Hi, great board! Hope to be able to contribute to it. My questions is this. I have an orifice style bench with incline and vertical manometers. I can get good flow numbers, but since the air varies from day to day, validating comparisons from different days tests are virtually impossible. Is there any formulas out there to convert the test days' baro/temp/humidity/air density etc. into standard day conditions? I figure if the formulas out there, and someone could share that info, I could make a simple spreadsheet program for us all to share.
Thanks.
Formula to standardize tests?
20 posts
• Page 1 of 2 • 1, 2
by 84-1074663779 » Sat Feb 14, 2004 7:57 pm
I am not sure that it does vary from day to day.
The same identical volume of air passes through the test piece and then through the measurement orifice, so if you crank up your flow to give a known fixed test pressure, this automatically adjusts for variations in the actual air.
Suppose your test pressure was 28" and your orifice pressure exactly 2.8" now change the ambient air density, temperature and humidity. The same 10:1 pressure relationship will still hold because the same air flows through both.
It is a ratiometric measurement method, which does not depend on ambient conditions.
The same identical volume of air passes through the test piece and then through the measurement orifice, so if you crank up your flow to give a known fixed test pressure, this automatically adjusts for variations in the actual air.
Suppose your test pressure was 28" and your orifice pressure exactly 2.8" now change the ambient air density, temperature and humidity. The same 10:1 pressure relationship will still hold because the same air flows through both.
It is a ratiometric measurement method, which does not depend on ambient conditions.
- 84-1074663779
by Vintagent » Sun Feb 15, 2004 6:59 am
I'm with rx7carl on this one. The air temperature affects the density of the gas (fluid, technically). This has a knock-on effect on its behaviour, changing all sorts of parameters regarding its flow characteristics. I think that the humidity will also play a part, in that it becomes a two-phase flow (ie there's the air (gas) and tiny suspended water droplets, though this will be quite minor. Again, this changes characteristics.
I'll have a look through some papers/books and try and get back asap. Most laboratory studies keep the room conditions constant (not always possible in the amateurs case), but I think I've seen something concerning these correction factors. They'll certainly make a difference, the question is, how much!
I'll have a look through some papers/books and try and get back asap. Most laboratory studies keep the room conditions constant (not always possible in the amateurs case), but I think I've seen something concerning these correction factors. They'll certainly make a difference, the question is, how much!
- Vintagent
- Posts: 5
- Joined: Mon Feb 09, 2004 7:52 am
by rx7carl » Sun Feb 15, 2004 12:06 pm
Well to give an example. What I do is I have a stock carb that I use only for benchmarking the flow from that day. Then I can see if my readings will be high or low depending on the air that day. I've seen it vary from 300-326 cfm's. So I know theres a measureable difference which throws off the readings. Thats one reason that racers are always re jetting their carbs as conditions change.
- rx7carl
- Posts: 11
- Joined: Sat Feb 14, 2004 1:51 pm
by 84-1074663779 » Sun Feb 15, 2004 8:06 pm
I think we might be talking at cross purposes here.
Sure the flow through anything is going to change with the relative density of whatever flows through it, no argument there. But what are we actually trying to do with our flowbench ?
My own thoughts on this are that we should try to refer measurements back to some definite permanent repeatable reference standard.
So you might say for example that my carb flows exactly the same as a thin steel plate with a one and three quarter inch hole. That thin steel reference plate becomes the standard to which you compare your carb.
Now some guy up a mountain in the tropics says that carb flows 75 CFM. Another guy in Alaska says it flows 65 CFM. They are both probably right.
I am saying it flows the same as my reference plate, which could be anything between 65 CFM and 75 CFM, depending on ambient conditions.
This is the simple beauty of an orifice bench, it takes all the fiddle factors into account. It measures the pressure drop across an orifice restriction, just the same as the test piece restriction. It measures exactly the same thing in exactly the same way. The same identical air is used, no corrections are required.
Now the problems start when instead you try to measure the ACTUAL airflow, and then have to try to correct back for all the various factors. So does that carb flow 65 CFM or 75 CFM. It cannot flow both. So which is correct ?
If you try to compare the pressure drop across the test piece, to the flow VELOCITY downstream, you are measuring two completely different things. Ambient air conditions will shift things quite a bit. It is no longer a direct ratiometric measurement. So you get to "correct" the measurement.
