전사 🔗
WEBVTT
00:00:27.679 --> 00:00:36.399
In today's episode, we'll cover what low cali what low pressure calibration is, the factors that can affect it, best practices and the industries that rely on it.
00:00:38.000 --> 00:00:41.920
You're listening to Measure Matters, a podcast by Mencer.
00:00:43.039 --> 00:00:43.920
All right, Philip.
00:00:44.000 --> 00:00:47.600
Well, this October marks your 15-year anniversary with Mencer.
00:00:48.000 --> 00:00:49.200
That is a huge milestone.
00:00:49.280 --> 00:00:49.759
Congratulations.
00:00:50.079 --> 00:00:50.240
Thank you.
00:00:50.320 --> 00:00:51.280
Yeah, yeah, it's been great.
00:00:51.359 --> 00:00:51.600
Yeah.
00:00:52.079 --> 00:00:55.200
So, what is it that keeps you at Mencer so long?
00:00:55.359 --> 00:00:59.039
And if the doors are rigged, the calibration lab doors just blink twice.
00:01:00.719 --> 00:01:02.880
Uh no, it's just a really great place to work.
00:01:02.960 --> 00:01:06.560
I mean, uh, you do have there's several people here that have been here for a long time.
00:01:06.640 --> 00:01:08.719
Uh, so it's just a just a great atmosphere to work here.
00:01:09.040 --> 00:01:09.439
That's awesome.
00:01:09.519 --> 00:01:09.680
Yeah.
00:01:09.920 --> 00:01:12.719
And can you talk a little bit about what you do here at Mencer?
00:01:13.200 --> 00:01:17.519
Um, my main responsibilities are as a calibration lab manager and the metrologist.
00:01:17.599 --> 00:01:24.319
Uh, so the manager position is just you know managing the lab, making sure you know people are where they are, getting stuff in and out of the lab.
00:01:24.480 --> 00:01:34.159
Um, the metrology side of it is to make sure all the equipment we're using in the lab is calibrated, it's well that's in within its interval, it's behaving in a good way that we can continuously use it for calibration.
00:01:34.239 --> 00:01:36.000
But that's pretty much those two positions.
00:01:36.239 --> 00:01:36.640
That's awesome.
00:01:36.799 --> 00:01:36.959
Yeah.
00:01:37.200 --> 00:01:41.519
And do you have any exciting projects coming up that you can talk about or recent projects you've done?
00:01:41.840 --> 00:01:44.719
Uh, we are working on expanding our temperature lab.
00:01:44.879 --> 00:01:47.680
Um, our scope has been relatively narrow for a long time now.
00:01:47.840 --> 00:01:53.840
And that's one of the big challenges that's gonna be coming for me is it's learning everything I can about temperature uh to expand it.
00:01:53.920 --> 00:01:59.200
Because to expand it in the areas that we're gonna need to expand it, it's gonna mean new equipment, new measurement standards that we're gonna have to use.
00:01:59.280 --> 00:02:01.040
So that'll be a lot to learn.
00:02:01.200 --> 00:02:02.159
Yeah, that's exciting though.
00:02:02.239 --> 00:02:03.200
Yeah, yeah, definitely is.
00:02:03.280 --> 00:02:03.599
Yeah.
00:02:04.000 --> 00:02:06.560
So let's get into the topic of the podcast.
00:02:06.719 --> 00:02:10.879
What is how does MENSER define low pressure calibration?
00:02:11.280 --> 00:02:17.680
For us, it would be really the ranges that fall too low uh for our other lab standards.
00:02:17.759 --> 00:02:28.319
So, really anything less than about 7.25 PSI, um, anything less than that, or within like a bidirectional pressure range of minus five PSI to positive five PSI.
00:02:28.639 --> 00:02:28.879
Okay.
00:02:29.599 --> 00:02:35.120
So the last time you were on, you talked about uh false acceptance risks and uncertainty ratios.
00:02:35.360 --> 00:02:40.719
Can you give us like a brief overview of how that fits into the topic of low pressure calibration?
00:02:40.960 --> 00:02:48.400
Yeah, so that's just in any pressure calibration, you're gonna want to be concerned with what is your test uncertainty ratio and what is your probability of false accept.
