A tour of Agilent Australia's new VOSCAL (Volume On-Site Calibration) portable NATA accredited calibration lab in a customised expandable shipping container, with Metrologist Peter Daly
The gear alone inside is worth a cool $3M.
Photos: http://www.flickr.com/photos/eevblog/sets/72157632748853904/
Forum Topic: http://www.eevblog.com/forum/blog/eevblog-424-$3m-agilent-portable-calibration-lab-tour'>http://www.eevblog.com/forum/blog/eevblog-424-$3m-agilent-portable-calibration-lab-tour
EEVblog Main Web Site:
http://www.eevblog.com
EEVblog Amazon Store:
http://astore.amazon.com/eevblogstore-20
Donations:
http://www.eevblog.com/donations/
Projects:
http://www.eevblog.com/projects/
Electronics Info Wiki:
http://www.eevblog.com/wiki/

Tell us about our volume Onsite calibration service is uh based around our 20ft shipping container. so it's a little blue and white box, travels around the country, disappears from one place, reappears in another. It's actually bigger on the inside than is on the outside. Aha.

Sounds a bit familiar, don't it? It does sound very familiar and it's blue and it's blue and white. So like our our lab inside. Yep, same quality system, same traceability and we use um, a little bit different setup procedure. We arrive at a customer site um and within 4 hours we're ready to start taking work right? We don't do Precision Time bases for 48 hours M generally.

but um, due to the warm up. Oh the ridum our Atomic Our atomic clock in here is different to the atomic clocks we use inside. Got it? Our atomic clocks inside are um, season beam tubes that are cross referen to GPS system. Here, we're using a ridian beam that is cross referenced uh to in its calibration.

it's checked and calibrated against our season beam tubes. and as part of our setup, we check it against the GPS system. How much power does the whole thing take? Um, it needs a 32 amp line single phase? Yep. uh.

When we designed our original onso facility, Uh, a lot of our customers, we we queried them. Um, many of our customers could give us three-phase power. Uh, but not all of them, right? So a solution was based around all of our customers needs. So we based it on a a 22 amp sorry 32 amp singlephase power supply.

It's a nice warm day here, so the air conditioner gets a little bit of a caning, but most handles it most of the most of the Heat's actually generated Bya via the Uh test itself. What we have is, um, an air conditioning system that's built to transfer that heat outside. The box is insulated. Um, and it's environmentally controlled.

Part of our environment is monitored same as we are inside. Uh, and it's the broader environment. It's 23 Plus or - 5 uh, and 5 to 80% humidity. 5 to 80% humidity doesn't seem like much of a challenge, but when you get to Al Springs and that's Z 0% outside, find the water somewhere and I Reckon we can fit a lot of people in here I Reckon we can too.

Let's try it. 30 people out of one box Dave That's pretty. That's not not bad. We normally normally have a crew of uh, two people, right? but it fits.

Heap We can fit 30 people in a what is it? At least plenty of room to spare Is actually bigger on the inside than on the outside. and if we take a look, it's actually a standard, uh, well, what looks like a standard 20t shipping container from this side. But if you head on over here, you'll notice that it's actually bigger on the inside than what it looks. It is expandable.

The uh, whole side of it just uh pulls out on one side. All the instrument racks are all on the far side of it, so that's where all the weight, well, a good majority of the weight of it is. and then this section over here pulls out and uh, gives them greater room inside. So it really is quite versatile and it's absolutely chalk of block at the moment.
And of course, there's the obligatory fire extinguisher just in case they, uh, try to calibrate one of those one hung low cheapy meters and it catches fire. so they've got a decent size. uh, air conditioner in it? Check it out. It really has to, uh, cool this sucker down in all sorts of, uh, temperature extremes.

The sides are, of course, um, the walls are, of course, and the roof Are all thermally, uh, insulated of course, cuz this would you'd be able to fry eggs on the top of this thing in the middle of uh, an Australian summer. And there's the old one, folks. it's uh, wasn't expandable at all. and uh, it was very little room inside that thing.

they had of course, the same uh racks, and uh, stuff that you saw inside this one. but uh, yep, that was pretty claustrophobic. Tada that's mine. Just step there.

Dave Yep, M The Tripod. But here it is, folks. Check this out. Racks and racks of pornographic calibration gear and I'll get some high-res photos of it all, no doubt.

