How do you measure zero pressure simply???

A diaphragm is indeed trivial in concept. Actually building one is yet another matter. Using one is a third. Gravity is a bitch. So is a droplet of water that weighs more than the force of the pressure difference.

Off topic alert:

I've been building heat pipes. If you take a copper pipe and fill it half full of freon and half full of vacuum, it makes a great heat pipe.

If you do the math, it's easy to conclude that you can do the same with water. But they don't work. Then you realize that the vapor pressure of water at room temperature is less than the weight of a few millimeters of water. So the surface boils, but any heat applied more than a few mm below that can't get out past the weight of the water column above it, so all you're left with is convection in the water.

The concept of the heat pipe is the same, but the physical realization doesn't scale well.

If you're gonna make a water heat pipe, you need a wick structure and no water column suppressing the boiling. If anybody knows how to make a useful wick structure from stuff you find in the average kitchen, I'm interested.

Just like a diaphragm. The diaphragm concept itself is easy. The structure to support it is the hard part.

My interpretation of this thread is that most people don't realize how little pressure is represented by a mm of water.

That's why, for a DIY sensor, I skipped directly to the indirect measurement of heat flow. NO moving parts except the air...and even that doesn't move at zero pressure differential. It becomes an electronics problem, and the purpose of this thread was to eliminate most of the electronics.

. o c ) n o

CC is dirt-simple. Are you reading the differential resistance without using any active circuitry? Does that result not go to some active control or monitoring device? I don't understand what we're interfacing to.

You could simply put the two in series, constant voltage, in-line. That way the up-wind unit heats the down-wind thermistor. But CV =3D positive feedback with an NTC.

Right, but you haven't told us what this is for, what it connects to, what power is available, etc. We can only guess.

It sounded like you wanted battery-powered, long-life, sensitive, analog output, connected to a uC. So, I suggested something to minimize power. If you just want a visual indication, something like an ultra-flimsy colored mylar flap in the air-path might do.

25mW ain't bad for the purpose, but yes, that'd draw down a battery.

You might improve that thermally by adding heat "sinks" to the thermistors, improving their coupling to the air. That'd speed the time constant too.

Cheers, James Arthur

You tell me. Which parts of an auto care about a mm of water pressure?

The first two lines of this page say what I'm wanting to do.

The HRV has two opposing blowers. The objective is net zero pressure differential in the vicinity of the HRV.

That's a true statement. But the devil is in the details. As I refine my ability to measure small pressure differences, I'm finding that you can have pressure differentials throughout the house even with all the interior doors open.

I don't have any upper floors.

Holler if you think my math is wrong... Using round numbers to make the math easy. You can scale for more appropriate numbers.

If you have 1400 square feet and 8 foot ceilings, you have 11,200 cubic feet of volume. If you want 0.3 air changes per hour, you have to heat 3360 cubic feet/hour of incoming air == 56 CFM.

If you have a 30F inside/outside temperature differential, you need

1.08 X CFM x dT = 56 x 30 x 1.08 = 1814 BTU/Hr. That's 1,306,368 BTU/Month blowing out the cracks. If gas costs $1/therm, that's $13/month. That represents about a third of my gas bill.

It's desirable to seal the place and get control that flow. A heat exchanger can recover a significant percentage of the lost BTU's. But if it's not pressure neutral, it also forces air thru the cracks.

I'm toying with the idea of dynamically adjusting the fan speeds to compensate for infiltration due to wind. I'm not optimistic, but worth a look.

Once you get it set up, you don't have to mess with it...but I'm still in the messing with it stage.

If you buy a commercial HRV and have it installed and pay for the electricity to run it, you can argue that it won't ever pay back. If you're building it out of coroplast and duct tape, the payback can be relatively quick.

And you can have a lot of fun learning how to weld polypropylene.

First prototype of the Hillbilly HRV:

Preliminary results suggest that you can get significant improvement in indoor air quality while recovering half the BTU's for almost no investment.

.me...

. o c ) n e

So you don't half-fill your heatpipe with water, but just add enough water to keep the wicking surface wet.

If the heat-pipe has been properly evacuated, the milligrams of water that you need to evaporate and condense to move the heat around travel very quickly in the gas phase. Capillary action is rather slower in pulling it back to the warm end of the heat pipe, which is why you've got grams of water keeping the wick wet, but you don't "half-fill" the heat pipe because that blocks the space through which the vapour has to flow (rapidly).

You don't find copper gauze in the average kitchen, but people do seem to stock it

CU5152

Support and stretch ...

-- Bill Sloman, Nijmegen

. o c ) n o

Active circuitry can be pretty simple.

That something can be a simple as resistor in series with the thermistor. You do need to limit the power dissipated in NTC thermistors to less than about a milliwatt - too much self-heating and you start to form "hot channels" inside the device, which wouldn't be too bad if they were stable, but I've seen an NTC thermistor resistance measurement fluctuate because the multi-meter doing the job was dissipating too much power in the device - the fluctuations stopped when we changed range to one that dissipated less power in the part, and they had been big enough to have been detectable on the new range.