00:02:48.479 --> 00:02:58.319
Uh the test uncertainty ratio um is really just the ratio between, you know, the device under test accuracy and the total measurement uncertainty of your entire process, of your entire calibration process.
00:02:58.560 --> 00:03:04.639
And the probability of false accept would be what is the likelihood that we tell you your calibration is good when it's actually bad.
00:03:05.280 --> 00:03:10.960
So really it's once you add your measurement uncertainty to your measurement errors, what is the probability your device is actually out of specification?
00:03:11.199 --> 00:03:12.479
And we try to keep that below 2%.
00:03:13.120 --> 00:03:13.680
Below 2%.
00:03:14.000 --> 00:03:14.159
Okay.
00:03:14.240 --> 00:03:16.639
Is does that go into the four to one ratio?
00:03:16.719 --> 00:03:18.080
I know you kind of touched on that last time.
00:03:18.400 --> 00:03:31.840
Yes, typically once your ratio gets so high, then you can kind of ignore what your your PFA is because your your insurance ratio becomes so high that you don't have to really do any guardbanding, you can be confident that your your measurement process is good.
00:03:32.240 --> 00:03:32.400
Okay.
00:03:32.879 --> 00:03:38.560
And you may have answered this, but when you go into the low pressure calibrations, like how does that factor in?
00:03:38.639 --> 00:03:41.599
Like, is there a certain process you use to ensure that it stays?
00:03:42.319 --> 00:03:42.639
Yeah.
00:03:42.800 --> 00:03:55.439
So with uh high accuracy sensors that we provide, like the double eight percent FS for the low pressure sensors, we are kind of with our current processes have to use the process that we're using with the two deadweight testers.
00:03:56.000 --> 00:04:06.960
Um we could use different methods, but then we may not guarantee that we're getting, you know, the four to one test and certainity ratio, or that we may have to widen the PFA percentage from 2% to 5% or something like that.
00:04:07.120 --> 00:04:10.879
But um, so there's lots of different reference standards that you could use.
00:04:11.039 --> 00:04:14.879
What makes a deadweight
Load full transcript
tester the right choice for low pressure calibrations?
00:04:15.199 --> 00:04:27.120
It is typically the best accuracy for a reference standard you're gonna get is gonna be with with with usually a traditional debris tester of any kind, but but specifically for low pressure, there's really only one kind that that we know of that you can use.
00:04:27.439 --> 00:04:27.680
Okay.
00:04:27.839 --> 00:04:31.519
And what what makes it that that good, I guess?
00:04:32.079 --> 00:04:44.240
Uh it's it's a combination of you know, uh you know, low uncertainty with the determination of the area of your piston cylinder, the which is also a stable area over time, the stable masses because you'll you'll get masses.
00:04:44.319 --> 00:04:46.879
It uses like intrinsic values for these things.
00:04:46.959 --> 00:04:49.920
It doesn't use like uh any digital digital part really.
00:04:50.160 --> 00:04:50.319
Okay.
00:04:50.480 --> 00:04:54.000
Um, it's physical artifacts that that are extremely stable over time.
00:04:54.240 --> 00:04:55.519
Oh, that's really cool.
00:04:55.839 --> 00:05:00.240
Um, so what type of deadweight tester does Mentcer use specifically?
00:05:00.560 --> 00:05:03.519
We use the Ruska 2465, is what we use.
00:05:03.600 --> 00:05:04.319
Yeah, cool.
00:05:04.560 --> 00:05:08.160
Can you go into why that it was the right choice for Mentzer?
00:05:08.480 --> 00:05:15.680
Yeah, so it offers one of the best uh uh uncertainties that you can get, which is double one percent of reading down to one PSI.
00:05:15.839 --> 00:05:16.000
Okay.
00:05:16.160 --> 00:05:21.360
Um, at least for the type of piston we're using when we're using it um for low pressure calibrations, but that's that's why.
00:05:21.680 --> 00:05:21.920
Cool.
00:05:22.079 --> 00:05:23.920
And is it gauge or absolute?
00:05:24.319 --> 00:05:27.920
We use for those calibrations, we do use the Witester in gauge mode.
00:05:28.160 --> 00:05:28.319
Okay.
00:05:28.399 --> 00:05:28.639
Yeah.
00:05:28.879 --> 00:05:41.519
Is that for across all low calibrations, or can you use other like absolute or is it so for higher pressure differential calibrations, we we could do the same thing where you could put the the bell jar in in a vacuum.
