Measuring CFM?

[vaclab]

cheesewonton,

If you are actually testing a wide variety of motors with significantly difference performance ratings and only getting 52-56 hose CFM there is a HUGE problem somewhere. You may have a very leaky hose (or some other catastrophic damage/leak) or you're not measuring correctly. I've been doing this for years and have made great efforts to match my numbers to actual OEM specs (Sanitaire, SEBO, Prolux, etc.). For example, the cheap GM8901 anemometer you see in my profile picture can give CFM numbers if you multiply the ft./min. by 0.026099. Through a series of adapters, one can easily attach any vane based tool to the vacuum/hose in question.

What really bothers me is when you falsely claimed:

"Try using a Kirby on the hose to dust with. Miserable. It doesn't pick anything up."

Kinda kills your credibility right there for a multitude of reasons. Kirby's have been picking up dust very well from the hose end for over 100 years!

Bill

 

[cheesewonton]

cheesewonton,

If you are actually testing a wide variety of motors with significantly difference performance ratings and only getting 52-56 hose CFM there is a HUGE problem somewhere. You may have a very leaky hose (or some other catastrophic damage/leak) or you're not measuring correctly. I've been doing this for years and have made great efforts to match my numbers to actual OEM specs (Sanitaire, SEBO, Prolux, etc.). For example, the cheap GM8901 anemometer you see in my profile picture can give CFM numbers if you multiply the ft./min. by 0.026099. Through a series of adapters, one can easily attach any vane based tool to the vacuum/hose in question.

What really bothers me is when you falsely claimed:

"Try using a Kirby on the hose to dust with. Miserable. It doesn't pick anything up."

Kinda kills your credibility right there for a multitude of reasons. Kirby's have been picking up dust very well from the hose end for over 100 years!

Bill

I have a low hour well cared for Avalir 2 I dote on and while it does an ok job on our big area rugs and with the dusting attachment attached to the bottom of the nozzle does a good job on hard floors, something I didn't expect but am happy about, it is not very good for above floor cleaning. I was surprised given the reverence for Kirby among vacuum enthusiasts but in our home with our caliche dust and pet hair the Kirby cannot match a good canister vacuum for dusting. The thing only generates 30 inches of water lift at the hose end, the lowest of any vacuum I have tested. It is inadequate. I had high hopes because the Kirby dusting brush is pretty nice but it left a lot of dust.

I have small high powered Japanese canister vacuums from Panasonic and Hitachi that run on 100 volts and they have vastly more suction and airflow than any Kirby. So does a good running Kenmore Elite or either of my big Swedish Luxes, the D795 and D820, that I brought over from Japan. All of these have more cleaning power than anything from Kirby. The little Japanese vacuums, bagged canisters with power nozzles, peg a BAIRD meter at the bottom of a fully extended wand. Even a Kirby can't peg a BAIRD meter at the end of their hose much less down at the bottom of the wand. And they peg that BAIRD meter with a loud click. That's power! That is what I am accustomed to using to dust with. I am not trying to offend anyone here but it is the truth.

 

[Vacuum Facts]

Ive been led to believe that even if a vacuum has high CFM it doesn't ensure success or offer best performance - again like air watts, its all about whether the vacuum in question has a sealed suction unit.

This is correct. Suction (pressure difference) within the cleaner head, specifically, is the critical aerodynamic parameter in a Venturi system that principally determines the acceleration of particles necessary for their removal. High air currents within these systems that act to reduce suction are actually bad and measuring volumetric air current into a machine indicates a common misunderstanding of the relevant fluid dynamics. I have a full lecture which details the formal science of how vacuum cleaners work in a way that I hope is accessible. I derive from first principles what physical parameters determine cleaning performance—which is what counts, and cover everything downstream, including filtration. Take a watch and if you need me to clarify any technical points, I can.

 

[panasonicvac]

I have a low hour well cared for Avalir 2 I dote on and while it does an ok job on our big area rugs and with the dusting attachment attached to the bottom of the nozzle does a good job on hard floors, something I didn't expect but am happy about, it is not very good for above floor cleaning. I was surprised given the reverence for Kirby among vacuum enthusiasts but in our home with our caliche dust and pet hair the Kirby cannot match a good canister vacuum for dusting. The thing only generates 30 inches of water lift at the hose end, the lowest of any vacuum I have tested. It is inadequate. I had high hopes because the Kirby dusting brush is pretty nice but it left a lot of dust.

I have small high powered Japanese canister vacuums from Panasonic and Hitachi that run on 100 volts and they have vastly more suction and airflow than any Kirby. So does a good running Kenmore Elite or either of my big Swedish Luxes, the D795 and D820, that I brought over from Japan. All of these have more cleaning power than anything from Kirby. The little Japanese vacuums, bagged canisters with power nozzles, peg a BAIRD meter at the bottom of a fully extended wand. Even a Kirby can't peg a BAIRD meter at the end of their hose much less down at the bottom of the wand. And they peg that BAIRD meter with a loud click. That's power! That is what I am accustomed to using to dust with. I am not trying to offend anyone here but it is the truth.

