Thread: DE Measurements

Lappa, I looked on line at the App for the probe and it looks like it might have 2 really useful features in the App for duct air speed measurements.

One is a time averaged measurements which might really help with turbulent air flow.
I downloaded the app but cannot get any further into the app without a probe.
What sort of time averaging does it do e.g. what time period and over what time intervals.
Does it output any sort of standard deviation or range of results.
I realises it can give you all the data but just wondering what it shows on the screen rather than having to post-process the data.

The outer is multi point measurements in square and round ducts.
Does it allow for radial multi point measurements on round ducts.

If these are catered even on a rudimentary basis then the slightly larger accuracy limits would be a minor detail and it might even tempt me to get one
Its these measurements that take up time in doing duct measurements.

Originally Posted by BobL

If you say there is no uncertainty then clearly you have no idea about uncertainty.

I think we may be relating the word uncertainty to different situations. My statement was "There is no uncertainty with the equipment we currently use" . I was referring to the equipment operating correctly and the readings being correct. If I was asked to take readings to diagnose a problem and I was uncertain I would get the correct readings from my equipment, I wouldn't/couldn't do the job. I do understand the role of tolerances of both equipment and systems in analysing readings. I also understand that when diagnosing unusual problems readings may not be what you expected BUT I believe the instruments I'm using because I've used them for sometime and have them checked annually.
If I was buying new equipment to do a task and I was uncertain that it would correctly do the job, I wouldn't buy it - that's why we have "in shop" demonstrations on systems we know, before purchase.

Relying on instruments and new technology to fix uncertainty is inherently unreliable.

Once again, I rely on my equipment to give me accurate readings. The use of new technology and equipment to speed up a task or increase accuracy is something that is becoming more important as our systems become more complex, however an understanding of the systems, their operations and parameters is paramount.

I am reminded of the dudes in a lab who bought the $1000000 German mass spectrometer because it could measure to 0.000001% when I asked them what their external uncertainty was he quoted his internal uncertainty - in other words he didn't know the difference.

The main uncertainty in DE system measurements is usually unrelated to instrumentation uncertainty but due to the variability caused by turbulence.
Unless this is understood and allowed for then folks will go round and round in circles making meaningless changes to DE systems based on meaningless measurements

Totally agree with this last statement.

Originally Posted by DomAU

Lappa, no measurement can be 100% certain. There is a level of uncertainty associated with any measured value, whether it is reported or not. Uncertainty is usually expressed as an expanded uncertainty, usually in +- percentage terms (or measurand units) at a 95% confidence level and an associated coverage factor (k) and degrees of freedom.
For example a properly considered measurement would report a value of 20m/s, uncertainty of +-1.4m/s at 95% confidence and coverage factor k =2. This would then inform the user of the measurement on how to best treat this result and/or allow them to incorporate this measurement uncertainty into their greater uncertainty if this was to form part of a larger test/measurement. Sorry it's hard to write a good explanation on a damn phone.

I see where you are coming from, but i was referring to me believing in what my instruments are telling me. Rather that repeat myself, see my reply to BobL re my use of the word uncertainty.
I believe what you relate to above could broadly be called a tolerance. I, for one, never believe a manufacture that tells me that the Vref should be 5V. That makes 4.99v or 5.01v incorrect and no 5v source or Voltmeter is that accurate in automotive.

Originally Posted by BobL

FWIW I note the instrument is made in China, probably by a Testo subsidiary.
This is not a criticism, just reporting what I saw on the suppliers website.

As is most equipment including what is/was the king of meters and handheld scopes used in automotive - Fluke

Bob, I'll look over the settings and run some tests over the next couple of days and get some data and screen shots and post them.

Cheers

Lappa
Thanks for tracking down a meter that may be something we have all wanted for some time. Inexpensive and capable.

As the handle suggests I was an quality inspector of manufactured aircraft parts. I was more like you in that I used the measuring tools to verify that the parts I was checking were good. Unlike you I couldn't trust the accuracy of the tools I used, even when they were recently back from calibration. So I would verify they were still in calibration each time I started to use them and at intervals during the session against standards. If I found that there was a drift of the reading I would have to retake all the measurements since the last good check.

Now I understand where the boys are coming from with their use of "uncertainty" as it deals with all the posible reasons that influence the accuracy of an instrument and its application. Some of it may seem like a tolerance but it takes into account much more. It is all the factors that influence the accuracy of a reading or measurement, from the atmospheric conditions, the way the instrument is used, the training, skill and consistency of the operator, source of it's power supply, and so on. It's a long "so on" list. So even if the tool is a real good one, there are factors that make the reading you just took "uncertain" beyond what the reading says. They know that if that isn't at least partially understood the wrong conclusions can be drawn from the numbers and we may take the wrong actions to correct or think none are needed.