I would just say that I flowed it a year ago and it measured 84% on orifice 2. And I just measured it now, and it still shows 84% on orifice 2. No corrections required. I would be very happy if I could get a simple result like that.
Trying to argue that my bench reads right any yours is 2% wrong is a pretty futile argument. The more magic correction factors you add, the deeper hole you dig for yourself.
All this is wide open to opposing opinions of course. I suppose it depends on what is important to you. I just want repeatability, and to see small improvements nothing more.
Sure the flow through anything is going to change with the relative density of whatever flows through it, no argument there. But what are we actually trying to do with our flowbench ?
My own thoughts on this are that we should try to refer measurements back to some definite permanent repeatable reference standard.
So you might say for example that my carb flows exactly the same as a thin steel plate with a one and three quarter inch hole. That thin steel reference plate becomes the standard to which you compare your carb.
Now some guy up a mountain in the tropics says that carb flows 75 CFM. Another guy in Alaska says it flows 65 CFM. They are both probably right.
I am saying it flows the same as my reference plate, which could be anything between 65 CFM and 75 CFM, depending on ambient conditions.
This is the simple beauty of an orifice bench, it takes all the fiddle factors into account. It measures the pressure drop across an orifice restriction, just the same as the test piece restriction. It measures exactly the same thing in exactly the same way. The same identical air is used, no corrections are required.
Now the problems start when instead you try to measure the ACTUAL airflow, and then have to try to correct back for all the various factors. So does that carb flow 65 CFM or 75 CFM. It cannot flow both. So which is correct ?
If you try to compare the pressure drop across the test piece, to the flow VELOCITY downstream, you are measuring two completely different things. Ambient air conditions will shift things quite a bit. It is no longer a direct ratiometric measurement. So you get to "correct" the measurement.
I would just say that I flowed it a year ago and it measured 84% on orifice 2. And I just measured it now, and it still shows 84% on orifice 2. No corrections required. I would be very happy if I could get a simple result like that.
Trying to argue that my bench reads right any yours is 2% wrong is a pretty futile argument. The more magic correction factors you add, the deeper hole you dig for yourself.
All this is wide open to opposing opinions of course. I suppose it depends on what is important to you. I just want repeatability, and to see small improvements nothing more.
- 84-1074663779
by willeng » Sun Feb 15, 2004 11:50 pm
Ok guys:
Firstly i'm not going to agree or disagree with any one here because in reality it depends on how you look at things.
I have worked with super/flow benches for quite a few years & like everything they have there good points & there bad.
I built my own bench a few years ago & it has proven itself & produced race winners.
I have no doubt that i have setup my bench a little different than yours are.
I read volume, velocity & (quality of flow).
I have tried to phase out the problems i have found with other benches i have worked on.
Firstly I use (manometers) because in reality this is how you set the standard for gauges etc etc (Physics).
I have my manometers set so i can adjust for differences in weather patterns, therefore no correction factors are needed ie
I can literaly see the manometers drop if the weather is changing etc & i re-zero to correct for change, giving me the correct readings regardless, it works.
It does'nt matter if it is hot & dry or wet & cold you just adjust them to suit.
I can on my bench, calculate the potential H/P of the engine & then run it on a dyno (not mine at the moment) & taking into account volumetric efficiency etc come up with fairly close H/P figures.
This will either add fuel to the fire or help, think about it. The stop watch tells the story!
I better get back to work catch you later.
Firstly i'm not going to agree or disagree with any one here because in reality it depends on how you look at things.
I have worked with super/flow benches for quite a few years & like everything they have there good points & there bad.
I built my own bench a few years ago & it has proven itself & produced race winners.
I have no doubt that i have setup my bench a little different than yours are.
I read volume, velocity & (quality of flow).
I have tried to phase out the problems i have found with other benches i have worked on.
Firstly I use (manometers) because in reality this is how you set the standard for gauges etc etc (Physics).
I have my manometers set so i can adjust for differences in weather patterns, therefore no correction factors are needed ie
I can literaly see the manometers drop if the weather is changing etc & i re-zero to correct for change, giving me the correct readings regardless, it works.