00:05:41.600 --> 00:05:44.000
Um, but but for these we just use them in gauge mode.
00:05:44.079 --> 00:05:44.160
Yeah.
00:05:44.399 --> 00:05:44.560
Okay.
00:05:44.720 --> 00:05:46.959
And do you know what the typical percent of reading is?
00:05:47.519 --> 00:05:49.839
Double one percent of reading down to one PCI.
00:05:49.920 --> 00:05:50.160
Yeah.
00:05:50.240 --> 00:05:50.319
Okay.
00:05:51.040 --> 00:05:56.639
Um, so is that why it's important to use two deadweight testers in the process?
00:05:57.040 --> 00:05:58.720
That is that is one part of it.
00:05:58.959 --> 00:06:11.120
Um I'm sure we'll get into more details later about how we're using this setup, but but really the reason we have to use two is because we are actually simultaneously applying
pressure to both the measure side of the sensor and the reference side.
00:06:11.360 --> 00:06:22.720
And we could use uh something else on the reference side, but we really haven't found anything that will give us a pretty much a stable pressure that will not change over the 30 to 45 minutes we're doing a calibration.
00:06:22.879 --> 00:06:27.199
Yeah, the the pressure that we're applying is extremely stable uh for those pressures.
00:06:27.519 --> 00:06:33.680
Okay, could you kind of walk through the process of how it's done with using the two deadweight testers?
00:06:33.920 --> 00:06:37.839
Yeah, so we'll we'll have one of the deadweight testers, and this is a complete deadweight tester.
00:06:38.000 --> 00:06:40.720
It has a base, a piston, mass set, temperature probes.
00:06:40.800 --> 00:06:45.120
We are capturing aming conditions while we're using it, and so we just have two of those systems.
00:06:45.199 --> 00:06:51.439
Uh one system is connected to the pressure side of the or the measure side of the sensor, and one is connected to the reference side.
00:06:51.600 --> 00:07:02.319
Uh, we will float them to a static pressure and determine an offset, and then we will keep the reference pressure the same and vary the pressure on the measure side.
00:07:02.480 --> 00:07:02.639
Okay.
00:07:02.800 --> 00:07:16.720
So, as an example, if we wanted to generate 0.2 psi of differential pressure, is what we would call it kind of pressure, we would put uh you know one point 1.0 psi on the reference side and 1.2 psi on the measure side, and that will give us a difference of 0.2 psi.
00:07:17.120 --> 00:07:23.519
Um, so can you touch on some of the different factors that could affect low pressure calibration, like environmental factors, anything?
00:07:23.839 --> 00:07:32.399
Uh two of the the big ones are um you know changes in the pressure of the room and vibrations.
00:07:32.480 --> 00:07:37.199
Um, and it was really funny in the old building, uh, before we built this this much, much better building.
00:07:37.600 --> 00:07:42.879
Uh, we would typically have to run to the machine shop and let and tell them, hey, can you guys close the door in the machine shop?
00:07:42.959 --> 00:07:44.720
Because it would be affecting low pressure calibration.
00:07:44.800 --> 00:07:46.480
Because we had it kind of shared the ceiling.
00:07:46.639 --> 00:07:46.800
Okay.
00:07:46.959 --> 00:07:50.959
Um, so now, at least in this case, the low pressure room actually has a sealed room.
00:07:51.439 --> 00:07:53.759
It does not share the ceiling with the rest of the room.
00:07:53.920 --> 00:07:57.279
So once you shut the door, it's completely isolated from everything else.
00:07:57.920 --> 00:08:05.439
Um, but you can see, like if you have a device connected, if you open that door, you will see a spike in pressure on the sensor because they're just that sensitive at that lower pressure.
00:08:05.600 --> 00:08:08.079
Um, the other major factor is vibrations.
00:08:08.240 --> 00:08:16.639
If I'm in there doing a calibration, I actually do not have the keyboard on the desk because if I go to type anything, you can actually see fluctuations in the pressure reading.
00:08:16.720 --> 00:08:17.199
Oh, wow.
00:08:17.360 --> 00:08:26.319
Um, so we actually have the entire slab that the low pressure room is on, which also uses the our mass calibration lab is also in the same on the same slab.