In all likelihood for the purposes you are using the instalment and your background you will not have to many if any issues but if you do get readings that aren't making sense or are not what you expect it will be because some of these uncertainties are at play.

Looking for ward to seeing how it works for you.
Pete

Originally Posted by QC Inspector

Lappa
Thanks for tracking down a meter that may be something we have all wanted for some time. Inexpensive and capable.

As the handle suggests I was an quality inspector of manufactured aircraft parts. I was more like you in that I used the measuring tools to verify that the parts I was checking were good. Unlike you I couldn't trust the accuracy of the tools I used, even when they were recently back from calibration. So I would verify they were still in calibration each time I started to use them and at intervals during the session against standards. If I found that there was a drift of the reading I would have to retake all the measurements since the last good check.

Thanks for the reply. I use both mechanical and electrical tools/instruments for measurements. When you speak of constantly checking calibration and drift I can relate that to eg. micrometers. If i'm measuring up engine components, I know the heat of the day, the heat from my hand, can cause the micrometers to expand and the reading drift, so in those circumstances I, like you, am constantly checking the calibration (zero point) during the course of measurements. The reason for this is that the measurement tolerances are extremely fine which I can relate to your experience with aircraft components.
As related to my electrical measuring instruments, they are checked yearly, but I know problems can occur in between the checks. For example, when reading resistance, I always check lead resistance before measuring and after measuring. I'm lucky enough to have two or more of the instruments so I can always check one against the other if I'm uncertain of a result. I also have access to precision power supplies and function generators but I cant remember the last time I had to use one to check a meter.

Originally Posted by QC Inspector

Now I understand where the boys are coming from with their use of "uncertainty" as it deals with all the posible reasons that influence the accuracy of an instrument and its application. Some of it may seem like a tolerance but it takes into account much more. It is all the factors that influence the accuracy of a reading or measurement, from the atmospheric conditions, the way the instrument is used, the training, skill and consistency of the operator, source of it's power supply, and so on. It's a long "so on" list. So even if the tool is a real good one, there are factors that make the reading you just took "uncertain" beyond what the reading says. They know that if that isn't at least partially understood the wrong conclusions can be drawn from the numbers and we may take the wrong actions to correct or think none are needed.

In all likelihood for the purposes you are using the instalment and your background you will not have to many if any issues but if you do get readings that aren't making sense or are not what you expect it will be because some of these uncertainties are at play.

Looking for ward to seeing how it works for you.
Pete

When I was referring to using tools/instruments, I was referring to myself and what I do for a living so your statement "It is all the factors that influence the accuracy of a reading or measurement, from the atmospheric conditions, the way the instrument is used, the training, skill and consistency of the operator, source of it's power supply, and so on" is fully understood.

I also understand the fact that if I get readings that don't make sense or re not what you expected you need to recheck both the readings and equipment. Generally, for me, things like atmospheric conditions have already been taken into account. This is one of the reasons I have two multimeters. If I get an unexpected reading, I can check it with the other meter. Looking back over many, many years, I can only remember about a dozen instances where i have taken a measurement, be it electrical or mechanical where I have gone " Whoooo - didn't expect that!". Funnily enough, none of those involved equipment or testing procedure fault.

Fortunately, in my game, basic systems are pretty well known by now and to some extent, outcomes predictable. Systems designers, like Bosch, design systems for many manufacturers so sensors, actuators, although different in looks, basically function the same way so I guess I'm lucky in that respect.

Thanks again for your reply

I was meant also for the many following that are puzzled by the way threads like this take off in a tangent from where it began as much as for you. You grasp it.

Pete

BobL, had a chance to play with the Testo today and learn the various tests and screens. I have snapped the screens to give you an idea of how the data is displayed and what tests it is capable of and what it cant do. I also carried out tests on my system which I will cover in a separate post.
First of all, you have a choice of a Basic Screen or screens associated with the various Testo tools you can link to and the various tests they can do.
IMG_0203.PNG
You can choose your favourites so I selected Basic and Volume Flow (duct)

1) Basic has three test screens, List, Trending and Table. The readings update every second. It only shows air velocity and temperature (you can change the units they are measured in)
List shows single readings IMG_0199.PNG


Trending shows a graph IMG_0200.PNG

Table shows all readings 1 second apart IMG_0201.PNG

All data can be exported as a csv, PDF or graph as a jpeg.