But uh woo, that is a lot. a lot. of gear. give us some specs, give us a rundown.

Well, the reason why it's bigger on the inside. It expands, it expands. So this wall here comes up to about this point. Yep, uh for transport.

and once once we landed the customer site and set up, we push this whole unit out. Um, it can take one and a half ton as a load Right in there. Uh, where that that is its rating. We're probably putting about 900 kilos on it.

No more than that, right? So it gives us our as opposed to our previous onsite facilities. Uh, we. If you imagine that our workspace in our old mobile facility started here. Uh, we had shells coming to here and we had the length of the container right.

It was a tight fit, all right. and uh, as a result of some of us putting on weight, we bought this box. Uh, we. we had this box purpose built.

Our other our other boxes. um, were old containers that we refurbished and um, got a lot out of our second. Our second facility we built in 2000 and we've just retired it. Um, it was probably about 25 years old when we got it.

Mhm. Uh, had been around the world endless times. This box, uh, we put together through a company who had it built in China. It's only done one trip.

a boat, right? That's the only only that's the only one to get here. only slow. Took about took about 6 weeks to get here from China but it's not Chinese gear in here folks? No, it's not. We actually had it built in China by an Australian engineering company.

Uh, very good. Uh, cabox International they did. They did a a particularly good job of looking after our needs. um, particularly when uh, they went over they their Engineers went over and did all the inspections, the engineers didn't went over, did all the Uh testing and acceptance testing on it and and had it accredited for transport internationally right? So as it is fully loaded, it was fully Ked out.
This was just the this is just the Box this is just the thermal, the box and the airon, the box and the airon and the power. So um, that was all done built in China Unfortunately, but um, they were able to to do that. It was Australian engineering Australian design for the expanding unit Y which allowed us to Kit out the rest of it. Right now, the rest of the kit has been transferred from our older units.

m Um, it's got the pretty much the same capability as that laboratory inside. Yep, some areas not the same. Precision We don't have a seasian boom tube tube on board, we use a aidian Let's cross check the GPS Uh, but our 50 gig uh s parameters and 50 gig power. So accreditation from 20 HZ to 50 GHz same traceability, same accreditation DC to Daylight D The daylight.

Almost. We don't actually do daylight on. we tend to stop at 50 gig, right? Um, we haven't had a lot of. We haven't had a lot of um, uh, interest in doing optical on site, right? So we really, we're focused on the RF side of things with this box.

Um, let's face it, if you're paying 3040 or $50,000 for Network analyzer, or even in the case of a PNA like this one here, Mhm. Um, you might be you might be paying 150 $160,000 for that Network analyzer because you want Precision measurements. All right. If you're in a production environment, you want that calibra and turn around as fast as you can if you have a number of them.

Oh yeah, if you have a number of them, that's where we come into play. Um, we could go on. We could try and build a business going on site and providing calibration, handheld moding meters, and oscilloscopes for our customers. Yeah, if you can.

If you have a system that's based on a handheld M meter or oscilloscope or general purpose, That sort of general purpose stuff. Um, you can generally afford to buy a spare, buy spares, and then put those in production. You're not going to have a spare one of these. No, that's right.

All right, you're not going to have a spare signal generator. Yep, you're not going to have a spare Spectrum analyzer. Nope, They're all expensive gear in in. all in all, uh, for someone outside of our company to put this box together M with all the equipment in it would probably cost about 3 million bucks.

3 million bucks is what this worth that would that would would be what it would cost cost you to buy the gear. just the gear in the gear in this box in this shipping container. Don't forget. it's not just the gear you got to have the guy that operat exactly all right without blowing.

Trump You' Got to. You've got to have. You got someone who knows You got to have someone that that knows how how it's driven. You've got to have the systems to back it up.
Our systems are monitored globally for Quality Mhm. If I develop a test card in Australia for a particular Australian customer I published it GL globally. If I happen to make a mistake on that test card, every Tech in the world world, every engineer in the world has got access to that mistake. Y And if you if on the OD occasion you do make that mistake, someone's going to tell you about it.

Mhm. All right, because that's that's the whole purpose of having equality system. If you're in our industry and you can't afford to have someone check your work, you've picked the wrong trade. Exactly All right.

So we we. We are not ashamed of the fact that we make mistakes. but we correct those mistakes. There's OD occasions when something fails or uh We've made a mistake.

