And that's the standards lab. But Pete's gonna show us around the working lab. All right. This is our laboratory of General General Laboratory and service area.

It's an electrostatic safe area so we have an electrostatic floor that's it's grounded and and requires electrostatic protection. So I'll ask Dave to keep his hands in his pockets and not touch any gears. Don't touch anything I'll just slip it all right. and there's probably.

There's probably a couple of reasons why we need Dave to keep his hands in these parts and that's and there's a tester you stand on that. Normally it's normal, normally a procedure. So in here we're broken up broken up into probably three three areas, three and bit areas. We've got our automated calibration area where we put a high-volume equipment through automated procedures.

again a standard procedures that maintain globally. On top of that, we have our manual Cal area where we do Cal procedures that need a little bit more operator intervention and a little bit more skill from our from our calibration engineers. It's sort of a progression path to go from automated to to manual cows to developing your own Cal Procedures When we developed Cal procedures, they have to be validated by a metrology team and metrologist. Super Guys being one at one out of three I know was a three minute metrology guys.

So we you were telling me you have to get your rear credited every four. Yeah, it's not a part of your accreditation. You basically have to research your entire trade every four years. So it's it.

Gives our customers a bit of confidence that the lastest but quality qualities had in my time. If you do it properly, if you also do it properly, it's it's It's a benefit to everybody concerned. Take shortcuts. We we spend about two nearly two full time.

He'd get devoted to quality and metrology across our organization. Fantastic! So I go for A - I Show us each individual. So here's here's one of our automated areas. Here you can see in the background we've got a computer control system that drives all their reference equipment.

In this case, this is a low frequency station so that we can test a book device under test to a validated procedure. It's important. It's not only a test equipment is traceable, but it's also that important that it's done to a validated procedure and we validate globally. When a manual guys develop a procedure for a multi vendor box or even in an old box that isn't necessarily currently sold, we have to publish it.

We publish it globally. All right as its. We don't keep secrets within their organization. If if you can't afford to have someone look over their shoulder, you've picked their entree.

All right. So we everything we do is audited and ordered to the ball. And when we publish globally, Agilent's are fairly fair sized company. you can imagine if we publish a web based a web based tool that demonstrates your procedures every to Agilent technician and engineer and the planet.

It's got access to it and if you make a mistake, they're going to tell you about it. they're going to tell you about it. They don't miss a trick. Nope.

And one of these two racks here and more automated. Yes, this is a higher frequency station. some microwave station so microwave signal generators would go through here. Counters: Spectrum analyzers up to 50 gigahertz.

This is an older station, we've got an old 83 48 there so as a standalone station it would only go to 26. Knife geek will drag in PSG or 83 650 in to extend the range to 50 gigahertz as needed. And what's involved in calibrating or calibration? Verifying a spectrum Analyzer: A typical spectrum analyzer, you had too much how many hours. Okay, you suspected through an automated procedure, calibrating a spectrum analyzer to the our OEM procedure would probably take depending on the frenzy range because the the frequency that the Ire functions are common, but the frequency range extends, Your frequency response tests, your your noise tests, your residual tests extends it.

So a a three gig spectrum analyzer might take you three hours to do on an automated system, A twenty six and a half key analyzer might take you four or five hours depending on the on the bottle and and the options. Doing it manually. doing the same one manually. it's the same level of testing can take you eight or nine hours and these can spit out automated reports.

And yes, our airports, airports and certificates are generated so that they we can give a customer a hard copy or a soft copy. Now soft copies are our air records and they put their published globally and ordered originally. So original auditors come through and tell us that you've missed something on your on your on a stivic. they'll they'll pick pick a fault under a system called Lighthouse that tells us that the Associated time based report from this device is missing.

Where is it? and if you can't find it, you have to recall the Box just an example of something might happen. So that's all there. It all that report information and certificate information. the phase noise test, the time based tests, the the measurement report, and the certificate and the uncertainty analysis is all available online to us and to our customers.

All right. So that begs a question. If your suppliers charging you $100 $200 to generate a report, what's he actually charging you for to printing out a copy of a PDF document that you should already have I'll leave that one for you Dave All right. Next and here's a whole bunch of gear.

This is all some old stuff, some current stuff. the this is general purpose test equipment. It's still calibrated. Oh right.