Both of my G4s' (one was my grandparents') did a fine job with dusting at my grandparents' house. Even better than their Dirt Devil bagless that has more suction than their G4 when I remember fitting the G4 dusting brush on the Dirt Devil to dust.

 

[vaclab]

cheesewonton claimed:
Even a Kirby can't peg a BAIRD meter at the end of their hose much less down at the bottom of the wand.

As usual, completely false information. Watch this very old Sentria II video.

My Best Canister Airflow Is


Most Kirby's made in the last 15 years or so can EASILY "smack a 10" on a Baird airflow meter. Again, there is something wrong with your machines. Maybe you have a 240 volt Kirby plugged into a 120 volt line? Also, SUCTION barely matters. It's a pressure, not a movement. Newer Kirbys test out about 30" at the nozzle and 40" at the hose. More, much more than necessary actually.

Why don't you make a video showing your poorly performing Kirbys? I've made over 1500 so far. And here's a kicker: Kirby, Sanitaire and Royal's have so much airflow (paired with low suction) that they can clean the carpet UNDER the carpet. Yup, made many videos clearly showing this.

Here's an old video showing a Sentria II cleaning flour from UNDER a medium pile carpet:

Kirby Sentria II Flour Under the Carpet Test


Again, I have hundreds of videos clearly showing high airflow/low suction wins. Even put all these measurements into a nifty spreadsheet.

Bill

 

[vaclab]

I haven't posted the Baird meter CFM numbers in years, so here it is once more. The meter MUST be periodically cleaned so the disk doesn't get stuck though.



Bill

 

[cheesewonton]

When the Kirby sales person was in my home his new Avalir 2 pulled 9.5 on his BAIRD meter at the end of the hose. My personal Avalir 2 won't pull 10 either. That is the honest truth. Consider I live up in the high desert so the air is thinner.

 

[cheesewonton]

I haven't posted the Baird meter CFM numbers in years, so here it is once more. The meter MUST be periodically cleaned so the disk doesn't get stuck though.

View attachment 168420

Bill

What kind of anemometer are you using? Are you able to calibrate it for different diameter hoses? That matters a lot. If you are using a propeller anemometer your measurements are not accurate. Like I mentioned above there are people posting airflow numbers that are beyond the rated performance of the motor in the vacuum being tested at a wide open orifice, much less at the smaller orifice sizes of anything but a shop vac. .

All the anemometer measures directly is air velocity. It has to know the square feet of the orifice to accurately calculate airflow. You can have a given velocity of air but if the hose is only 28mm ( Tristar for example ) that velocity the anemometer sees represents a lot less airflow than the same velocity coming out of the 38 mm hose on my Electrolux Epic 8000 or the 46 mm hose on one of my Japanese canister vacs. My anemometer requires the user to enter the area of the opening in square feet. On my anemometer an Electrolux Silverado produces around 66-67 cfm at the hose end, using a ten foot long vinyl hose. That figure is a bit high. The true airflow would be slightly less because the opening on an old Electrolux 1 1/4 inch diameter hose has an area of 0.0088 square feet where my anemometer only goes down to 0.01 square feet. On my BAIRD meter the SIlverado is it pulling 6-6.5 at the end of the same hose, same as your Olympia did. So that is consistent.

 

[blackheart]

Okay before I add to this just know I don't consider my measurements to be the be all end all I think of them as a Ballpark as we're going to have environmental factors that effect our measurements, from electrical variants to environmental factors like elevation and I'd imagine things like humidity and what not also effect the density of the air.

Anyways Bill's multiplier of .026099 for the GM8901 vane style anemometer was calculated using the area of the vane. we're measuring how much air is passing through that vane space. Now if we had two machines of equal hose airflow with two different hose diameters they would still measure the same FPM wouldn't they? Especially with the cone adapters we use to allow that air to better fill that space.

 

[vaclab]

cheesewonton,

Blackheart is correct...now onto other matters.

This will be my last response in this thread. You literally have no idea what you're talking about because you obviously have never been in a lab, written college grade reports and been overseen/graded by PhD professors. I have (think graduate level). I have also taught Math, Physics and various computer science (think programming/networking/hardware) courses in High School and College. I actually make great efforts to obtain the most accurate, repeatable laboratory grade measurements as a rule on my channel. That's why I started it back in May of 2017.

I have two vane type anemometers, a GM8901 and a HP-846A. The HP-846A has a built in calculation (CFM) function, the GM8901 does not, so you use the ratio I stated before. You will see these in my videos on my YouTube channel.

From cheesewonton: "If you are using a propeller anemometer your measurements are not accurate."