We turn customer and let them know. Yep, and why hide it? It's crazy. Well you know, well you know if won't go into that too much. But um, if if you're afraid of telling your customers the truth yeah, what are you selling your customers Exactly.

So in here we get we go from DC to to 50 GHz Um, any piece of equipment on this wall here yep can be controlled from either end. We have a switching Network enables us to share equipment there. the the equipment in the middle two racks. Mhm tends to get used by both sides right on the machine.

So our two operators will be here. Um, it might be Mark or or Bruce or Luke at the other end, it might be here me or Kate or Shan at this end. and um, we work work with each other to enable us to to share those common piece of equipment that enables us to get best best value out of it. So how many instruments can you count at one time? I Guess how many different types and we can do how many people you got? Well yeah, we can do a number of different measurements at the same time we can do while we're running an automated system.

We can be do doing manual tests right? All right? So we can be doing manual tests or running two complimentary systems at this end. For example, I can be doing RF work and on an independent system, the same operator can be doing the multimeters, handheld multimeters, and copes. Got it That are a value add to our product. Yeah, all right, they're not.

They're not your target market. You, you wouldn't ship this container somewhere to cuz somebody wants 500 mm calibrated Noth It wouldn't be worth it to them. Yes, we're happy to charge it for it, right? but it wouldn't be worth it. How much does it cost logistically to ship a container like this? I Guess just throw it on the back of a truck right and just drive it.

Well, no or is more to it. We? uh, we have it shipped through. Uh, a company called Toll Toll Transport. Yep, we get it picked up by sighter.

We insist on a S side lighter, not a crane, right? Not a forklift. Why that? Um, because side loaders are expensive pieces of a kit and the people that drive them want to look after them. Got it because it's in their best interest to look after them. So they tend to not knock their side loader about.
and they don't knock the load about. so it goes to a rail head. Um, funnily enough. Uh, it goes to the rail head and gets picked up by a side of the other other other side of the country In some some cases.

How? how does vibration and shock in the transport of this thing? Factory to such precision calibration equipment? I Assume that you'd have to calibrate. or you know, check all this gear traceably calibrated more often than something inside a regular standards cow Lab is the same rules apply. It's a measurement between two points in time, right? If we've had about we probably in the history of this equipment, we've probably had about two or three failures over the last 5 or 6 years, right? Uh, Which is normal? Are they gross failes or are they just slightly drifting out of SP Uh, tend to be tend to be either gross or a slight slight change, right? Very. R Rarely is there there's something significant in there, right? Um, and once or twice we've had to, we've had to recall a couple of products very rarely.

Our system for these things uh, is is based on the idea that if we, if we were to mount these racks rigidly to The Container Mhm um, and to The Container wall and the base uh, every shock gets transferred. the momentum of that shock gets transferred into into the equipment and straight into the reference all the other stuff. So um, and you imagine if this is a probably about 6 1/2 ton worth of container, right? If it's at a rail head and happens to get shunted by a 30 30 ton container, Ooh as a lot of momentum. So what we what we rely on is a suspension system.

That means it can B Okay, so that so it's actually compliant. It actually bounces, can can actually bounce around. it's got. so each individual rack is is is independently mounted.

All right, So it's it's. got rubber mounts on the back to bounce off the wall mhm uh. on on the rails and it's got resilient mounts underneath it to allow it to move. Did you actually do when you as part of the Uh configuration of this thing and the Um accreditation I Guess Did you have to actually measure the performance of all these shock mounts? No, no, You just put it in and then go still each.

Each item. We've got about 170 170 calibration standards in this box, right? Yeah, By the time you count everything up, we got about 170 calibration standards. some of them on a 2ear cycle, some of them on a 12 month cycle. We're just at the end of the calibration period.

For it, it's about a month's worth of work to get everything calibrated and and back into space. Again, Each each piece of equipment is calibrated Mhm between two points in time. Yep, all right. So the physical environment is important, but the calibration is what is what ultimately matters at the end of it.
That's right. So regardless of the envir, we have to maintain the environment as best we can. Um, regardless of that, we have to monitor that environment. But it still boils down to those two points in time.

Your calibration. Anytime you're using a A A quality system that relies on calibration, it's two points in time. Yep, everything that happens in those two points in time is your confidence level. All right, our 95% confidence level that we we have on a standard Cal or two Sigma Um is is what our general offer for us Got it all right, We don't do 100% Do.

