passive component to an active component going noisy, or a bad connection to a tube pin (I have that with one of my input tubes; wobbling it stops the problem. Note to self: must fix more permanently), or something else entirely.

RIP the victims of the Berlin truck attack on shoppers. 9 confirmed dead so far. RIP The Russian ambassador to Turkey, gunned down by an off duty police officer. WTF just happened in the last few hours?

Makes you feel like the front of your face is about to explode off your head. Get better soon Dusty!

She certainly lived a full life in the fast lane! Quite a character. RIP Zsa Zsa.

RIP the maneuver man - who lived to a ripe old age, and contributed greatly. Not a lot of people end up with a life saving technique named after them.

I forget now which liquefier we bought when we moved from Eynsham to Tubney. But it was big. Because this was cryogenic systems manufacture on a large scale we had to deal with a serious amount of helium. Every magnet could take typically 5 (and sometimes many more) training quenches. And because these were in less than efficient test cryostats, each quench took at least 100 litres of helium (for cool down and reservoir). So 500 liters plus per magnet, at a production of one a day or thereabouts. Not counting all the other systems (variable temperature inserts, He3 and dilution fridges). So a daily use of 600-700 litres was typical, which had to be compressed and stored (two monster pressure tanks outside), and then liquefied 24/7.

Yup - I did say "about". But as a reasonable ball park those prices are comparable enough. What would be the comparison for LHe? A fairly decent quality wine I was thinking.....

Finally something that looked less scripted (it still was of course) in Ep5 Hammond and May play battleships and cruisers using cars packed with explosive dropped from cranes, the idea being to hit the opponent's cars parked on a grid. By far the best sequence of the series to date.

Seriously though - you're right. In defense I got sucked into a conversation and lost track of the thread subject

Oh wait - this is the Stax thread, right. This is head-case though

That was interesting. For seven years in the '90's I ran engineering at Oxford Instruments, and apart from some outrageously complicated superconducting magnets, we pretty much had the world market in dilution refrigerators. These are the leaping-off point for many techniques to get even colder, using processes like adiabatic demagnetization. Anyhow, the dilution fridge works by a really strange quantum mechanical effect using a mixture of regular helium 4 and a lighter isotope of helium 3. As with all these cooling processes (brought out in the video) it is multi-stage using a series of heat exchangers until at the bottom of the machine is a mixing chamber less than the size of an ice hockey puck. In this the helium 4 becomes a superfluid and sits at the bottom, and helium 3 floats on top and has the viscosity of corn oil. The hotter atoms of helium 4 diffuse upwards into the helium 3 (no viscosity, remember), and a truly impressive pump attached to the mixing chamber sucks these hotter helium 4 atoms out. That shifts the equilibrium point, and the next hottest helium 4 diffuses upwards, etc etc. We designed and shipped products using this process to labs around the world - about 1 per week - which produced a base temperature somewhere between 1 and 3 milli-Kelvin. A long way above pico-Kelvin sure. But the race for ever colder temperatures all starts off with a dilution fridge - other than the Bose-Einstein crew, which use a different cooling process altogether. Anyhow, here is the business end of a dilution fridge. The thing at the bottom is the mixing chamber, the disc-like things are a series of counter-flow heat exchangers filled with sintered silver, and the copper coil has a capillary inside and is the next highest temperature. Not shown is something above all that called the "1K pot" which is what is says on the tin - so everything in that image sits at less than 1 Kelvin, and getting colder as you go towards the bottom of the image until 1 milli-K in the mixing chamber.

General rule of thumb is that liquid nitrogen is about the same price as milk, give or take, for about a gallon of each. Bulk price is much cheaper than milk.