Completely false statement made by someone who isn't familiar with the equipment. Stop repeating it!

At this point I can guarantee that you don't know how to use your hotwire anemometer either. Have you calibrated it for your environment?
1) Elevation correction
2) Temperature/Humidity correction
And don't forget a standardized supply load voltage for your vacuums.

Through using a series of cone shaped adapters (easily made), ANY vacuum port can be fitted properly to ANY vane adapter. Basic Physics dude.

My test lab typical specs: loaded test voltage range (120-121 VAC), temp. 67-69 F, humidity 35%-45%, elevation 840 feet (close to USA populated avg. of 636 feet).

Here's an elevation versus pressure pdf
https://www.nwflowtech.com/media/0y0aizb3/nwft-barometric-pressure-vs-altitude-table-122120-v2.pdf

I've spent about 1000 hours in high-end labs (chemistry, physics, biology (yuck)) before I fully understood how to setup my own correctly.

Best of luck in your testing journey but you'd better learn how to use your equipment properly and refrain from posting false info which you have here.

Bill

 

[cheesewonton]

cheesewonton,

Blackheart is correct...now onto other matters.

This will be my last response in this thread. You literally have no idea what you're talking about because you obviously have never been in a lab, written college grade reports and been overseen/graded by PhD professors. I have (think graduate level). I have also taught Math, Physics and various computer science (think programming/networking/hardware) courses in High School and College. I actually make great efforts to obtain the most accurate, repeatable laboratory grade measurements as a rule on my channel. That's why I started it back in May of 2017.

I have two vane type anemometers, a GM8901 and a HP-846A. The HP-846A has a built in calculation (CFM) function, the GM8901 does not, so you use the ratio I stated before. You will see these in my videos on my YouTube channel.

From cheesewonton: "If you are using a propeller anemometer your measurements are not accurate."

Completely false statement made by someone who isn't familiar with the equipment. Stop repeating it!

At this point I can guarantee that you don't know how to use your hotwire anemometer either. Have you calibrated it for your environment?
1) Elevation correction
2) Temperature/Humidity correction
And don't forget a standardized supply load voltage for your vacuums.

Through using a series of cone shaped adapters (easily made), ANY vacuum port can be fitted properly to ANY vane adapter. Basic Physics dude.

My test lab typical specs: loaded test voltage range (120-121 VAC), temp. 67-69 F, humidity 35%-45%, elevation 840 feet (close to USA populated avg. of 636 feet).

Here's an elevation versus pressure pdf
https://www.nwflowtech.com/media/0y0aizb3/nwft-barometric-pressure-vs-altitude-table-122120-v2.pdf

I've spent about 1000 hours in high-end labs (chemistry, physics, biology (yuck)) before I fully understood how to setup my own correctly.

Best of luck in your testing journey but you'd better learn how to use your equipment properly and refrain from posting false info which you have here.

Bill

Lol, I just retired from a career in one of the Navy's premier weapons labs. I made a career in R&D. I think I know my way around instruments and data.

 

[cheesewonton]

With the GM8901 to obtain an accurate airflow measurement you would multiply the speed in feet per minute by the area of the hose opening in square feet. Hoses have different diameters and thus different surface areas at the opening, so there is no one size fits all factor. If you are measuring a 1 1/4 inch diameter hose you would multiply the observed airspeed in ft/min times the surface area 0.0088 square feet. If you were measuring a 35mm hose on something like a Miele you would multiply the observed airspeed in ft/min by 0.0104. Ect. Each hose diameter will have a different square foot opening to use in the calculation.

Likewise with the HP-846A, directly from the operating manual, you have to manually enter the surface area of the hose opening of the vacuum being tested. If you are not entering the correct value the resulting CFM calculation will be incorrect. As before for accuracy there is no one size fits all factor to multiply airspeed by.

A general shortcoming with rotating vane anemometers are bearing friction and the amount of obstruction to airflow inherent in their design. Hot wire anemometers, like you find in automobile Mass Airflow Sensors do not create as much of an obstruction to the airflow so there are less losses and greater accuracy. HVAC pros will use capture hoods with differential pressure sensors or a thermal ( hot wire ) sensor.

 

[vaclab]

cheesewonton,

I simply had to respond to such a grossly ignorant comment. While I have GREAT respect for our armed forces, if what you claim is true, you couldn't pass an 8th grade Algebra or 9th grade Geometry test. Geez, what a disappointment. I posted these calculations many years ago on Vacuumland, but here they are again.

For the GM8901 anemometer:

Diameter of GM8901 Anemometer Detector = 2.1875 inches
Radius = 1.09375 inches = .0911458 feet
Detector Area = 3.1415926 x .0911458 x .0911458 = 0.0260990 (rounded)

Soooo, if I obtain a reading of say 5747 ft/min, that converts to about 150 CFM.
5747 (ft/min) x 0.026099 (area) = 149.9 CFM

I also noticed you refused to provide ANY of your lab specs after I stated mine (as expected). No pics, no vids, just false claims. Par for the course I suppose.