It does'nt matter if it is hot & dry or wet & cold you just adjust them to suit.
I can on my bench, calculate the potential H/P of the engine & then run it on a dyno (not mine at the moment) & taking into account volumetric efficiency etc come up with fairly close H/P figures.
This will either add fuel to the fire or help, think about it. The stop watch tells the story!
I better get back to work catch you later.
- willeng
- Posts: 14
- Joined: Fri Feb 13, 2004 1:59 am
by 84-1074663779 » Mon Feb 16, 2004 1:15 am
rx7carl,
Have you tried using an identical measurement orifice and test piece, to see that you get identical pressure drops across both ?
On the first flow bench I ever built, I could never do this successfully. The reason being the first orifice (test piece) was in still air. The measurement orifice was seeing some pretty disturbed incoming airflow upstream, so operating conditions through the measurement orifice were rather unstable and repeatability was poor.
I am not suggesting that this is your problem, but it is something to be aware of.
In theory the design pressure drop across the measurement orifice is not really relevant, in practice though, the higher the operating pressure drop across the orifice, the more tolerant it will be of slight air disturbances upstream. The obvious disadvantage is that blower pressure and power requirements can quickly become excessive.
willeng,
Not sure what sort of manometers you are using where the zero changes with barometric pressure. These cannot be simple water filled U tubes obviously. Are you using a vacuum referenced mercury column? ie a barometer ?
Have you tried using an identical measurement orifice and test piece, to see that you get identical pressure drops across both ?
On the first flow bench I ever built, I could never do this successfully. The reason being the first orifice (test piece) was in still air. The measurement orifice was seeing some pretty disturbed incoming airflow upstream, so operating conditions through the measurement orifice were rather unstable and repeatability was poor.
I am not suggesting that this is your problem, but it is something to be aware of.
In theory the design pressure drop across the measurement orifice is not really relevant, in practice though, the higher the operating pressure drop across the orifice, the more tolerant it will be of slight air disturbances upstream. The obvious disadvantage is that blower pressure and power requirements can quickly become excessive.
willeng,
Not sure what sort of manometers you are using where the zero changes with barometric pressure. These cannot be simple water filled U tubes obviously. Are you using a vacuum referenced mercury column? ie a barometer ?
- 84-1074663779
by willeng » Mon Feb 16, 2004 5:20 am
Here's a couple of Q/basic programs that may be of help.
They were given to me.
weather corrections
Calculates Temperature
there are (2) Loops , either computes the same answer
i prefer the 9 Loop version
'| |
'| Program Name: REVERSE.BAS , QBasic/QuickBasic version |
|
'| Purpose: Compute temperature from dewpoint vapor pressure or |
'| from saturation pressure (all in inches Hg.) |
'============================================================================
WIDTH 80 '<--set screen to 80 columns , default
SCREEN 0 '<--set to text screen , default
KEY OFF '<--turn-off 25th line GW-Basic's menu text
VIEW PRINT '<--enable printing on 25th line
LOCATE , , , 7, 7 '<--set cursor to default shape
COLOR 15, 1 '<--bright white on blue
CLS '<--clear entire screen
start:
LOCATE 3, 5
PRINT "Input choices: dewpoint,vapor, or saturation pressure in inches Hg.";
LOCATE 5, 5
PRINT USING "Pressure inches Hg.= ##.###### "; psat;
LOCATE 5, 38
INPUT psat$: IF LEN(psat$) > 0 THEN psat = VAL(psat$)
IF psat = 0 THEN psat = .5216 '<--default value if no input, (60 deg.F)
x = LOG(29.9213 / psat) '<--29.9213 inches Hg. at sea-level at 59 F
b = psat
a = 672 '<- 672 Rankine, (672 = 460 + 212) , 212 F = boiling pt. H2O
c = 35.9381# '<- Constant , 35.913 works with 671.67 and 459.67 (original)
d = -1.152437# '<- Constant , best constant value for d,(-1.152437 orig.)
logicloop:
FOR cnt = 1 TO 25 '<---- 25 is best value for number of loops
t1 = x - (a - b) * c * (b ^ d)
b = b + t1 / (a * c * d * b ^ (d - 1) - c * (d + 1) * b ^ d)
NEXT cnt
dry1 = psat
FOR cnt = 1 TO 9
dry1 = ((x + (c * (dry1 ^ (-.152437)))) / (a * c)) ^ -.867726392#
NEXT
dryb = (dry1 - 460)
LOCATE 10, 5
psatf = b - 460 '<-Fahrenheit conversion from Rankine degrees (459.67 orig.)