You can also prevent thermal runaway by driving the thermistor with a constant current source, but the added extra circuit complexity doesn't usually buy you enough extra sensitivity to be worth the trouble.

-- Bill Sloman, Nijmegen

se.

to

or

tic

ce)

can

or

to

It would help if we knew what we were driving. If it's a uC, then what's the big deal with a simple interface circuit? If not, your idea of a SPDT diaphragm-operated switch is as simple as it's going to get.

CC was so the thermistors could still be used in bridge. I s'pose a single current-limiting resistor split in two pieces could do as well.

Talking about simplicity, it might even be advantageous to capture a tiny bit of that thermal-runaway positive feedback, increasing sensitivity.

-- Cheers, James Arthur

[snip] [snip]

I'm with dagmargoodboat. The diaphragm approach is simplicity in itself. Using it to drive a switch... zero power. But you could get fancy and move the core of a differential transformer, or even do differential capacitance fairly easy. ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

use.

e to

tor

stic

n

ice)

can

is

or ~

run it

nce

n

ok

.

Hi Mike, Why the 'bleep' do you need so much resolution. If I'm doing my numbers correctly.

1 atm. =3D 32 feet of H20 ~ 400 inches. So 0.001 inches is about 2.5x10**-6 atm. That's equivalent to a temperature difference of about 1 milli-Kelvin (ideal gass at 300K)

I should also put that into a height difference... What's the difference in pressure from the top of your head to the floor? (I'll work that out in a minute.)

George H.

[snip] [snip]

Sno-o-o-o-ort ;-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

=A0 =A0 ...Jim Thompson

=A0 =A0| =A0 =A0mens =A0 =A0 |

=A0 | =A0 =A0 et =A0 =A0 =A0|

=A0|

=A0 =A0 =A0 |

I got that a cubic meter of air has a mass of about 1kg.

(which is a nice number)

George H.

So design it so the plane of the diaphragm is vertical and gravity isn't a factor. The weight of a drop of water or heat pipes has as much relevance here as the pressure of sunlight, cosmic ray impacts, or the phase of the moon. Art

Sno-o-o-ort :-)

Take a piece of tire inner tube and mount it in a needlework hoop :-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

That way of using a diaphragm wasn't my idea.

A typical thermistor in still air has a thermal resistance of 1K/mW. The resistance goes down about 4% per degree Kelvin, so you'd need to dissipate a lot of power in the thermistor to see anything approaching run-away or even a perceptible increase in sensitivity.

Dissipating even 1mW in the thermistor gets you close to the region where hot-channel formation in the thermistor can introduce a rather different - and less useful - form of instability. A constant voltage drive (from say a transistor emitter) would mean that a 1K rise in temperature would lead to a 50uW increase on that 1mW, or 0.05K of extra self-heating. Not exactly significant, and that running at an already imprudently high power dissipation.

-- Bill Sloman, Nijmegen

Until you try to make it work.

I tried to get the Melbourne University Chemistry Department machine shop to make me one back in the 1960's, when I was doing my Ph.D. thesis, but they'd already made me bellows gauge in which the position of the bellows was sensed by a linear variable differential transformer (which worked), so it wasn't going to happen.

By then you could find diaphragm gauges with capacitative sensing in the literature - and buy them off the shelf if you had enough money. The MKS Baratron has been around since then

though I imagine that the electronics have changed a bit in the last forty-odd years.

The capacitances involved are on the low side and it pays to excite the bridge circuit involved with relatively high frequencies - 455kHz came up frequently presumably due to its popularity as the IF in AM radios.

-- Bill Sloman, Nijmegen

Should be pretty simple. Aviation rule of thumb is 1" Hg per 1000' altitude change.

For a 6-footer that's 6 milli-inches Hg from nose to toes.

In water (13.6 x Hg) that's 81.6 milli-inches

In psi that's 3 milli-psi

Your turn.

Jim

I think I got something like one part in 10**4. A cubic meter of air is about

1 kg (a bit more). So F=3Dmg is 10 newtons over a square meter is 10 Pa. (1 atm =3D 10**5 Pa)

(I'm slowly converting to MKS units)

George H.

=A0(I'll

About 1.3 kg/m^3 assuming 75% N2 and 25 % O2.

I started with Rydberg's constant, 22.4 L/mole.

Totally inappropriate application of f=3Dmg. Open field air pressure is derived from the weight of the air column above it. Just like = hydrostatic pressure, with the added fun that air is compressible. Wind pressure is even more fun.

Wrong name

Avogadro is the name that came to my mind, but Avogadro's number is

6.02214129(27)=D71023 mol-1

and the number you are quoting is actually the volume of a mole of ideal gas at a "standard" temperature and pressure

You have opted for zero degreee Celcius as your standard temperature. Europeans mostly opt for 20C and Americans for 25C.

-- Bill Sloman, Nijmegen