FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups 
 ProfileProfile   PreferencesPreferences   Log in to check your private messagesLog in to check your private messages   Log inLog in 
Forum index » Electronix » design
Pressure Transducer Linear algorithm needed
Post new topic   Reply to topic Page 1 of 1 [8 Posts] View previous topic :: View next topic
Author Message
Stanislaw Flatto
electronics forum addict


Joined: 10 Sep 2005
Posts: 74

PostPosted: Wed Jul 19, 2006 12:06 am    Post subject: Re: Pressure Transducer Linear algorithm needed Reply with quote

William Hightower wrote:
Quote:
I have a pressure transducer that I am trying to make its output linear to
drive a digital readout.
The digital readout takes a voltage of 0 to 2 Volts. The problem I am having
is Cumming up with an algorithm to crunch the following numbers to feed the
digital display. Any and all help will be greatly appreciated.

What IS the accuracy of your pressure standard? As I remember from doing
those things this part has to be at least 5-10 times more accurate than
the calibrated transducer. So in my case it was ordered from
manufacturer for placing at precisely measured gravitional force in the
test group building (6-7 digits). This was THE standard from which other
standards (mobile) were compared.
Also the precise zeroing of output at start of calibration and its
stability over time is very important.
Then comes the transducers full scale and its +/- FS error promissed by
manufacturer. Readings <30% FS are considered "gossip".
BTW is your transducer absolute or atmospheric as I havn't seen 0
pressure reading in your numbers.
And last, from experience, "on a plane you can draw a straight line
through any three points if the line is thick enough".
The rest is statistics.

Have fun

Stanislaw
Slack user from Ulladulla
Back to top
Bill Sloman
electronics forum Guru


Joined: 12 May 2005
Posts: 1080

PostPosted: Tue Jul 18, 2006 11:25 pm    Post subject: Re: Pressure Transducer Linear algorithm needed Reply with quote

Tim Wescott wrote:
Quote:
Spehro Pefhany wrote:
On Tue, 18 Jul 2006 12:24:06 GMT, the renowned "William Hightower"
william.hightower@earthlink.net> wrote:


I have a pressure transducer that I am trying to make its output linear to
drive a digital readout.
The digital readout takes a voltage of 0 to 2 Volts. The problem I am having
is Cumming up with an algorithm to crunch the following numbers to feed the
digital display. Any and all help will be greatly appreciated.



Fitting a simple quadratic polynomial to your data yields:

P = 5.621655 * Vin - 0.00047733 * Vin^2 -56.02

...and gives an answer that's within +/-12 psi over the range.

(+/- 0.35% FS)

Your data looks a mite "noisy" to me. Perhaps better measurements
could yield a better fit.

If not better, then perhaps a whole bunch of them. Assuming that the
noise is zero mean (or at least consistent mean) then you'll effectively
average it out when you do a curve fit.

Noise decreases as the square root of the number of observations being
averaged - better measurements are always the place to start.

One of the traps for young players in curve fitting is trying to fit
too many parameters for the data available. Good curve-fitting programs
give error estimates on the parameters allowing for co-variance in the
parameters, and this usually makes it obvious when you are trying to go
too far.

--
Bill Sloman, Nijmegen

--
Bill Sloman, Nijmegen
Back to top
Tim Wescott
electronics forum Guru


Joined: 04 May 2005
Posts: 707

PostPosted: Tue Jul 18, 2006 4:18 pm    Post subject: Re: Pressure Transducer Linear algorithm needed Reply with quote

Spehro Pefhany wrote:
Quote:
On Tue, 18 Jul 2006 12:24:06 GMT, the renowned "William Hightower"
william.hightower@earthlink.net> wrote:


I have a pressure transducer that I am trying to make its output linear to
drive a digital readout.
The digital readout takes a voltage of 0 to 2 Volts. The problem I am having
is Cumming up with an algorithm to crunch the following numbers to feed the
digital display. Any and all help will be greatly appreciated.



Fitting a simple quadratic polynomial to your data yields:

P = 5.621655 * Vin - 0.00047733 * Vin^2 -56.02

...and gives an answer that's within +/-12 psi over the range.

(+/- 0.35% FS)

Your data looks a mite "noisy" to me. Perhaps better measurements
could yield a better fit.

If not better, then perhaps a whole bunch of them. Assuming that the

noise is zero mean (or at least consistent mean) then you'll effectively
average it out when you do a curve fit.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Posting from Google? See http://cfaj.freeshell.org/google/

"Applied Control Theory for Embedded Systems" came out in April.
See details at http://www.wescottdesign.com/actfes/actfes.html
Back to top
Spehro Pefhany
electronics forum Guru


Joined: 01 May 2005
Posts: 2326

PostPosted: Tue Jul 18, 2006 4:17 pm    Post subject: Re: Pressure Transducer Linear algorithm needed Reply with quote

On Tue, 18 Jul 2006 12:24:06 GMT, the renowned "William Hightower"
<william.hightower@earthlink.net> wrote:

Quote:
I have a pressure transducer that I am trying to make its output linear to
drive a digital readout.
The digital readout takes a voltage of 0 to 2 Volts. The problem I am having
is Cumming up with an algorithm to crunch the following numbers to feed the
digital display. Any and all help will be greatly appreciated.


Fitting a simple quadratic polynomial to your data yields:

P = 5.621655 * Vin - 0.00047733 * Vin^2 -56.02

....and gives an answer that's within +/-12 psi over the range.