Lastly, you did mention a 9.5 on the Baird meter, which equates to 109 CFM. And you somehow can't manage to clean (dust) with 109 hose CFM? What world do you live in anyway? The typical hose CFM is around 80 CFM, so your Kirby has plenty of above the floor cleaning power.

Bill

 

[vaclab]

Here we go again. If you had bothered to read/understand my previous posts and/or view some of my extensive YouTube videos, you would clearly see the cone adapters I use. These type of adapters have been used in the industry for decades. They adapt (without interference) diameter "A" to diameter "B" very easily. Try and make one rather than whining so much. How on Earth do you think that airflow measurements were made say, 50 years ago?

Example #1: if a hose is 1" diameter and you want to measure its flow, you can with any measuring tool that is 1" or larger. Yup, you can create a 1" to 2" cone adapter and proceed to mate it to a 2" vane.

Example #2: if you have a large 12" tube/square (say an A/C vent), you can use the same 2" vane detector to measure part of the area (say 10%), then multiply by 10.

 

[cheesewonton]

You are multiplying airspeed in ft/min times 0.26099. That is the diameter of the fan. You are using the instrument incorrectly. You need to multiply airspeed by the diameter of the hose opening. The operators manuals for both anemometers ( I looked at them ) tell you to multiply airspeed by the surface area of the duct. In this case the duct is a vacuum hose. This explains your overly high airflow readings. That is also how my anemometer explains the airflow calc.

 

[vaclab]

Here we go again...and you typed 0.26099 instead of 0.0260990. Different by a factor of 10. So here's a picture of the least expensive way to measure vacuum airflow (i.e. the hose is smaller than the vane). Note: my wife isn't holding everything exactly straight.

The vane is a detector and the cone (properly sealed at both ends of course) will spin at a speed proportional to the flow at its given diameter. That same airflow enters the hose at a different diameter. The total flow of the air MUST be the same (minus about 0.25 CFM for the blade drag of the GM8901). Sooo, if 100 CFM enters the vane, 100 CFM MUST enter the hose.


One method of using a small vane to calculate airflow in a larger airduct is to first find the area of the duct (i.e. H x W). Find out the ratio of duct area divided by vane area and now you have a simple multiplier you can store in your head.

For example: if the area of the duct is 10 times the area of the vane, simply take your (calibrated CFM) measurement and multiply by 10. Say the vane reads 10 CFM, then the total airflow through the duct would be 100 CFM. Pretty easy huh?

Do you require another picture showing the vane being held up to a duct (area comparison)?

 

[cheesewonton]

Here we go again...and you typed 0.26099 instead of 0.0260990. Different by a factor of 10. So here's a picture of the least expensive way to measure vacuum airflow (i.e. the hose is smaller than the vane). Note: my wife isn't holding everything exactly straight.

The vane is a detector and the cone (properly sealed at both ends of course) will spin at a speed proportional to the flow at its given diameter. That same airflow enters the hose at a different diameter. The total flow of the air MUST be the same (minus about 0.25 CFM for the blade drag of the GM8901). Sooo, if 100 CFM enters the vane, 100 CFM MUST enter the hose.
View attachment 168426

One method of using a small vane to calculate airflow in a larger airduct is to first find the area of the duct (i.e. H x W). Find out the ratio of duct area divided by vane area and now you have a simple multiplier you can store in your head.

For example: if the area of the duct is 10 times the area of the vane, simply take your (calibrated CFM) measurement and multiply by 10. Say the vane reads 10 CFM, then the total airflow through the duct would be 100 CFM. Pretty easy huh?

Do you require another picture showing the vane being held up to a duct (area comparison)?

That is incorrect. The diameter that matters is the diameter of the hose opening. My anemometer is a simple probe. There is no shroud like the propeller ones. You don't use an adapter cone. Just hold the probe in front of the hose with the probe oriented for the direction of airflow ( a dot on the top of the probe gives you orientation ).

For use measuring airflow for an HVAC situation the technique is to take airspeed measurements in feet per minute at the four corners of a duct and at the center, then average those readings. If it is a round duct like my house has you would take readings at four of five points around the perimeter and at the center and calculate their average. Take the average airspeed you just calculated and multiply that by the area of the duct in square feet. For a smaller duct like a vacuum hose you do the same, but you have a smaller duct. But the salient area is the area of the duct, which is the hose opening.

When an HVAC tech takes an airflow measurement using a shrouded airflow meter they too use the surface area of the duct opening, not the smaller opening of the shroud, to calculate airflow. Look at some of the CFM figures shown at the beginning of this thread. The CFMs shown for the suction inlet of the vacuum are literally greater than the motors in them can produce based on the performance charts supplied by the manufacturers. These airflow values are literally impossible for those vacuums to achieve. The numbers I calculate are less than the maximum values for a given orifice size, which makes sense given airflow losses going from the motor to the hose end.