In the Volume Flow (duct) screen, you have more choice of settings plus it calculates Volume flow in CFM or what ever unit you wish. You can turn OFF temperature or air velociy but not Volume flow.
IMG_0219.PNG

You have a selection of three measurement screens. Single, Multi-point Average and Timed Average. In this screen you can also program in round or rectangular ducts and their size. You can also select whether you are measuring Supply Air or Return Air.

IMG_0211.PNG

Single screen - this is the one I have found myself using most often. It has two screens; List or Results

List shows the current results being measured.

IMG_0217.PNG
First click of the button on the meter starts the measurement, Second click Holds and displays the average of the measurements taken over the time that it was measuring. Click again and its starts measuring again and clicking again Holds the average of the second round of measurement.

The Results screen is the one I use most often as it displays each of the "snapped averages" down the screen. This is great as you transverse across the duct to measure from one side to another.

IMG_0221.PNG

The current set of readings is cleared by selecting the second icon from the left on the bottom task bar, the third from the left has the same function as the button on the instrument and the one on the right allows you to export the data as a CSV or a PDF. The one on the left is for changing settings.

NB. Please see my post below as in this mode it does not average - apologies.

Thanks for the info.

I could see the first couple of screens on the App but was stuck after that.

Originally Posted by Lappa

The Results screen is the one I use most often as it displays each of the "snapped averages" down the screen. This is great as you transverse across the duct to measure from one side to another.

IMG_0221.PNG

That's one I was really hoping it would do as this takes by far the most time and helps even out the effect of turbulence.
Can you go back and see the individual data points of each of the averages. I'm was hoping it displayed these plus a Std Dev, or range, so that some indication of a final uncertainty could be determined for each annual cylinder of air air flow.
No biggie though if the individual points are available.

We could also suggest they add that in an upgrade of the software.

After further testing this morning, I have found I unfortunately provided some wrong info on the Testo Volume Flow (duct) - single measurement post.It in fact does not do an average of the readings between clicks. It actually stores the measurement when you click it ie. on first click it starts reading and on the second click it holds/stores the measurement it saw on that "hold" click. As the name implies - single measurement. Apologies

I have put a note in the above post to this effect.

It will, in fact, do an average of the readings, but it's more complex and I will describe the process in my next post.

In multi-point average you can take a series of readings and then display an average of these readings. There are two selections, List and Results.

In the List selection it shows Volume Flow, Temp and Air velocity. You can turn off Air velocity and Temp if you wish.

IMG_0213.PNG
This displays constantly varying readings conditions change while measurements are being taken.

As you click the instrument button it snaps and stores the readings (each individual stored reading is not displayed). In the grey band near the top of the display, you can see how many samples you have taken (in the above example its 10).

The Orange disc on the display at the bottom centre of the screen above, performs the same function as the button on the instrument.

On the RH side of the Orange disc there is a circle with a square inside. When you press this, it averages the readings for this series of measurements, and displays that averaged result in the List screen.


Once you average a set of readings, the Orange disk changes to an arrow which you have to press to start a new series of readings.


IMG_0223.PNG

You will also see a number in Blue in the Results heading. This shows how many "averaged" results you have stored for this set of tests.

In the Results selection, it lists each of the "average" tests you have carried out.

IMG_0221.PNG


The second icon from the bottom LH side (page with a turned down corner and + symbol) clears all current tests results.

Once again, all data can be exported as a PDF or an Excel csv file


Unfortunately what you can't see, even when you export the data, is the individual readings that were taken to reach each averaged result so no min, max or STD deviation.

The last measurement in Volume Flow is Timed Average

As the name suggests, it takes a series of readings (1 per second) over a time period selected by the operator.

The opening screen looks like so.


IMG_0231.PNG

To start the timed measurement, you click the button on the instrument or the arrow symbol in the bottom middle of the display. A timer displays at the top of the screen.

During measuring the arrow had changed to a "press to pause" symbol

IMG_0232.PNG

Pressing the instrument button or the pause symbol on the display stops the sampling

IMG_0233.PNG

If you press the button or press the arrow simple on the display, the instrument will continue to measure adding more samples to those taken previously and the timer advances.

To store the average for that first test, you need to select the circle with the square at the bottom of the display. This can be done while sampling or when paused. The display now shows the average of the readings. You will notice that each reading had a circle with a line through it indicating it is the averaged result. The Result table show the number of tests done, in this case 1.

IMG_0234.PNG


Result Screen

This just shows the results of the tests in table form.
IMG_0237.PNG

The problem with this display and all the test displays and printouts is they don't show what test is being carried out, and in this last example, how long the tests ran for which I think would be important as its labelled as a Timed Average.