00:08:26.480 --> 00:08:28.639
And it's actually isolated from the rest of the building.
00:08:28.800 --> 00:08:33.840
Um, because we've got the train tracks on one side and I-35 about a hundred yards on either side of us.
00:08:33.919 --> 00:08:36.240
Uh so that was try to to minimize vibrations.
00:08:36.320 --> 00:08:36.399
Yeah.
00:08:36.879 --> 00:08:42.080
So even 100 yards away, if a train goes by, those vibrations we could see it on the device in our test.
00:08:42.480 --> 00:08:42.639
Wow.
00:08:42.720 --> 00:08:43.679
So they're just that sensitive.
00:08:43.919 --> 00:08:44.960
They are that that sensitive.
00:08:45.039 --> 00:08:47.919
It's really the the resolution of the sensors are are so large.
00:08:48.000 --> 00:08:50.559
Like if we had a 15 psi gauge, you may not be able to see it.
00:08:50.639 --> 00:08:56.639
But if you have a 10 inches of water sensor that's 0.36, you have so much more resolution and you can see these small changes in in pressure.
00:08:56.720 --> 00:08:57.440
Yeah, definitely.
00:08:57.679 --> 00:08:58.159
Wow.
00:08:58.399 --> 00:09:08.000
So when you moved buildings, were there any other factors like that you guys kind of had to remedy, or was it mainly just the vibrations and the pressure in the room?
00:09:08.240 --> 00:09:10.720
Uh mainly was those two, specifically for low pressure.
00:09:10.799 --> 00:09:10.960
Yeah.
00:09:11.279 --> 00:09:15.840
Vibrations, uh, changes in in the pressure of the room itself were the two big ones.
00:09:15.919 --> 00:09:16.080
Yeah.
00:09:16.240 --> 00:09:17.600
So temperature, nothing like that.
00:09:18.320 --> 00:09:29.039
Well, we do we do consider those, but as far as what was, you know, something that could affect in real time the calibration really was the vibrations and the and the changes in uh the pressure of the room.
00:09:29.200 --> 00:09:29.279
Yeah.
00:09:29.519 --> 00:09:30.159
Very cool.
00:09:30.399 --> 00:09:35.279
So do you know what industries use low pressure calibration or how many?
00:09:35.600 --> 00:09:37.279
Yeah, I know of two, actually.
00:09:37.360 --> 00:09:39.919
Um, one of them is for surgery rooms.
00:09:40.000 --> 00:09:44.240
They want the pressure in to in the room to be higher than the pressure outside of the room.
00:09:44.399 --> 00:09:48.879
That way nothing from outside the room comes inside the room to canam contaminate the surgery room.
00:09:49.039 --> 00:09:49.200
Right.
00:09:49.360 --> 00:09:56.159
The other use case is kind of the opposite, where if the room is for quarantine, you want the pressure to be lower in that room so nothing is getting out.
00:09:56.240 --> 00:09:59.840
So you want all the air coming into that room so that nothing, no viruses or anything get out.
00:09:59.919 --> 00:10:02.080
But those are two two use cases I'm familiar with.
00:10:02.320 --> 00:10:02.720
Very cool.
00:10:02.879 --> 00:10:03.200
Yeah.
00:10:03.440 --> 00:10:06.480
Um, well, Philip, those are all the questions I had for you today.
00:10:06.639 --> 00:10:08.720
Thank you for shedding some light on this topic.
00:10:08.799 --> 00:10:10.159
And I know I learned a lot.
00:10:10.320 --> 00:10:12.480
Hopefully, people watching learned something.
00:10:12.639 --> 00:10:13.679
So thank you for coming on.
00:10:13.759 --> 00:10:14.559
We really appreciate it.
00:10:14.720 --> 00:10:15.519
Thank you for having me.
00:10:16.799 --> 00:10:17.759
Thank you for joining us.
00:10:17.919 --> 00:10:19.600
We hope you enjoyed today's podcast.
00:10:20.000 --> 00:10:24.559
To listen to more topics, you can visit info.mensor.com forward slash podcast.
00:10:24.720 --> 00:10:29.519
And if you'd like to learn more about low pressure calibration, check out the white paper linked in the description below.
00:10:30.080 --> 00:10:32.879
I'm Sally Pittman, you stay classy metrology lovers.