Okay, so this isn't customer stuff This? No, no, this is. no. this is lab gear that's leftover, some of it's not in use and you'll be able to tell the stuff it's not used because it's got got a normally go to stick around alright quarantine sticker So it's not calibrated on calibration. We even can't wait.

What? We even calibrate past Claws? Yep, yes, some people need their well, well, they they want you. Yeah, it's part of the standard if you do. If you're doing switching systems for for data and the like, a switchable power supply and load generally match for it. So is part of our manual cal area we have.

We have a pretty competent Emu. We've got three people operating manual Cal area experienced calibration engineers. We've got another station here that's a semi semi semi automated station where we do some particular precise measurements. So we do some more precise measurements in this area than we do in the automated area.

Traceability past the same. Certain these are similar so, but this is more manual. This is more for for an item globally. we don't generate a an automated procedure across the planet for any item that there's less than 100 of.

So if if we start seeing a call the multi vendor boxes a box that's mined by another manufacturer. If customers start to ask us to do that, we'll make a decision whether we're going to do it I'm not, whether it's worthwhile and in most cases that will do it. But we're supporting a customers box our competitors box so we're going to charge a little bit more foot. Inevitably, they the small volume of competitors boxes we see per model number, they become manual celts.

So this is where the manual Cal comes in. What sewing menu When you do it If you if you out to a measurement manually that you can do an automated system and there's something wrong, right? right? Yeah, Yeah, yeah, it's still an analog analog universe and this is all old-school GPIB Controlled. Yes, mostly mostly mostly. we also do.

We also dual in for the the lens stuff so that the field Foxes can be calibrated either end for example, and they're there. The reference equipments all run by GPIB The device under test is controlled by the link cable and what's this little? look? This little look here is the passing warning folks. Voltage test area. All right.

So this is this is. This area is devoted to testing passport and can throw in CML Sierra. You're out of the blue, but these two racks here are totally devoted to AC and DC power supplies and loads. So we test that.

we test the loads for under full load and test the power supplies under full load. We test them for things like RF RF noise up to 20 megahertz. On top of the DC we test, we test them for RMS and peak peak noise. We test them for the transient response time between full load and half load.

So we switch that we switched the loader to Kilohertz right on and off between full load and half load and check the response. It's any good, power supply is going to be spec for that successor. I'm an for no is usually up to 20 megahertz. Yeah, Yep, yeah, so that's that.

That's a sort of. It's not just a minute when you do a power supply. It's not just a matter. Okay, more than eight across and saying it's got 24 volts.

Yeah, that's that's items are waiting. Quite acceptance or repair parts all right. We have an optical area here where we we do optical testing. So we do essentially we do in fiber measurements, multimodal and single mode fibers, wavelength, optical power, optical attenuation.

Again, it's a traceable traceable calibration and covered by a quality system. Is there much call for fiber? Not a lot about testing, not a lot. It's the RF. The RF sort of of the industry tends to be more focused on on quality and traceability.

There is less, there is less less demand in the optical side for this sort of work and we cover in two air repair area repair air repair areas run boy to competent engineers. it's It's led by mr. Luke Kindt Iski whose air appear manager and it's and he's lovely assistant. Joey And on there's a repair engineer so these guys these guys to probably M&E How many different product numbers did you say look 120 different products that we support that we do repair repairs in Australia over the last year.

So hundred different model numbers not just different units, different model numbers. and is there any particular type of instrument that fails more than others do like as endoscope sale more than multimeters. Do power suppliers fail more than but you'll find a lot of multimeters my file because it's just a nature of instrument. They've actually abuse it.

Correct Yes device gets dropped, gets damaged that it gets not abused but you know one Wrong measurement and pain. Yeah there is no time trying for you to the correct mistake. You just make that mistake. What happens to a network? A network analyzer when you put sixteen hundred watching the front end of it? We had a customer who put who managed to put six hundred watts in the front of our 8753 yes in front of us because the person setting it up didn't actually know what it was doing like We stood back with their hands in their pockets and said this isn't this isn't right but when they try to collect claim claim or and you don't we said I know go around to a face noise area again Oh Live Free station.