Do customer needs vary I Can assume that some customers needs vary greatly in their calibration requirements. They do. They do many, many. C Many customers want to want a sticker for their calibration, uh, system? or sorry for the quality system.

They want a sticker on the box that says it's calibrated. Yep, and that's it. Yeah, those are the customers that we generally don't see because um, they're interested in price and they just send it to the cheapest CB they can get. which is yeah, Generally yeah, yeah.

or you know, backyard a backyard boy, that uh, that hasn't got the quality system or or Precision or the accreditation. Most importantly, the accreditation to do the work. Anyone can say that they can calibrate something, but your accrediation is what is what? uh, your bread and butter is based on And what accrediations are you guys rated to? um, we're that's the correct term we're accredited by Nat Yep, Um, which is the Australian accreditation body for our industry? Um, not only we we accredit, we accredited to a breadth and precision within our scope. Um, that is probably unmatched by anyone outside of Nml.

and that's the key, isn't it? Because anyal any calibration lab any tinpot guy in his backyard could potentially get n calibrated with within a specific narrow field of could meur. you can say you're not accredited Y Without going too far into your scope, you can say you're not recuited for RF power. Yep, um, but over what? Over what range to what? Precision is that we? That's that's exactly right. Um, some of our competitors RF accrediation some of our significant competitors.

their RF accreditation. Uh, might be 50 MHz 1 M Yep, which is great if you're doing 50 Me: mehz that. so that's as high as they'll go, that's as high as they'll go, not lower 50. MHz We're still DC right? URF guys to us here.

But um, so our our scope uh is. Not only does it go for RF and S parameters, not only does it go from 20 HZ to 50 GHz, but it's also the power ratios as well. So we go from not only 20 HZ to 50 GHz But in our Power Scope we go from plus 20 Dbm down to- 136 dbm and you can do that in this portable. Most of it most of it can.

Um, the same for RF Yes, M for RF DC DC Not so much we can't do Precision Equipment We can't do the 3458 for example. Oh okay, you wouldn't do those in here. We wouldn't do those in here because they're level of precision that can only be done in a standards lab. Got it all right? So there there are some things that we draw the line at because it's just not commercially viable for us to have that tighter scope in our mobile facility.
Well, the thing is, most cow Labs I've been to will actually use the 3458a as their primary transfer standard. Yeah, so like is in two calibrator mulet. so you can't send a you. In theory, you couldn't send a 3458a to them.

Well, could, right? you could. But you get what you pay for it, right? All right, we have 2 3458 in here. Yeah, right. they're used as working standards y They're calibrated against our gold standards in the standards laboratory and our gold standards go away for calibration every 6 months.

So so give us the level breakdown. There's gold standard at the top. Yep, or is there another one above that? Well, our gold stand. Our gold standard is a transfer standard is it is is what you call transfer standard, right? It's it's.

it's transfer standard. that's that's been calibrated in Loveland in the United States Yep, against primary standards, right? Bank You just that's that's the the highest level. that's the highest level. that we have available.

Uh, in here? Um, the reason we use Love Lovelin instead of Nmis is it's a fast turn around time for us. Uh, we we get the the same level of precision we get. Um, not only do we get the report, but we get the data. Yes.

So all the offset data is transmitted to us uh via The Ether and a straight into M. So when our Go standard gets back within 48 Hours of of warming up and running, we can do the cross checks on it against its offset standards, against our other gold standard as a cross check and we have that we have maintain that confidence level for our customers for our quality system. Got it? Because these something like this 3458a, it actually contains offset data doesn't It's not just a Precision reference and then that's it. It has to each range.

How is it? Each range contains one offset data. You know? I don't don't want to go into that because that's bit of propriety. Is it? No, No. I Just have to pull a book out to look it up.

Oh right. All right. Um, but the offset data is built into the machine. but the offset data is also.

um, also a calibration data for our standard lab. That's that's that's that's part of. So that's the error. That's the ABS.

