Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

why sensor LM35 temperature sensor return same data in fixed conditions???

Status
Not open for further replies.
stackprogramer

stackprogramer

Full Member level 3
Joined
Jul 22, 2015
Messages
181
Helped
0
Reputation
0
Reaction score
1
Trophy points
1,298
Activity points
2,668
why sensor LM35 temperature sensor doesn't return same data in fixed conditions???

i designed a PCB board ARM.a part of it,is temperature sensor .this sensor LM35 is on a engine car
that when it is vibrated it in temp same before it return data wrong.
vibration is caused data is changed.my question is which sensor is accurate in vibration cases.:bang:
please help me
i want to it has some screw and pins too to i can connect to body engine or ....
broken link removed
**broken link removed**
the result obtained from LM35 is not repeatable in same conditions!!!!1
 
Last edited by a moderator:

It doesn't look like it should be sensitive to vibration, when I look at the data sheet.
It does have a line regulation spec. If you are running the device off 12V from the vehicle, the typical 10mV/V spec would look like about a degree C variation per volt.
There could easily be interference concerns too.

If you are not completely certain that vibration is the cause, I would suggest regulating the LM35 power supply, shielding the sense voltage backto the A/D conversion point, and if you are using a metal body version of the device, that the electrical connection from the metal body to chassis ground that you are using to measure temperature is a solid connection. Perhaps this should be the single point ground for your PCB.
 

ionp said:
It doesn't look like it should be sensitive to vibration, when I look at the data sheet.
It does have a line regulation spec. If you are running the device off 12V from the vehicle, the typical 10mV/V spec would look like about a degree C variation per volt.
There could easily be interference concerns too.

If you are not completely certain that vibration is the cause, I would suggest regulating the LM35 power supply, shielding the sense voltage backto the A/D conversion point, and if you are using a metal body version of the device, that the electrical connection from the metal body to chassis ground that you are using to measure temperature is a solid connection. Perhaps this should be the single point ground for your PCB.
Click to expand...

thanks i examined all connection ,tacking , and...
i think LM35 is weak...
is there powerful sensor that has not this problem.


but LM35 is not accurate in strong vibration ,my means it had not same data in same conditions.it is randomly increased or decreased....
 

stackprogramer said:
thanks i examined all connection ,tacking , and...
i think LM35 is weak...
is there powerful sensor that has not this problem.


but LM35 is not accurate in strong vibration ,my means it had not same data in same conditions.it is randomly increased or decreased....
Click to expand...

The temperature is sensed by a semiconductor junction.
The sensor itself is immune to vibration, as far as I can see.
I think your problem has to be more of a system problem.

This device can source current, but is almost floating if it has to sink current. Please pull the output to ground with 4700 ohms or so.

Please put a grounded shield around the device and see if it is still sensitive to vibration, If it is not, you may have a magnetic or electric field that the device in moving within.

Make sure there is less than 50pF of capacitance from the output to ground, from any cable or bypass connections. The device cannot drive such loads directly and may be on the verge of oscillating, which could be triggered by almost anything.
 
I agree with the comments saying that the vibration sensitivity is unlikely caused by LM35 sensor. But whatever causes the problem in your circuit, it seems obvious that this type of sensor isn't suited for engine measurements. I would consider industry standard resistance thermometers and thermo elements in a first place. Special types with mounting options are available,
 
FvM said:
I agree with the comments saying that the vibration sensitivity is unlikely caused by LM35 sensor. But whatever causes the problem in your circuit, it seems obvious that this type of sensor isn't suited for engine measurements. I would consider industry standard resistance thermometers and thermo elements in a first place. Special types with mounting options are available,
Click to expand...
thanks for reply can you introduce some sensor that is suitable for these works?
 

Sensor form would be selected according to application requirements which you didn't tell yet.

You find many suitable sensors at catalog distributors like Farnell or temperature measurement specialists, e.g. https://www.omega.com/
 

I guess you have not soldered the sensor leads well. It is the loose contacts that are creating havoc.
 

Last edited:

Considering your problems with LM35, I would not suggest the "naked" Pt100 chip. A similar chip is safely molded in the industrial sensor forms. Which one you choose really depends on your application requirements that we don't know. Besides the standard forms that you show, there are hundreds of specialized types. Everything depends on:

- probed medium (solid, liquid, gas)
- size and shape of probed volume
- temperature range
- response time
- special mounting requirements
- environmental conditions
-
 
stackprogramer said:
hi, i want to buy a good sensor, i want to know your opinion,
is this sensor PT100 good for vibration case like engine car?
Click to expand...

I would like to know the cause of the noise or error caused by vibrations in LM35 sensor. If this works happily, others too will work and vice versa.

I also do not understand clearly the title: why sensor LM35 temperature sensor doesn't return same data in fixed conditions...

First suspect is some loose connection. I am lost afterwards...
 
c_mitra said:
I also do not understand clearly the title: why sensor LM35 temperature sensor doesn't return same data in fixed conditions...
Click to expand...
thanks for reply,my means when sensor is worked now ,the temperature result is for example:
t=1 s T=100 C
t=2 s T=120
t=3 s T=130 and so on
but it is not periodic !!it is random...
we want to use it in a feed back circuit,and we have not same data for deciding in in microcontroller ???

- - - Updated - - -

i even i tested LM35 in a environment that temp is fixed like a metal wall ,when metal wall(body) has not no vibration it returns fixed data,
when we vibrate wall with special frequency it results not fixed data!
 

Are the reported temp values statistically random? I mean do you think that it is picking up noise because of vibrations? (the opposite question will be why it is not picking up noise while stationary) Is the sensor physically located at a distance from the measuring circuit? Are all the signals being carried by shielded cables? Is it sensitive to low frequencies (if you mount the LM35 on a pendulum) or high frequencies (fixed to a tuning fork)?

If you have a small speaker, fix the LM35 to the speaker cone with a drop of rubber cement and feed the speaker from the oscillator and see the response vs frequency.
 
hello



it seems you have a LM35D model

LM35, LM35A −55°C to +150°C
LM35C, LM35CA −40°C to +110°C
LM35D 0°C to +100°C

so overrange measure if temp is above 100°C !
maybe your device is a bit detroyed ...
 
My friends thanks very much for your invaluable offers.i will tested all and again send post here

best regards stackprogramer:thumbsup:
 

After examine LM 35 is set ted properly:
i putted sensor lm35 on car door.
now in temperature 35
when my engine is off it is showed 35
when my engine car is on.,it is showed 35.1
which sensor should i use for in both of state it show 35
why it added .1 to temperature:bsdetector::bsdetector:
 

The best test will be on broken ice (0C) and boiling water (100C) and room temp (25 or whatever) and see if the reported values are correct.
 

yes,i tested it for ice now it is good.it is show 0 zero
for boiling water i need some time...

- - - Updated - - -

for vapour water it was showed about 99.5
but my problem is when it is vibrated it has error in temperature
i want to find a sensor temp that is sensitive to vibration??
i even change lm35 sensor 5 times
 

So the device is in good health.

If you are using long wires to connect the device and the sensor is far from the measuring circuit, please twist the wires nicely (2-4 turns /inch) to avoid stray pick up. I do not expect the sensor to find out whether the engine is running or not!!
 
thanks for reply.i tested all type model wiring.

I do not expect the sensor to find out whether the engine is running or not!!
Click to expand...
yes,your words is completely true.
i thought there is a sensor that is not sensitive to vibration
do you offer i test pt100?
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top