Looking You want to do the introduction on the lead free station with across in in the US and in Europe. You know that row has compliance is becoming more and more important and we're starting to see that in Australia I Don't think the legislation is has is likely to come in the short term in Australia but in the long term we'll probably as a manufacturer. If we want to be a manufacturing base and we want to be able to sell to the US and Europe, we're going to have to comply. So it's Row has compliances, removal of hazardous substances, things like LED So it's a lead-free soldering station, even the tools isolated here.

we can't You: We can't take a set of pliers or side cutters from here and go use it on leaded solder and then and then bring them back here because we're going to contaminate the lead-free station. so it's a different type of soldering the lead soldering. For those of us that did high reliability hand soldering as part of our trade, you're always aim for a shiny joint, join it. always cleaned it off with flux removers and everything else and you know when you inspected it with a you we're looking for a perfectly mirror mirror that best finish with the lead-free soldering station if you looked at it under the same conditions.

Every every joint looks like it's a frosted joint. because you're not using a eutectic point, you're not using flux remover and though that that same sort of technique so that's gone by the wayside with the row has compliance. Unfortunately I don't do that stuff anymore which is which is good when when look he needs a hand and I offered to give him a hand and go and give him a hand with the repairs you normally answer me and bro no, it's not not quite normally fit parts run contacts Yes! So now we'll go into a phase material so of course it's locked I Went I went unlock it, you've sent It's similar to the phase noise assistant that we use in the TARDIS except it has a higher frequency range. This this allows us to go to 26 and a half gig directly there using the external mixes and and down to the normal normal frequency range of low-frequency to get down to 10 - making only differences we're using a different FFT it gives us a much better dynamic range than normal due to a converter which gives us a lower noise floor closing In particular, so it's very nice is it Agilent's policy to develop state of the art you know, gear that can be used in calibration? That's because almost every bit of gear you've got here is Agilent Yeah, it is.

it is. If you look at the reference signal generator here, think back to the old days when you had the 86 62a and the 86 63 A, which were the industry standards for phase noise across the plant. They were the industry standard. so we now have this one here, which is the E 86 63d which has developed particularly for the low frequency low phase noise performance on up and in at the low frequency end where residual phase noise, not absolute phase residual frozen phase noise, that's we where the time bases have to be tied.

The problem with Absolute phase noise is that as you get close in and and we're talking 3 Hertz to 200 million Hertz offset from the carrier, nice and close. If you're using a phase lock loop system, any instability of your device under test within that phase lock loop is lost within the phase lock loop. So when we, especially when we specify a residual phase noise, a time base of the device under test, and the reference oscillator a lot. so any instability of the device under test actually shows up in that residual measurement.

It's a much tighter method than absolute or a phase locked loop method. It shows the inherent stability of the two devices in the case of this one. Here, at low frequencies, it's actually got a divider built into it, so at low frequencies you can switch in a divider and which mathematically gives you 6 DB better phase noise because you divide the signal alright so automatically and because you're not harmonically mixing that lower frequency Ben you, you're dividing it. You probably get another 3 DB benefit because you're not going through that that heterodyne mixer to get the lower frequency so effectively get around about 9 DB In practice, about 90 be better performance on those those lower frequency phase noise things, particularly for radar sites.

It's it's critically important. You've got a got a radar signal that's going echelons away and coming back. Your face nor your face. Noise and noise figure all add up to to determine what the dynamic range of your radar system is.

So that's why it's That's why it's particularly important for those guys. it's that way. most of your most of the stuff you most of the call for equipment and calibrated equipment comes from is in A in Australia Munis in Australia defense is certainly large, large interest to us. We don't want to go into too much detail about what we do or where we go, but it is significant for us.

But we also have significant manufacturers as well. They're associated with it. So some of those people that you met this morning, some of our significant manufacturers in Australia and we support those people directly is that it. That's pretty much it.

We're back to it. We're back to the unstable atomic clock. So this is where most general-purpose gear is calibrated as opposed to the standards lab which is the cell. Yeah, that's that's for really.

The standard slab is for for the low frequency standards, the thermal voltage converters, the DC cells resistance standards, their precision multimeters so it's 3458 gold. No testing in Stargirl centers our the calibrate us that we we use get calibrated in there as we discussed yesterday and the only other thing it's it's in there more for isolation than necessity that its parameters and sensor system is in there as well. So it's and they're accredited differently. The to note is the one accreditations to one a corrosion so got it.