Not only not only are girl standards calibrated against a good reference, but their offset or error is taken to account as well. right? So it's not just a specification, but it's offset. So would it be when you send one of your transfer standards back to to to L to get calibrated? Yep. Do they tweak it? Do they tweak it spot on? Or do you just get the offset data? Get the offset data.
You just get the offset and then would you tweak? No, No. You wouldn't touch it. Never touch it. Why would you? Why would you do that? Why would you spend all that, invest all that time, money and effort to measure something within two points in time and then adjust it all right.

if we measure it between two points in time and it doesn't need adjustment. if it does need adjustment, we have to go back and hit in time. we have to look back in time. We have to look back in time if if we have to adjust that 3458 gold standard, we have to look back in time and see what it's been used on y for all our customers and internal jobs that that particular item is used for.

Our our quality system is such that every time we do a job, we list every piece of uh, calibration equipment or every Cal standard that's used on that on that um, calibration so that we can track it. Got it? The importance of being able to track it is when it dies or when it has a has a fault or an adjustment. At that second point in time, we can look back in time and determine what it's been used on and determine what impact it has on our equipment and the customer equipment. It's more important that we do that than just have a a cow sticker at the end of it and say yep, it's it's calibrated Now it's working.

We have to be able to determine what the impact on a customer product is, and on those rare occasion where it there is a likely impact on that customer's product, we will recall it and we will recalibrate the customers customer box right? So you send this this 3458a, You know it was a value 12 months ago and let's say it was right at the edge of what you would deem to be acceptable. Okay, and then 12 months later it's at the other end of, but still acceptable still within your acceptable range. What would you do? This is where this is where your metrologist comes into play. Your metrologist has has to, um, analyze that data and and determine what the impact is.

If it's unstable, like if it's year in year out, swing. Yep, one way to the other. Um, either the reference that you're checking it against is unstable or your box is unstable. If you're using that as a transfer or working standard, then you you've got a serious problem because there's two different issues with something with a transfer standard like this isn't there.

There's absolute level, the absolute value of it and its stability. and they not necessarily and or it's drift and they're not necessarily the same Two different there are things that that impact different devices. y Um, if your the stability of in many cases device under test is is the largest source of instability. mhm if you're doing a Time base If if you're doing a most of our instruments run off of 10 Meg time base, Right right? Most of the Agilant time bases um, are stable within within uh, 4 hours.
However, we we wait. Uh, if we're testing a Precision temperature controlled Crystal oscillator, we'll have it running for 48 hours, right? All right before we. So this van will have to be on site powered up for 48 hours before we start doing those. Precision Time bases General time base a normal time base Um, your air Crystal oscillator or your um, your old culprit style oscillator RC oscillator.

Um, if you're wait more than an hour, you're just wasting time. Yeah, all right, that that sort of that sort of sort of precision is not there. So that's right we gen. We generally um, need that Precision for the temperature controled crystal oscillators Mhm.

Um, and that's the limit of the time bases we do here. Oh okay, so you couldn't do anything, couldn't do. You couldn't do a Reidan in here because you're using a ridium as a reference. That's right, How how many orders do you have to go above? Precision Generally your test accuracy ratio is is what comes into into account and if you've got say you do a measurement that that is Uh, you've got a specification of plus or minus one unit.

Yep, if your uncertainty is plus orus 1 and2 units, You kidding yourself, right? Yeah, Exactly. All right. So if you've got a specification of of uh, plus or minus one unit, um, then your uncertainty level should be better than 0.25 right? Uh, that gives you a test of accuracy ratio of 4 to1. So four times better is that's AB minimum minimum.

Most of the what would you run in practice most of the stuff is probably 10 to one, 10 to one, 10 to one. All right. So you know if if your device under test is Um has got a specification in percentages, your reference will be in parts per million Mhm. All right, that just just to just to give you the difference, can we have a breakdown of what instruments you got here? Cuz I'm sure people would actually want to know each bit of Kit in here each bit of Kit So you want me to go through 170 items not not quite 170? I Mean you've got maybe what 20 20 here in the rack is that? I'll try I'll try and go through them all right? So here's we'll start from here.

We got a AAL kit. Oh sorry folks I'm uh I don't think I can. Uh, we're using our wireless mics here I don't think I can separate my camera from the oh yes, I can There we go. Okay, all right so we have a PNA Precision Nwork analyzer.

Yep. Below that we have a test test linearity system that looks homeade. Uh, it's not homemade, but it's uh, it's SSU made right. Over here we have a low lower frequency Network analyzer.