PRINT USING "Temperature =####.#### deg.F (####.#) (2nd formula=####.####)"; psatf; psatf; dryb
GOTO start
=================================================================================
Calculates Pressure
, QBasic/QuickBasic version
'| Purpose: Compute vapor pressure or
'| saturation pressure (all in inches Hg.)
'============================================================================
WIDTH 80 '<--set screen to 80 columns , default
SCREEN 0 '<--set to text screen , default
KEY OFF '<--turn-off 25th line GW-Basic's menu text
VIEW PRINT '<--enable printing on 25th line
LOCATE , , , 7, 7 '<--set cursor to default shape
COLOR 15, 1 '<--bright white on blue
CLS '<--clear entire screen
start:
LOCATE 5, 5
PRINT USING "Dry bulb = ###.## "; dryf;
LOCATE 5, 30
INPUT dryf$: IF LEN(dryf$) > 0 THEN dryf = VAL(dryf$)
dryr = 459.67 + dryf
satp = 29.9213 / (EXP((671.67 - dryr) * 35.913 * (dryr ^ -1.152437)))
LOCATE 8, 5
PRINT USING "Saturation pressure =###.###### inches Hg."; satp
LOCATE 9, 5
PRINT USING "Saturation pressure =###.###### psia"; satp / 2.036020696#
LOCATE 10, 5
PRINT USING "Saturation pressure =###.###### mm Hg."; satp * 25.4
GOTO start
They were given to me.
weather corrections
Calculates Temperature
there are (2) Loops , either computes the same answer
i prefer the 9 Loop version
'| |
'| Program Name: REVERSE.BAS , QBasic/QuickBasic version |
|
'| Purpose: Compute temperature from dewpoint vapor pressure or |
'| from saturation pressure (all in inches Hg.) |
'============================================================================
WIDTH 80 '<--set screen to 80 columns , default
SCREEN 0 '<--set to text screen , default
KEY OFF '<--turn-off 25th line GW-Basic's menu text
VIEW PRINT '<--enable printing on 25th line
LOCATE , , , 7, 7 '<--set cursor to default shape
COLOR 15, 1 '<--bright white on blue
CLS '<--clear entire screen
start:
LOCATE 3, 5
PRINT "Input choices: dewpoint,vapor, or saturation pressure in inches Hg.";
LOCATE 5, 5
PRINT USING "Pressure inches Hg.= ##.###### "; psat;
LOCATE 5, 38
INPUT psat$: IF LEN(psat$) > 0 THEN psat = VAL(psat$)
IF psat = 0 THEN psat = .5216 '<--default value if no input, (60 deg.F)
x = LOG(29.9213 / psat) '<--29.9213 inches Hg. at sea-level at 59 F
b = psat
a = 672 '<- 672 Rankine, (672 = 460 + 212) , 212 F = boiling pt. H2O
c = 35.9381# '<- Constant , 35.913 works with 671.67 and 459.67 (original)
d = -1.152437# '<- Constant , best constant value for d,(-1.152437 orig.)
logicloop:
FOR cnt = 1 TO 25 '<---- 25 is best value for number of loops
t1 = x - (a - b) * c * (b ^ d)
b = b + t1 / (a * c * d * b ^ (d - 1) - c * (d + 1) * b ^ d)
NEXT cnt
dry1 = psat
FOR cnt = 1 TO 9
dry1 = ((x + (c * (dry1 ^ (-.152437)))) / (a * c)) ^ -.867726392#
NEXT
dryb = (dry1 - 460)
LOCATE 10, 5
psatf = b - 460 '<-Fahrenheit conversion from Rankine degrees (459.67 orig.)
PRINT USING "Temperature =####.#### deg.F (####.#) (2nd formula=####.####)"; psatf; psatf; dryb
GOTO start
=================================================================================
Calculates Pressure
, QBasic/QuickBasic version
'| Purpose: Compute vapor pressure or
'| saturation pressure (all in inches Hg.)
