Temperature sensor opinions!

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
User avatar
Lenp
Posts: 1316
Joined: Thu Oct 26, 2006 7:11 pm
Location: Maryland
Contact:

Temperature sensor opinions!

Post by Lenp » Tue Oct 03, 2017 6:24 pm

I am into a design for a test jig to consecutively measure the temperature at 4 different locations on a common display. The critical temperature for all points is 40C. A micro driven relay, or ssr, switching system to scan and select the points is being considered.

With all the error considerations with thermocouples when switching and dissimilar junctions, I don't think that is the way to go, plus the TC switches available are mostly manual switches and too costly

Thermistors are low cost but inherently a bit innacurate. The platinum sensors (PT-100, PT-1000) have attractive enhanced specifications, but I noticed that with seveal different sensors connected to the same display there still are some differences. I would like to 'trim' each sensor for optimum accuracy. In view of the fact they are 'resistance devices' I am considering a trimmer resistor in each sensor circuit. Linearity is not a big issue since the target temperature is that 40C, single point reference with little concern on either side of that value..

If the trimmer shifts the sensor loop resistance, out of proper adjustment, I can easily compensate for that in the display with a temperature offset value, that is agreable to all the sensors.

Before I get too far into all this, any opinions to wake up this forum would be welcomed!
Len

“To invent, you need a good imagination and a big pile of junk.” (T. Edison)
"I must be on the way to success since I already have the junk". (Me)

User avatar
Externet
Posts: 1778
Joined: Fri Feb 01, 2002 1:01 am
Location: Mideast USA
Contact:

Re: Temperature sensor opinions!

Post by Externet » Wed Oct 04, 2017 7:43 am

Hello Len.

Trying to contribute with my grain of sand, I always encounter this doubt when people addresses thermostats...

Your "The critical temperature for all points is 40C" ---> means that is the maximum or minimum you want to achieve ? temperature should be controlled from above or under 40C ?

Give these a look ----> https://getconnected.honeywell.com/en/evohome
----> https://www.carrierenterprise.com/carri ... cc4kit01-b

(Search for "four zone thermostat") Their sensors may be decent or trimmable...

Miguel
- Abolish the deciBel ! -

User avatar
haklesup
Posts: 2829
Joined: Thu Aug 01, 2002 12:01 am
Location: San Jose CA
Contact:

Re: Temperature sensor opinions!

Post by haklesup » Wed Oct 04, 2017 1:09 pm

Have you considered a thermal IR camera, I bet you can rent one and there are some reasonably priced options to own too

http://flir.com/flirone/ at $400 the FLIR One Pro is more than accurate enough to spot measure and monitor this object in real time. The cost of the setup and labor you described exceeds $400 and with a thermal camera you get loads more data that's way easier to understand.

User avatar
Lenp
Posts: 1316
Joined: Thu Oct 26, 2006 7:11 pm
Location: Maryland
Contact:

Re: Temperature sensor opinions!

Post by Lenp » Wed Oct 04, 2017 2:08 pm

Miguel,
The temperature that is of interest is 40°C. Linear tracking above or below that is unimportant.
I just need to know when the probes are at the desired 40°C +/- 0.5° temperature.

Hackelsup
IR is out of the question because of space and cost. A simple display will be more than adequate.
We have an IR gun for a quick test and it is always low by a few degrees since it reads the exterior temperature and not the interior where a probe will be placed.

The question presented really was about trimming the PT-xx sensors for optimum accuracy at 40°C

Thanks all
Len

“To invent, you need a good imagination and a big pile of junk.” (T. Edison)
"I must be on the way to success since I already have the junk". (Me)

User avatar
haklesup
Posts: 2829
Joined: Thu Aug 01, 2002 12:01 am
Location: San Jose CA
Contact:

Re: Temperature sensor opinions!

Post by haklesup » Wed Oct 04, 2017 2:18 pm

40C is pretty close to body temp, anything optimized for body temp reading would be a good place to start. I don't have a specific recommendation.

I agree a IR spot meter often is inaccurate especially on shiny surfaces and its difficult to know exactly where the measurement spot is sometimes, I think the camera would do a better job though I understand your spec.

User avatar
CeaSaR
Posts: 1621
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Temperature sensor opinions!

Post by CeaSaR » Sat Oct 07, 2017 7:38 am

I think the big question here is:
Is this an analog design or a digital (microcontroller) design?
This would greatly affect how you would deal with the issue at hand.

CeaSaR
Hey, what do I know?

User avatar
Lenp
Posts: 1316
Joined: Thu Oct 26, 2006 7:11 pm
Location: Maryland
Contact:

Re: Temperature sensor opinions!

Post by Lenp » Sat Oct 07, 2017 1:21 pm

Hi CeaSar
Guess I didn't make that too clear. The sensor is expected to be either a PT-100 or PT-1000 platinum RTD and the display is a rather standard process control digital display that has many features including choices for different sensors.

Thermocouples can involve numerous errors with inadvertent cold junctions, plus the low output can be problematical in the switching side of the design.
For that reason I am leaning towards the RTD. The charts show that at 40°C, the PT100 is 114.54 ohms and the PT-1000 is 1155.5 ohms. The low on resistance of either a conventional or SSR relay should introduce tolerable, or at least correctable, errors.