 

[cheesewonton]

Attached is the manual for my anemometer. Read page 4. It explains how to determine airflow.

 

[vaclab]

You're simply not grasping how measuring tools work here. You should know that your hot wire doesn't actually measure airflow directly just like a vane type. It measures airspeed and then you enter in the area calculation (and type, i.e. square, round). Hot wire anemometers also have to constantly correct for temperature and humidity since the wire has to report a temperature difference via voltage/resistance. Also, how the hot wire thermometer is used will affect the readings. For example, stuffing it straight into the flow versus a vertical position. The vane type is only used one way, which simplifies things somewhat.

The vane type only cares about the air flowing through itself, not the hose or airflow box hole, etc. That's a known calculation and is given above. Then, various cones are used translate the flow to the vane in the proper manner and thusly CFM is measured.

Let me say it another way. I have gathered and created data from many OEM's and not only compared them to my readings but also have been the actual SOURCE of verification for some of those OEM's as well. In other words, My measurements stand confirmed as accurate, even by third parties. I've even compared notes with the likes of Wirecutter (see my YouTube video).

To be used correctly, a hot wire thermometer must be calibrated with various known wind sources and I'll let you Google that for verification.

The vane type doesn't care about temperature or humidity and usually needs no extra calibration.

Bottom line, your false assumption that vane types are worse (not to be trusted) compared to hot wire types is quite disappointing.

Now onto duct measurements:

My simple example was exactly that. Something that assumes an area and approximately equal flow at all points in the space. I have no desire to type a dissertation length post. Obviously, with things like corrugations, angles, vents, splitters, turbulence is introduced.

Airflow CFM Measured with a Hot Wire Anemometer!


If you watch carefully, you see the tech perform a wide range of measurements that involve a HUGE amount of human error (speed of insertion/removal, tilting, etc. Then everything has to be averaged, yippee! While this method will certainly give a general measurement of airflow, I hope you see just how sloppy the entire procedure is. What would the error bar calculations be, plus or minus 10%? Larger? Not to mention performing measurements with several open holes, geez!

Anyway, good luck with your measurements but unless you make the attempt to fully verify them (as I have for years), your results won't gain any credibility.

Bill

 

[kenc184]

It's nice to see that at least one gentleman in this cfm debate is not casting insults at the other and "throwing his toys out of the pram".

 

[cheesewonton]

All I can say is I asked the manufacturer how to use the device to measure airflow from vacuum cleaners and use it the way the manufacturer described. I e-mailed the for advice before forking out that much money to be sure it was the right tool for the job. The meter has a temperature compensation but not one for humidity. As a long time military and civil helicopter pilot before going over to the weapons side of the Navy as an Operations Research Analyst I can tell you with absolute assurance anything with a propeller lives and dies on density altitude. What is that? Air density corrected for altitude, temperature and humidity. That vane type airflow meter is very much affected by all three.

As I said before some of the airflow numbers shown on this site are beyond the capabilities of the motors in the vacuums tested. They are simply not possible. There is no suction motor made for an old Tristar CXL that will deliver over 130 cfm at the hose opening. None. Suction motors that fit into a CXL and draw ten amps under ideal conditions will pull somewhere in the high 70s to high 80s cfm through a 1.125 inch orifice with no other restriction. That is the best anyone can expect at the hose connection of a Tristar. Expect even less at the hose end. The ten amp motor, Ametek 115923, is rated at 122 cfm max at a wide open orifice. When I see a post showing airflow in excess of 130 cfm I know with certainty there is an error. This comes from literally decades of working with numbers.

Something drilled into me since junior high math classes is to look at your solution and ask if it is reasonable. Same throughout my undergraduate and graduate classes where the math became far more complicated and all throughout my professional life. I know Vaclab is upset that I called out his technique but I stand by mine. I have been hided enough in peer reviews / murder boards of my own work to be careful what I put forth in terms of calculations and make sure my own numbers are reasonable. 50 something cfm at the end of a hose on a Tristar makes sense. Over 100 cfm from that small an orifice at the end of a hose with all the loses due to restrictions along the way does not make sense.

Vaclab, before you layer more criticism on me please familiarize yourself with the actual performance the suction motors in these vacuums are capable of. I am going to attach the Ametek performance bulletins for the OEM motor in a Tristar CXL or DXL , 116884-49, and one for the Ametek 115923, the typical ten amp two stage motor people retrofit to these old vacuums. I will also add a bulletin for the higher performance 11 amp motor used in Pro Team backpack vacuums, 119347-01, that I have used in Tristars in the past. It is slightly higher performance than 115923. The interesting thing about Tristars is that regardless of which of the three motors I use the result at the hose end is a constant, boring 52 cfm.

