Differential Pressure Transmitters

Mark Eatherton

Inexpensive differential pressure transmitter for use with Honeywell VFD pump motor controller. When the transmitters cost more than the controller, you KNOW you've got a problem.

Anyone got any idears??

TIA

ME

Dave_22

Try Wika or Dwyer.... I think they are somewhat affordable. Pressure sensor is one thing, a differential is a bit more pricey!!!! Good luck.

Mike T., Swampeast MO

I'll try to search

How much accuracy is required and what is the ballpark differential?

Mi39ke_2

Omega

Hey McGyver,

Have you tried Omega.com? I honestly went there to search for you, but spent all my free time ogling this little number: http://www.omega.com/pptst/OS-XL.html

It's new to me, anyway!

Have fun,

Michael Ward

tim smith

delta P sensors

I am trying to locate in my mess of paperwork the sensors I found before. I will keep looking and if I find I will let you know. Tim

Mike T., Swampeast MO

Differential Pressure Sensor

doesn't look like it could cost too much and ratings sound like they'd be great for circulator control...

Many more products at:

Advanced Custom Sensors, Inc.

Mark Eatherton

Thanks everyone!!

As usual, you've been extremely helpful.

ME

Blackoakbob

diff press transducers

Hi Mark,
To repay a reply you made to me once.... Try setra at this number 800-257-3872 and of course there's always Rosemount but they are sold by the ounce!
I have used setra for high pressure steam and vaccum they have a large selection and are very dependable.
Best Regards,
Bob

Blackoakbob

diff press transducers

Hi Mark,
To repay a reply you made to me once.... Try setra at this number 800-257-3872 and of course there's always Rosemount but they are sold by the ounce!
I have used setra for high pressure steam and vaccum they have a large selection and are very dependable.
Best Regards,
Bob

Mark Eatherton1

Thanks Bob...

I'll give them a call.

ME

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

jerry scharf_2

roll your own pressure sensor

Mark,

There are a bunch of ways to skin this cat. I'm not sure whether the this moto part are wet capable, but their automotive line pressure sensors are really nice when they can be used. The mpx4250d would do your job nicely and put out a 0-5V analog output of a 0-36 psi difference. It has a 1.5% range error and is temperature compensated from 0-80C. They usually run under $20 in quantity 1 , but digikey doesn't stock this part. Get the unibody package, so you can solder it to a breadboard.

To put this into the Honeywell, I would use a pair of Dallas Semi (Maxim) 1-wire chips. one for the A/D (12 bits would be plenty) and a second for the voltage D/A. Then the computer reads the pressure and adjusts the voltage until the pressure is met. The 1-wire chips run about $2-5 each, and can be bought quantity 1 from maxim. Since you need regulated 5V for the mpx, use a 3rd wire and don't use the parasitic powered chips.

If the moto part won't work, Omega has a $35 wet/wet diff pressure sensor. Since this is mV/V, you would want a 16 bit A/D and it's temperature compensation range is narrower. I think it takes a +10V supply.

I can't imagine doing this without something programmable in the middle, it's much easier to tell a computer what you want to happen than some knobs and pots. I think the 1-wire system is by far the easiest way to hook sensors to the computer.

$30 for the sensor, A/D and D/a seems quite cheap. Add a regulated 5V power supply (maybe 2 if you need +5V and +10V) and the USP 1-wire interface, and you're in business. Once you have this, you can drop $3 temperature sensors just about anywhere you want to and add them to the control package. A cheap trick for sensing a pipe is to put the sensor with the transfer plate contacting the pipe in a piece of split pipe insulation. It's a 30 second install and gives better readings than any other way I've found.

Lonworks is certainly more elegant, but the sensors cost probably 10x or more compared to the 1-wire.

enjoy,
jerry

Mark Eatherton1

Jeeze Jerry....

And here all along, I thought you were just a nice grief counselor. Now I find out you're a rocket scientist in disguise! Thanks for all the information buddy. Think you could de-greektify it for me? One wire? I thought for any communications to take place there had to be a minimum of 2 wires, with one of them being the return or ground? Am I too old school?