Below that we have a a 3 GHz Pulse Gen. Up here we have a low frequency spectrum analyzer and network analyzer. y we have 3458a We have another another manufacturer's product. oh oh no.

another manufacturers product which these are calibrators. Agant doesn't make calibrators. We have a counter. We have a pulse chin.
we have another pulse chin. We have a multimeter. We have a noise Source controller We have a noise figure meter. We have a Scala Network canalyzer We have a 50 gig Source We have a switch.

We have a power meter We have a PSG we have another switch box amplifier box we have a 1 GHz scope We have a switch box We have a 30 3 33 5A low frequency High Precision level generator It's an old unit but still works. Pulse Gen Amfm test set function generator 50 gig Network analyzer 50 gig oscilloscope another 30 G 3458 another one 26/2 gig counter Distortion analyzer measuring receiver down converter another Sig Gen 50 gig GPS System: dual power meter Ridium distribution app DC Load device under test counter Pulse Gen Counter AC Power Supply Phase Noise measurement system That one just does phasee noise that's totally devoted to to phase noise. It's um, it's driven by the E5500 phase noise system as a test set that enables us to do Um residual phase noise from 10 Meg out offset from the carrier down to 100 MZ offset from the carrier right Um and when we're doing residual phase noise, that gives us a d dynamic range of Uh minus 170 DBC per Hertz at the top end. Uh, these are just some phase noise plots.

This is just a cross cross check we do on our phase noise system uh on three different systems. So we we've measured the the 10 MHz out of the GPS system, our abidian reference, and the temperature control. Crystal Osat All right. Uh, it's not an absolute calibration, it's just a a cross check on our system.

Make sure make sure it's operationally ready to go. And how often would you do this or is it just one off? No, you would do it, um, fairly regularly. You would do it as part of your normal cross check as you start up right. Oh okay, so every time you take the this uh van, this uh container on site, you would redo these charts Before you do.

You would do a check on on your system, right? All right. So you do a check on your system. In this case, we've done it against the 10 Meg ones. All right.

It's just an example of of giving yourself a confidence level on what you're measuring. So we've only gone, we've only gone down to 100 HZ offset. Yep, because we're doing what's called absolute phase noise just as a relatively quick check. So if you look at them, it's it's taken That's only taken a couple of minutes to do each plot on there.

If if we wanted to go down to Um 100 MZ offset and do residual, it probably take us about 25 to 30 minutes to do each plot right. That's because as you know yourself as you get resolution gets as you get closer with the Fft to go from the Um, the analog to digital conversion takes a lot more samples, a lot longer time which gives you your your Allen variant. So that's right, we only use residual phase noise for those particular devices that custom is require it on or or those Precision boxes where it's um specified because it's it is directly proportional to the Allen variance of the system. Got it all right.
So if you've got uh your your average agilant uh time base as I said earlier is good. probably good after 4 hours a Precision Time base we won't test for 48. All right. And and um, some of it and most of our signal generators have that option in them, right? So that's part of their standard standard uh system or an option for them if some of some of our competitors have got.

and we do do calibrate competitors boxes because we're able to go on site. Yeah, we're able to calibrate the competitors boxes. Um, we can't calibrate their time bases because their time base has a 30-day stability spe. So so it's specification for stability and and short-term stability like phase noise.

Um, we're not going to hang around 30 days of course, not. Well, if they paid you maybe, but they'd have to pay a hell of a lot. That's right. So um so yeah, you always give it to paper.

All the other all our other standards are are smaller standards like our power sensors, power sensors, mixes. Oh yeah, please open the drawers, see, see it, look, look, oh it's wet dream stuff. It really is. And the reckon I need to get life all right.

So so not much to see here. They just just connected another bits in pieces. So yeah, and we have a number of boxes over over here. where, uh, each of our cow kits for example, come in um Precision box like that 34.

uh sorry. the 8332 F Cal kit. Oh yeah, all right. so these these devices.

We use these devices here as our reference standards. not for not just for our not just for our Network analyzers, but for customers. Economy calculates. We these these go overseas.

Um, they're directly traceable to physical dimensions. Um for their performance as well as the RF performance. All right. So that's it.

That's that's where we get our traceability not only for linear magnitude in S parameters um, but also for phase for the opens and shorts. So the opens and shorts calibrate. That's a low frequency one that's only good to 9 GHz Um, an economy C kit Like that one there. it's good to 18 GHz economy.