On these performance bulletins look at the rated airflow for a 1.125 inch orifice. If your instruments are showing significantly more than the motor is rated for you are probably doing something wrong and need to re-evaluate your procedure.

 

[cheesewonton]

A couple of news stories with excellent videos about the place I just retired from. We used to do some really cool videos for public release but people up the chain of command came to the opinion we were sharing too much information so since 2018 no more "Greatest Hits" videos. If you want to poke around Youtube you can find a whole series of videos called "Pictures of US" that depict China Lake going back to its inception in 1943.

I highly recommend watching the videos. The one showing a parachute test was for NASA. We did the parachute descent systems for several recent Mars landers. And yes the tow vehicles pulling the trucks getting blown up are unmanned remote controlled O_O

https://www.twz.com/25281/what-its-...cretive-china-lake-weapons-development-center

https://www.twz.com/24940/weapons-t...e-everything-in-this-crazy-greatest-hits-reel

 

[blackheart]

This is like watching the two smart kids with different answers arguing....

I'm going to play Devil's advocate here and ask this question. When we place machines like Kirbys or Royals onto our airflow boxes and the opening from said airflow box is 2" does this then mean our measurements are more accurate vs measuring from a hose and does that also apply to power nozzles which have smaller openings before it.
If that doesn't apply to machines which have smaller openings before the nozzles wouldn't this then mean that the airflow of a direct air would be substantially higher than non direct air then.... but if it does apply then we would have a higher airflow number at the nozzle wouldn't we?

What Bill is telling us is we're measuring 100% of the air coming through the vane area. It's not quite like a duct where the opening is larger than the vane itself where we need to figure out the total air moving based upon a speed given to us by a gauge....we are seeing ALL of the air moving through that vane.

I suppose I could ask our HVAC guy about this sort of situation to see what his thoughts on it are.

 

[cheesewonton]

This is like watching the two smart kids with different answers arguing....

I'm going to play Devil's advocate here and ask this question. When we place machines like Kirbys or Royals onto our airflow boxes and the opening from said airflow box is 2" does this then mean our measurements are more accurate vs measuring from a hose and does that also apply to power nozzles which have smaller openings before it.
If that doesn't apply to machines which have smaller openings before the nozzles wouldn't this then mean that the airflow of a direct air would be substantially higher than non direct air then.... but if it does apply then we would have a higher airflow number at the nozzle wouldn't we?

What Bill is telling us is we're measuring 100% of the air coming through the vane area. It's not quite like a duct where the opening is larger than the vane itself where we need to figure out the total air moving based upon a speed given to us by a gauge....we are seeing ALL of the air moving through that vane.

I suppose I could ask our HVAC guy about this sort of situation to see what his thoughts on it are.

That is a tough question I have often asked myself watching tests of uprights on Vacuum Wars. On something like a Kirby the suction inlet from the nozzle to the fan is a good two inches or more ( I haven't measured but it is much larger diameter than the hose ). On something like a Sebo, Panasonic, my Hoover Hushtone, etc, what you get is what is available at the end of the hose where it enters the nozzle base. On a Sebo G4 or similar, pull the hose out of the wand and measure the airspeed and multiply that by the area of the hose opening. It would be analogous to measuring the airspeed at the bottom of the wand on a canister vac and multiplying it by the area of the wand opening. That is all you get at the nozzle end.

I am going to go back to your post with the airflows you claim for that Tristar CXL and calculate the airspeeds you measured. Simple algebra. Take the airflow in CFM, divide by the area of a 2 1/2 inch diameter hole and that is your airspeed in ft/mim. I want to take those airspeed numbers and multiply them by the actual diameter of the hose opening and hose end, which I have already calculated based on a measured 28 mm ( 1.1 inch ) orifice to see what kind of airflow numbers that calculation produces. Give me a few minutes to go back in this thread three pages ( ! ) and find the values to use.

 

[cheesewonton]

So here is the math head in me. I will use Blackheart's numbers for the Tristar CXL and divide his CFM numbers by 0.026099 to arrive at the airspeed values he measured. Then I will multiply these by the area of the hose opening for each vacuum based on a 1.1 inch diameter opening for the Tristar.

For the CXL Blackheart shows 132 cfm at the canister opening and 104 cfm at the hose end. If you divide these cfm values by 0.026099 you get an airspeed of 5057.665 ft per min at the canister opening and 3984.827 ft per min at the hose end. If you multiply these by the area of a 1.1 inch diameter opening, or 0.0066 square feet you arrive at 33.38 cfm at the canister opening and 26.30 cfm at the hose end. Huge difference.