The more I learn, the less I realize I knew.

Tell us some more. BTW, I think I'll opt for the store boughten one on this trial 'cause the job is clear over in Grand Junction and that's an 8 hour trip along the beautiful Colorado river for me.

Technology... Ain't it sumpthin!!

ME

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

jerry scharf_2

here goes

I worked for a while at NASA, but on far more pedestrian things than rockets. I met some airflow simulation people who really did live up to the billing.

1-wire is a trademarked name for a simple instrumentation system designed by Dallas Semiconductor, now Maxim 1-wire

jerry scharf_2

here goes

I worked for a while at NASA, but on far more pedestrian things than rockets. I met some airflow simulation people who really did live up to the billing. My background is computers, control systems and large networks, but 25 years of silicon valley has burned me out (at least for now.)

If you want something off the shelf that can do the whole thing, it probably won't be cheap like I build it. You need some control system somewhere to take whatever the pressure transducer puts out, and turns that into the control signal that the VS pump needs to meet your goal. If you want constant delta P, then you need something like a simple PID controller to handle the pump and sensor delays and reseek the goal. You also need to operate the controller, which in a small job ends up being most of the work.

1-wire is a trademarked name for a simple instrumentation system designed by Dallas Semiconductor, now Maxim 1-wire. The reason they call it 1-wire is that many of the parts don't need a separate power line, just 1 signal wire and ground. It turns out that you're usually better off running a power wire as well, but since I end us using cat5 4pr wire for this, the extra wires are no problem. I like it so much because the sensors are so cheap and flexible. This shifts away from the limits of SBCs with fixed built in interfaces and lets me focus on the controls. It's definitely the best roll-your-own stuff out there.

The Motorola pressure sensors and a 1-wire A/D is a perfect example. Automotive pressure sensors are extremely rugged, accurate and cheap, so you can roll all the electronics for a delta P sensor for $30 plus wiring (handy for checking filters in the attic.) There are cheap computer boards for 1-wire in the $50-100 range that have everything I need including ethernet. Of course, my time is free. I expect to put well over $1000 of this stuff in my house, not counting the controls themselves.

You need to know your way around a electronics (no, you don't use a torch for soldering this stuff ) but there's no special electronic design work.

I'll write another thread in a while about the control and monitoring system for the house. I'd love to get people's input on this. It will have a few 1-wire TINI controllers handling polling and specific tasks (heating, A/C, ventilation, monitoring) and a linux box that manages all of those, does the web interface and the like. At last count, there are approaching 200 sensors for the house. good thing they're cheap.

Some maintenance person is going to scratch their head when someone after me owns the house.

jerry

Mark Eatherton1

Thanks Jerry! (nm)

ME

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

jerry scharf_2

I wouldn't choose this part

Mike,

This transducer is only going to put out about 3mV at 5psi with a 5V input. That's within the error range for the device, and you will absolutely have to use a 16 bit A/D. Also, it's not temperature compensated, so the readings will change as the temperature changes, and not in a necessarily simple way.

It's cool that it's designed as a pipe fitting.rahterh than the hose fittings that the automotive world loves.

jerry

tim smith_2

diff pressure differential transmitters

Mark, this is the company I talked to and could not find info on. I think their product will work. Tim.

Mike T., Swampeast MO

Thanks for reporting Jerry

My knowledge of the application of such devices is little more than nill, but at least I mainly understand why you say it's not suitable. The connections really did look nice though <;)

I presume that "temperature compensated" is yet another nasty little problem with "how much does of gallon of water weigh".

jerry scharf_2

what the temp compensation is about

Mike,

There are several ways to measure pressure using semiconductors. One is to use a piezoelectric crystal, which puts out a voltage when it's deformed. the second is to produce a strin guage using silicon fabrication techniques to build a really small beam and measure it's deflection. There are other less common ways.

In each of these, the semiconductors are stable with temperature. As the temperature rises, the output of the device increases. In both of the instruments above, the output curve does not to keep the same slope and the zero offset changes. It's much nicer to have the people who make the sensors handle all the compensation for their given process and give you an overall error spec across the range.

hope that's helpful,
jerry