That's an economy one there. but um, a sliding load C kit that we we take on site. That's good for 18 GHz That gives the sliding layer, gives you a different order of precision and you also get What's called the electronic Cal kits. Uh, where if we're doing lower order equipment Um, like a common common attenuator or a switchable attenuator.

Um, we can set the calibration up and hook the electronic ecal in as an insertable device and let that do it itself. self Cal doesn't give you the best performance. low frequency ranges. Uh so n for below 2 GHz We'd use a mechanical Cal kit for most most uh things above two above 6 GHz We' use a sliding load Cal kit.

Uh for most things, we also use a sliding load Cal kit for precision work between in that region from 2 gig to to 6 gig. Got it? and it's about 3 million bucks worth of gear. probably? Yeah, that'll be about it. That's how much it would cost you to set it up.
Unbelievable! Thank you very much Peter No worries.

Avatar photo

By YTB

21 thoughts on “Eevblog #424 – $3m agilent portable calibration lab tour”
  1. Avataaar/Circle Created with python_avatars Ian Cooper says:

    Peter definitely knows his stuff!!

  2. Avataaar/Circle Created with python_avatars Cirilomags 7 says:

    I AM USING WAVETEK AND FLUKES CALIBRATOR IN SQUARES METERS RPM VOLTAGES METERS PYROMETERS TEMPERATURE DRY BLOCKS CALIBRATOR PRESSURE CALIBRATOR ALL MULTIMETERS CALIBRATED IN FLUKES TORQUE DIAL INDICATOR TRANSSET

  3. Avataaar/Circle Created with python_avatars mine own says:

    Gimme !

  4. Avataaar/Circle Created with python_avatars Daniele Patane' says:

    As Hewlett & Packard Fan and collector I would watch this video for hours

  5. Avataaar/Circle Created with python_avatars StoesA says:

    That 19" rack equipments is worth most

  6. Avataaar/Circle Created with python_avatars john gabris says:

    hold the camera steady

  7. Avataaar/Circle Created with python_avatars bayareapianist says:

    An oscilloscope with flappy disc drive in 2013? I guess agilent never wants to drop floppy and use USB drive!

  8. Avataaar/Circle Created with python_avatars Ruel Tmeizeh says:

    Watched this a second time today. Very cool.

  9. Avataaar/Circle Created with python_avatars Mr. Simulator says:

    Can you still fit 30 people in there? LOL

  10. Avataaar/Circle Created with python_avatars Luis Herrera says:

    Holly molly, this video is pure gold, good old days before Keysight shows up.

  11. Avataaar/Circle Created with python_avatars Ura Nium says:

    Why not rewind the tape again and again. Makes it look longer. Or biGGer.

  12. Avataaar/Circle Created with python_avatars Persoana a 3-a says:

    $3M worth of gear and still using Windows XP

  13. Avataaar/Circle Created with python_avatars Paul Brisson says:

    Accreditation does not mean you are a good cal lab.

  14. Avataaar/Circle Created with python_avatars Pete D says:

    Bet they use Fluke MetCal and MetTrack for cal automation and asset tracking.

  15. Avataaar/Circle Created with python_avatars Matt Boland says:

    3 million bucks? I'll take 2 thanks…

  16. Avataaar/Circle Created with python_avatars N Bake says:

    Nice vid! Accredited Calibration Services (Marsh Metrology) in Canada has developed a dedicated mobile lab facility for RF/Microwave type test equipment for multiple manufacturers including Agilent equipment as well. A second mobile lab can calibrate parameters (electrical, temperature, pressure, torque and more) to ISO 17025! Contact us for more details!

  17. Avataaar/Circle Created with python_avatars Eric Helal says:

    I would watch as many videos as it took to explain what all of these tools do to a lay person. I am fascinated by your world.

  18. Avataaar/Circle Created with python_avatars Chuck does Hotlaps says:

    How casually, "so here we go from DC to 50 GHz"
    "So here we from 'easy does it' to 'black magic and the occasional mathematical model for things that kinda don't exist but fuck it it works.'

  19. Avataaar/Circle Created with python_avatars Gareth Jones says:

    Great video, beautiful gear!

  20. Avataaar/Circle Created with python_avatars SeltsamerAttraktor says:

    I'll take two then.

  21. Avataaar/Circle Created with python_avatars Foxy rollins says:

    ahem.. just connectors.. says the man with all the connectors.

Leave a Reply

Your email address will not be published. Required fields are marked *