My own numbers were based on the smallest duct area setting on my anemometer, 0.01 square feet. As I mentioned earlier using that setting means I am overestimating airflow for hoses with openings smaller than 35 mm since a 35 mm opening is closest to 0.01 square feet ( 0.0104 sq ft to be exact ). So for giggles lets take the 52 cfm number I get on my anemometer for my DXL ( and my EX-30, and my EXL, etc, Tristars are boring that way ) and divided it by 0.01 to find the airspeed my anemometer measured, or 5200 ft per minute airspeed, higher than what Blackheart's anemometer measured. Take that 5200 ft/min and mulitply it by the area of a 1.1 inch diameter hose opening ( measured inside diameter of the Tristar hoses I have ) and you get 34.32 cfm. Now you see that Blackheart and I are not so far off, and our numbers are well within the parameters of the publish performance for the motors you find in a CXL. It passes the reasonableness test. Also want to mention that an Ametek 115923 is rated for 96 cfm at a 1.125 inch orifice with no other obstructions. Pretty interesting to see how much the design of the vacuum degrades performance from the motor face to the hose opening.

 

[vaclab]

Cheesewonton,

Here's what you really need to spend more time doing, but I see you're on your way already. I said it before in previous posts but I'll say it one more time.

Search for manufacturers that have posted CFM and/or suction numbers (i.e. hose, canister, power nozzle, etc.) and attempt to duplicate them. Rainbow, Sanitaire, SEBO and even Dyson have periodically posted some specs. When I was first setting up my lab, I made sure to replicate/explain OEM specs fully before posting any procedures/numbers. When you first take lab classes in college, you are expected to use the instruments provided and duplicate past tests. It's one very important method of learning to be an accurate tester. If you can't obtain reasonable results, you'll need to explain why (may involve error rates, formulas, etc.).

Again, I have posted hundreds of measurement videos and more than a dozen OEM's have accepted them AS FACT. Additionally, a few OEM's have sent their products for official validation. In other words, my lab was the first to post any test results.

Do the work and you'll figure it out eventually I suppose but be ready to accept many failures as well as surprises.

You keep bringing up my TriStar numbers and you've conveniently forgotten that you don't have access my motor. The model isn't listed here. The most important measurements are at the point of cleaning! Hose and power nozzle. Get those right first.

Lastly, there are manufacturing tolerances that can vary. Some may be 10%, some less. And unless you setup a standardized environment as I have, you easily could be fairly far off. Make sure to test all AC motors plugged into a dedicated 20A line with a load voltage of at least 120VAC. My basement lab is very close to proper specs (i.e. temp, humidity, voltage, altitude).

Bill

 

[cheesewonton]

Cheesewonton,

Here's what you really need to spend more time doing, but I see you're on your way already. I said it before in previous posts but I'll say it one more time.

Search for manufacturers that have posted CFM and/or suction numbers (i.e. hose, canister, power nozzle, etc.) and attempt to duplicate them. Rainbow, Sanitaire, SEBO and even Dyson have periodically posted some specs. When I was first setting up my lab, I made sure to replicate/explain OEM specs fully before posting any procedures/numbers. When you first take lab classes in college, you are expected to use the instruments provided and duplicate past tests. It's one very important method of learning to be an accurate tester. If you can't obtain reasonable results, you'll need to explain why (may involve error rates, formulas, etc.).

Again, I have posted hundreds of measurement videos and more than a dozen OEM's have accepted them AS FACT. Additionally, a few OEM's have sent their products for official validation. In other words, my lab was the first to post any test results.

Do the work and you'll figure it out eventually I suppose but be ready to accept many failures as well as surprises.

You keep bringing up my TriStar numbers and you've conveniently forgotten that you don't have access my motor. The model isn't listed here. The most important measurements are at the point of cleaning! Hose and power nozzle. Get those right first.

Lastly, there are manufacturing tolerances that can vary. Some may be 10%, some less. And unless you setup a standardized environment as I have, you easily could be fairly far off. Make sure to test all AC motors plugged into a dedicated 20A line with a load voltage of at least 120VAC. My basement lab is very close to proper specs (i.e. temp, humidity, voltage, altitude).

Bill

Again. the airflow values shown by yours and Blackheart's technique are beyond the the published capabilities of the motors in those vacuums. That is a fact proven by the Ametek performance charts I have provided. If you are claiming airflow numbers for a given hose end size that exceed what the motor manufacturer says the motor is capable of then you have made a mistake.

I also do not for a minute trust the claims made by vacuum manufacturers. Miracle Mate claims the Platinum model has a 6 cubic feet per second airflow on their website. That translates into 360 cfm. Simply not possible. The Ametek motor in that vacuum is rated for a maximum unrestricted airflow of 122 cfm and through a 1.125 inch orifice with no other obstruction is capable of 96 cfm under ideal conditions. The inner and disposable bag plus the shape of the opening are going to reduce that number. Their publish claim is nonsense. Most of the claims vacuum manufacturers make are wildly optimistic. That is why I do my own tests. I know they are lying. But those data sheets from Ametek, along with similar ones from Electromotor Inc and Domel are reliable.