(+/- 0.35% FS)

Your data looks a mite "noisy" to me. Perhaps better measurements
could yield a better fit.


Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
Back to top
John Fields
electronics forum Guru


Joined: 24 Mar 2005
Posts: 3260

PostPosted: Tue Jul 18, 2006 3:22 pm    Post subject: Re: Pressure Transducer Linear algorithm needed Reply with quote

On Tue, 18 Jul 2006 12:24:06 GMT, "William Hightower"
<william.hightower@earthlink.net> wrote:

Quote:
I have a pressure transducer that I am trying to make its output linear to
drive a digital readout.
The digital readout takes a voltage of 0 to 2 Volts. The problem I am having
is Cumming up with an algorithm to crunch the following numbers to feed the
digital display. Any and all help will be greatly appreciated.

---
View in Courier:


+------+ +------+ +-----+ +-----+ +-----+ +---------+
| XDCR |---| GAIN |---| ADC |---| LUT |---| DAC |---| DISPLAY |
+------+ +------+ +-----+ +-----+ +-----+ +---------+



--
John Fields
Professional Circuit Designer
Back to top
HKJ
electronics forum beginner


Joined: 15 Feb 2005
Posts: 33

PostPosted: Tue Jul 18, 2006 2:52 pm    Post subject: Re: Pressure Transducer Linear algorithm needed Reply with quote

William Hightower wrote:
Quote:
I have a pressure transducer that I am trying to make its output linear to
drive a digital readout.
The digital readout takes a voltage of 0 to 2 Volts. The problem I am having
is Cumming up with an algorithm to crunch the following numbers to feed the
digital display. Any and all help will be greatly appreciated.

Your can do i with a polynom:

MiscEl can calculate the coefficeints:
http://www.miscel.dk/FitTest.png

or a table (MiscEl can also generate a table).


Link to program:
http://www.miscel.dk/MiscEl/miscel.html
Back to top
Mochuelo
electronics forum Guru Wannabe


Joined: 10 May 2005
Posts: 128

PostPosted: Tue Jul 18, 2006 12:57 pm    Post subject: Re: Pressure Transducer Linear algorithm needed Reply with quote

On Tue, 18 Jul 2006 12:24:06 GMT, "William Hightower"
<william.hightower@earthlink.net> wrote:

Quote:
I have a pressure transducer that I am trying to make its output linear to
drive a digital readout.
The digital readout takes a voltage of 0 to 2 Volts. The problem I am having
is Cumming up with an algorithm to crunch the following numbers to feed the
digital display. Any and all help will be greatly appreciated.



PSI Output (m volts)

3400 649
3300 629
[...]
22 15.7
20 15.6


http://80.35.195.131/ng/ng_20060718A00.png
That is quite linear, isn't it?
Back to top
William Hightower
electronics forum beginner


Joined: 06 Jan 2006
Posts: 8

PostPosted: Tue Jul 18, 2006 12:24 pm    Post subject: Pressure Transducer Linear algorithm needed Reply with quote

I have a pressure transducer that I am trying to make its output linear to
drive a digital readout.
The digital readout takes a voltage of 0 to 2 Volts. The problem I am having
is Cumming up with an algorithm to crunch the following numbers to feed the
digital display. Any and all help will be greatly appreciated.



PSI Output (m volts)

3400 649
3300 629
3200 610
3100 591
3000 570
2900 550
2800 532
2700 514
2600 495
2500 476
2400 456
2300 437
2200 417
2100 398
2000 380
1900 360
1800 341
1700 322
1600 302
1500 284
1400 265
1300 246
1200 227
1100 208
1000 190
900 171
800 152
700 134
600 115
500 98
400 79
300 60
200 42
100 27.5
98 27.2
96 26.8
94 26.5
92 26.2
90 25.8
88 25.5
86 25.2
84 24.8
82 24.4
80 24.1
78 23.8
76 23.5
74 23.1
72 22.8
70 22.5
68 22.1
66 21.8
64 21.5
62 21.2
60 20.9
58 20.6
56 20.2
54 19.9
52 19.6
50 19.3
48 19
46 18.7
44 18.4
42 18.2
40 17.8
38 17.5
36 17.2
34 16.9
32 16.7
30 16.4
28 16.3
26 16
24 15.8
22 15.7
20 15.6
Back to top
Google

Back to top
Display posts from previous:   
Post new topic   Reply to topic Page 1 of 1 [8 Posts] View previous topic :: View next topic
The time now is Wed Aug 15, 2018 1:42 pm | All times are GMT
Forum index » Electronix » design
Jump to:  

Similar Topics
Topic Author Forum Replies Last Post
No new posts Problem with output power-linear amplifier tzitzikas design 0 Thu May 20, 2010 3:32 pm
No new posts 12V DC to 1.5V DC needed 3T39 Repair 11 Fri Jul 14, 2006 10:02 pm
No new posts KLH 21 Tuning Cap needed Tim Schwartz Repair 1 Fri Jul 14, 2006 7:42 pm
No new posts 17 V linear voltage regulator Mochuelo components 6 Fri Jul 14, 2006 11:57 am
No new posts Non-linear model for SPICE wombat cad 2 Fri Jul 14, 2006 9:44 am

Copyright © 2004-2005 DeniX Solutions SRL
Other DeniX Solutions sites: Unix/Linux blog |  Unix/Linux documentation |  Unix/Linux forums |  Medicine forum |  Science forum  |  Send and track newsletters


Powered by phpBB © 2001, 2005 phpBB Group