The present test jig design uses 4 PT-100's and 4 separate indicators. Unfortunately there seems to be some drift between the different displays as time passes. I am suspicious that it is the display units drifting since the PT type sensors have a pretty tight tolerance. So, if only one display were used and the sensors were switched, it is felt that the display errors would be reduced, or at least more easily compensated....maybe?
Thanks!
Len

“To invent, you need a good imagination and a big pile of junk.” (T. Edison)
"I must be on the way to success since I already have the junk". (Me)

User avatar
CeaSaR
Posts: 1621
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Temperature sensor opinions!

Post by CeaSaR » Sun Oct 08, 2017 5:28 am

Okay. Wasn't sure if you were doing from scratch via a MicroController (software compensation) all hardware.

All hardware based. ;)
(Preaching to the choir, I know)
I would think that a solid state relay would be the best option to minimize offset within the switching since there are no mechanical parts and it's all on the same die. And then probably a high quality trimmer variable resistor near the SSR, so as to keep the ambient temperature fluctuations the same for all the trimmers, thereby minimizing / coupling any thermal drift. Of course you would have to use a calibrated setup to check the output of the probe vs. what the display says, and adjust accordingly.

I know it probably doesn't help much, but it's what I can give at the moment.

CeaSaR
Hey, what do I know?

User avatar
Lenp
Posts: 1316
Joined: Thu Oct 26, 2006 7:11 pm
Location: Maryland
Contact:

Re: Temperature sensor opinions!

Post by Lenp » Sun Oct 08, 2017 7:16 am

Remember even old choirs learn new songs!

Yes, that's where I have been headed. I may even put the SSRS and trimmers under a styrofoam wrapper for better stability but not go all the way to a crystal oven!

About accurate temperature measurements....
In an old British cookbook, a recipe's instructions, said to "cook the pudding until you could put your finger in and count to three, but not more"....
Len

“To invent, you need a good imagination and a big pile of junk.” (T. Edison)
"I must be on the way to success since I already have the junk". (Me)

User avatar
CeaSaR
Posts: 1621
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Temperature sensor opinions!

Post by CeaSaR » Sun Oct 08, 2017 8:22 am

You have that book too??!! ;)

That was along my line of thinking, keep what you can as constant as possible, as even as possible, as minimized as possible, and let the probes do their job.
Hey, what do I know?

User avatar
Lenp
Posts: 1316
Joined: Thu Oct 26, 2006 7:11 pm
Location: Maryland
Contact:

Re: Temperature sensor opinions!

Post by Lenp » Sun Oct 08, 2017 9:40 am

I think I'll hook it up with a rotary switch to see how close the sensors are, and the effect of trimming, before I go too far.
One error at a time!
Len

“To invent, you need a good imagination and a big pile of junk.” (T. Edison)
"I must be on the way to success since I already have the junk". (Me)

User avatar
Janitor Tzap
Posts: 1297
Joined: Sat Aug 12, 2006 4:17 pm
Contact:

Re: Temperature sensor opinions!

Post by Janitor Tzap » Wed Oct 25, 2017 1:09 pm

Hey, Lenp.

Are you still working on a temperature sensor circuit?

I found this one: http://www.goldmine-elec-products.com/p ... ber=G21366
It looks like it works with the Arduino.


Signed: Janitor Tzap

User avatar
Lenp
Posts: 1316
Joined: Thu Oct 26, 2006 7:11 pm
Location: Maryland
Contact:

Re: Temperature sensor opinions!

Post by Lenp » Wed Oct 25, 2017 3:43 pm

Thanks for the interest, but I am not working on a temperature controller design. I am working on using a commercial controller but only as a display, switching multiple PT-100 sensors onto the same control. I am using a rotary switch for testing but the end result will be a micro controlled switcher with several features. Since the PT-100 sensors are low resistance units I will use resistance wire for compensation so all the sensors are really close at one critical temperature. (40C)
A work that is still in progress!
Len

“To invent, you need a good imagination and a big pile of junk.” (T. Edison)
"I must be on the way to success since I already have the junk". (Me)

User avatar
Janitor Tzap
Posts: 1297
Joined: Sat Aug 12, 2006 4:17 pm
Contact:

Re: Temperature sensor opinions!

Post by Janitor Tzap » Thu Oct 26, 2017 3:49 pm

Oops, Sorry......

I forgot you already had the PT-100 sensors. :lol:

I think you'll be running into issues with the rotary switch contacts.
If they are dirty, or not making good contact with the wipers.
Your going to get inconsistent readings from the sensors.

What about using a HCF4066B Quad Bilateral Switch?
https://www.bucek.name/pdf/4066.pdf


Signed: Janitor Tzap

User avatar
Lenp
Posts: 1316
Joined: Thu Oct 26, 2006 7:11 pm
Location: Maryland
Contact:

Re: Temperature sensor opinions!

Post by Lenp » Thu Oct 26, 2017 4:47 pm

Yes, that's what will be used in the micro controller version.. The switch was cleaned and lubed and has been pretty good so far. The solid state switches are low resistance when on, but not zero. Since the PT-100 is a low resistance sensor, this will cause some error. I plan to use a short piece of nichrome wire to calibrate each of the 7 sensors to read alike, then I can enter a correction into the controller to make it read the value needed for all of the sensors.
Thanks for the feedback!
Len

“To invent, you need a good imagination and a big pile of junk.” (T. Edison)
"I must be on the way to success since I already have the junk". (Me)

Post Reply

Who is online

Users browsing this forum: No registered users and 2 guests