Something I learned at China Lake. I don't adjust my test to match the manufacturers claims. I test for reality assuming the manufacturer is not being completely honest. Make sure your test procedure is correct and let the resulting data speak for itself. Even in the defense world, maybe especially in the defense world, the manufacturers are not honest. Our job was to be the kill joys and show what the product or weapon could really do under real world conditions.

Some older 1970s Kenmore canister vacs had some performance data on a tag under the lid and I have found those values to be pretty close. Hard to say though when you are testing 45-60 year old vacuums. You don't expect them to nail the numbers on that tag necessarily.

 

[Vacuum Facts]

Again. the airflow values shown by yours and Blackheart's technique are beyond the the published capabilities of the motors in those vacuums.

I can help you here, as I have detailed knowledge of the physics of this situation and I feel you’re being unfairly demonised. You are correct in your observation that vane anemometers only measure airspeed, have to compute airflow, require corrections in certain physical setups, and values reported by others are likely overestimates. Using a vane anemometer in the configurations I’ve seen discussed here will unlikely yield accurate measurements of the volumetric airflow.

Furthermore, after reading through this thread, there is evidence of serious misunderstanding of important fluid dynamics principles, such as the relationship between airflow continuity in tubes and radial velocity profiles that affect the accuracy of vane anemometer readings. I’ve no confidence in the accuracy of any of the measurements shown and there are serious experimental weaknesses observable. There is also evidence of significant misunderstanding of basic physical parameters associated with vacuum cleaners, such as suction and airflow. This was all the more striking viewed alongside the mismatching credentials alleged.

Out of courtesy, I’ve produced a video that details the flaws in these kinds of measurements, quantifies the error, offers a practical solution, and correctly contextualises their relevance. Ultimately, airflow measurements, as shown in some of the amateur videos here, do not translate to cleaning performance, and the physics of dirt removal from carpets using fluid flow is seriously misunderstood. Claims to the contrary stem from a misunderstanding of what’s being measured. I encourage anyone interested to additionally watch my lecture on the topic to learn how these machines really work, which I’ve also linked to below to watch on YT. Happy to provide technical clarification but I wouldn’t entertain uncivil argument.

Accurately measuring the volumetric airflow of vacuum cleaners: .

The physics of dirt removal using a vacuum cleaner: .

 

[cheesewonton]

Eh, the problem lies in the fact that all of those vane style anemometers have a rotating vane that is much larger diameter than any vacuum hose and the area used for the airflow calculation is based on the diameter of the vane, and not the diameter of the hose. But yes because the vane is so much larger than the hose it is very hard for a vane anemometer to obtain an accurate velocity reading. Then you compound this by using the wrong area to calculate airflow.

Using a thermal, aka hot wire, anemometer and using the area of the hose opening avoids these errors. The probe in a thermal anemometer is much smaller than the hose opening and you can move it around the opening and see changes in airspeed and airflow.

I will also mention that early automotive fuel injection airflow measuring systems used swinging vanes connected to an electrical winding that changes a voltage signal to determine the volume of air entering an engine to calculate "pulse width", the amount of time to hold the fuel injector open to obtain the correct air:fuel mixture. Bosch L-Jetronic is a classic example of this kind of system. Analog and extremely reliable, but nowhere near accurate enough for fuel injected cars to pass modern emissions tests. Modern cars use a hot wire anemometer in their Mass AirFlow ( MAF ) sensor along with feedback from O2 sensors in the exhaust precisely because this is much more accurate and reacts faster to changes in airflow than those old vane style airflow meters could ever achieve. I love L-Jet for its reliability and smooth throttle response but there is no way it can meet any modern emissions standards.

 

[Vacuum Facts]

The reducer used accounts for that diameter mismatch, but introduces new problems—the physics of which are discussed in the video above. Unless you've got no changing radial velocity profile (completely uniform, laminar flow), a vane anemometer isn't really a good instrument to use unless you know the velocity profile and can correct for it. Best you can do without measuring it precisely is to apply a 20-30% correction factor.

Hot wire probes measure point locations conveniently if they're calibrated and designed to operate within their use environment. I've used many in my time and they're very convenient, although were questionable in some very harsh environments I researched in. However, a measurement within a bounded flow, such as a pipe, will read different values depending on its radial and axial location for the fluid dynamic reasons discussed in the video. If you are using a velocity measurement to calculate volumetric flow from a single location, then correction factors will still be needed that reflect the radial velocity profile; it can't be assumed the reading is radially uniform, which later calculations assume. You'd have to traverse the duct radially to correctly gauge the radial velocity profile to calculate the correct volumetric flow rate from the average velocity, following a similar integral.

It sounds like you've had an interesting history in industry.