'============================================================================
WIDTH 80 '<--set screen to 80 columns , default
SCREEN 0 '<--set to text screen , default
KEY OFF '<--turn-off 25th line GW-Basic's menu text
VIEW PRINT '<--enable printing on 25th line
LOCATE , , , 7, 7 '<--set cursor to default shape
COLOR 15, 1 '<--bright white on blue
CLS '<--clear entire screen
start:
LOCATE 5, 5
PRINT USING "Dry bulb = ###.## "; dryf;
LOCATE 5, 30
INPUT dryf$: IF LEN(dryf$) > 0 THEN dryf = VAL(dryf$)
dryr = 459.67 + dryf
satp = 29.9213 / (EXP((671.67 - dryr) * 35.913 * (dryr ^ -1.152437)))
LOCATE 8, 5
PRINT USING "Saturation pressure =###.###### inches Hg."; satp
LOCATE 9, 5
PRINT USING "Saturation pressure =###.###### psia"; satp / 2.036020696#
LOCATE 10, 5
PRINT USING "Saturation pressure =###.###### mm Hg."; satp * 25.4
GOTO start
- willeng
- Posts: 14
- Joined: Fri Feb 13, 2004 1:59 am
by rx7carl » Mon Feb 16, 2004 11:26 am
Tony, no I havent tired that. Once I calibrated the bench with the test orifice, I put it away. I do use the same carb to basline my tests for that day however. Thats how I caught the variation and figured it had to be from different ambient conditions. In this case I'm benchmarking it against a modified carb to see the changes. I run my tests at 20.3" water, standard for rating 4bbl carbs. Maybe Im just doing something wrong?
willeng,
thanks for the info, and can you answer tony's question about your manometers?
willeng,
thanks for the info, and can you answer tony's question about your manometers?
- rx7carl
- Posts: 11
- Joined: Sat Feb 14, 2004 1:51 pm
by 84-1074663779 » Mon Feb 16, 2004 7:10 pm
rx7carl,
It is an interesting test to carry out using two identical orifice plates, and can be quite revealing.
My understanding is that the carby manufacturers rate the flow of single and two barrel carburettors at 3 inches of water, and four barrel carburettors at 1.5 inches of water. This is a convention, but it is also pretty close to the typical running conditions found on real engines when running flat out.
The single and two barrel carbies are usually found on older or low specific power engines. The four barrels on what (were) considered modern high output engines back in the carby era.
It has been a very long time since I messed with carbies though.
It is an interesting test to carry out using two identical orifice plates, and can be quite revealing.
My understanding is that the carby manufacturers rate the flow of single and two barrel carburettors at 3 inches of water, and four barrel carburettors at 1.5 inches of water. This is a convention, but it is also pretty close to the typical running conditions found on real engines when running flat out.
The single and two barrel carbies are usually found on older or low specific power engines. The four barrels on what (were) considered modern high output engines back in the carby era.
It has been a very long time since I messed with carbies though.
- 84-1074663779
by Jocke » Mon Feb 16, 2004 7:18 pm
In the beginning of my flowbenching, I had setup inlet hole straight above orifice hole, the slightest change in position of testpiece would change my readings a lot... The soulution was th move the orifice as far from valves and orifices as possible, now my values doesn't change more than 1-2% (less if I am really thorough zeroing the manometers) from day to day...
- Jocke
- Posts: 5
- Joined: Fri Sep 17, 2004 3:07 am
by 84-1074663779 » Mon Feb 16, 2004 8:55 pm
I had exactly the same problem as Jocke. Get the test piece and measurement orifice well out of direct line and as far apart as possible. A large volume plenum area immediately after the test hole is excellent as well.
Let the air recover a bit. A higher measurement orifice pressure is also a step in the right direction if you can afford it.
Trying to measure angry air through your measurement orifice is going to lead to problems.
Let the air recover a bit. A higher measurement orifice pressure is also a step in the right direction if you can afford it.
Trying to measure angry air through your measurement orifice is going to lead to problems.
- 84-1074663779
20 posts
• Page 1 of 2 • 1, 2
Return to Flowbench techniques
Who is online
Users browsing this forum: No registered users and 1 guest