I need a set of equations or source code (VB) to calculate the
humidity of air as follows:

1. Input Temperature (dry bulb), pressure and relative humidity and
get out specific humidity in grains water/pound dry air.

2. Input Dry bulb temperature, wet bulb temperature and pressure and
get out relative humidity %.

3. Other related calculations if available.

---

Clarification of Question by yanos-ga on 31 Jan 2004 06:35 PST

Need it to be accurate between -30 C and +40 C.

---

Request for Question Clarification by pafalafa-ga on 31 Jan 2004 07:15 PST

Hello yanos-ga,

Are you familiar with Humicalc?  You can see this sophisticated and
very flexible program at:

http://www.thunderscientific.com/humicalcpg/humicalc.html#

and can try out the web version.  It may very well produce the outputs
you need (though in a quick run-through, I saw inputs expressed in
grains, but not outputs).  There is also a downloadable Help file at
the site that gives a few of the formulas used, but mostly references
sources for the formulas.

If you're lookng for a way to calculate variable, this program may do
the trick.  If you need the actual formulas, this is a good pointer,
but it isn't quite there yet.

Let me know if this comes close to meeting your needs.  


Thanks.

---

Clarification of Question by yanos-ga on 31 Jan 2004 08:45 PST

I did find that application during my search but I need source code in
order to embed it in one of my own applications.  The one that I could
not find was how to get specific humidity from RH, temperature and
pressure.  I could not find a place where the equation and units are
clearly stated for this particular calculation and I just need it
listed so I can copy it into my code and have it running by Monday.

---

Clarification of Question by yanos-ga on 31 Jan 2004 16:53 PST

Is anyone able to actually answer this question for me without
pointing me to links that don't work or making me do hours of research
on my own?  I am running a bit short on time now.

The required forumula and physical constants are clearly set out in
NASA Technical Note TN D-8401 "Equations for the determination of
humidity from Dewpoint and Psychrometric Data"
http://www.dfrc.nasa.gov/DTRS/1977/PDF/H-937.pdf

It would be difficult to type these equations as an answer without the
risk of transcription errors or confusion due to reformatting as plain
text, so I suggest you get them from that source.

Additionally, you will have to convert from "pounds per cubic foot of
air" to "grains per pound of air" (although the necessary constants
are there for you).

Your second formula is directly present in the referenced work, and
you will have to back-substitute the formula for RH into the presented
formula for absolute humidity, in order to obtain it from your
starting points (dry bulb temp, pressure and relative humidity).

Note also that different constants are needed depending on whether the
wet bulb is at ice temperatures or water temperatures. These constants
are clearly presented in the appendix of the referenced article.

For some reason the link freezes my browser.  I will continue to try
but pounds per cubic foot is a volumetric measure and grains per pound
of air so I still need density at the specific temperature and
pressure to get my required number.

Any ideas on why the link doesn't work for me?

Hi yanos,

I had a look at the link myself that eiffel found, and it's certainly
jam-packed with formula, but I'm not sure it has everything you need
for specific humidity calcs.  Hopefully, Im mistaken, and it's all
there.

Try "saving" the link to your PC first and then opening it with Adobe
(rather than opening it directly).  That might help you avoid the
freeze-up.

pafalafa-ga

Thanks for the input and I was able to access the PDF file using the
method suggested.  The methods given, however, do not include finding
Specific Humidity from Relative Humidity, temperature and pressure.  I
am beginning to think that perhaps it is more complicated than I
thought.

The initial link http://www.thunderscientific.com/humicalcpg/humicalc.html#
did provide the functionality that I am looking for but the equations
used in that do not appear to be in the reference document
http://www.dfrc.nasa.gov/DTRS/1977/PDF/H-937.pdf.  Perhaps I just do
not have the background to translate the equations but then that is
why I offered to pay someone to do it for me.  (I am too lazy or too
stupid I guess).

p.s. I have emailed humicalc and one other but I don't think I will
get an answer back before I need it.  I also don't know if they will
sell the DLL or source code for a reasonable price.

I'm glad the pdf trick worked out.  Wish I could help you in a more
timely fashion, but I'm afraid it would take me too long to slog
through the materials to even figure out if an answer were possible.

If you want researchers to continue looking out for possibilities for
you, you should make that clear in another clarification...otherwise,
they might all suppose it's too late to answer your question.

Best of luck.

pafalafa-ga

Still looking for an answer on this one.

Yanos,

If you look at the equations on page 22 of the NASA report, you will
see that the last one starts with U(T, T', p) and calculates the
relative humidity from the dry bulb and wet bulb temperatures plus the
air pressure. That's your second equation taken care of (just multiply
by 100 to get it as a percentage).

It's the first one that's difficult, because you want to get the
specific humidity starting from RH, dry bulb temperature and pressure.
Are you sure you can't start from dry bulb and wet bulb temperatures
plus the air pressure, just as you did for your relative humidity
calculation? If so, the fifth equation on page 22, starting with H(T,
T', p), is exactly what you need, although you need to take it further
to get it in grains per pound as you require.

If you are stuck with RH, dry bulb temperature and pressure as your
inputs, then you can still use the fifth equation, but first you have
to calculate the wet bulb temperature from your inputs. To do this,
you can manipulate the final equation to get T' on the left-hand-side.
This gives you T' (wet bulb temperature) calculated from dry bulb
temperature, relative humidity and pressure. You then have the inputs
you need to use the fifth equation.

Sorry, I don't trust myself to manipulate that final equation myself
and get the right answer (it's too many years since I've used all the
rules about manipulating powers and logarithms on equations as complex
as this one), but I reckon that's the best approach.

Finally, I don't know what level of accuracy you need, but if you're
looking for an "easy way out" you could consider using the online
calculators to produce a large table of results, and use table lookup
instead of formulae in your computer program.

I have found answers for you, but I'm not usually a VB programmer, you
need an answer right away, and I'm going out the door shortly. So
instead of taking your $100 I'll just share my sources -- which are
very straightforward code for exactly what you want, which happens to
be in C and Javascript at the moment. But all three languages do
ordinary algebra like this the same way -- this code is very close to
VB already.

* * *
The following page offers simple C source code for step 1. It might
look intimidating but it's just straightforward math code, and would
look about the same in any language. You should have no trouble
rephrasing it in Visual BASIC:

Hot List for PTM Example Applications, U of Kentucky
http://www.bae.uky.edu/~gates/psych/PTM_apps.htm

And the specific C program for this task:

http://www.bae.uky.edu/~gates/psych/PTM/dbrh2al.c

This program also fulfills part 3 of your question by outputting
several other interesting factors.

* * *

The following page offers straightforward Javascript code for step 2.
This will be even easier to translate. Just go to the page and hit
"view source:"

Wet and Dry Bulb Calculator
http://members.nuvox.net/~on.jwclymer/wet.html

You should have no trouble completing the project in a hurry with this
information, but if you really want me to translate it to VB for you
when I'm back at my desk Monday morning, by all means let me know and
I'd be glad to collect the $100 after all.

This is excellent thanks.  I will get someone who knows C to compile
it into a DLL for me and I am off to the races.  Thanks very much.