MCUmall EPROM BIOS Chip Burner Forum
MCUmall EPROM BIOS Chip Burner Forum
Home | Profile | Register | Active Topics | Members | Search | FAQ
Username:
Password:
Save Password
Forgot your Password?

 All Forums
 MCUmall Forums
 General Discuss
 why resolution lower than accuracy
 Forum Locked
 Printer Friendly
Author Previous Topic Topic Next Topic  

mozobata

2 Posts

Posted - 07/06/2017 :  02:02:06  Show Profile

Regards,

I've seen a few sensors which its datasheet claims to have a better resolution than the mean accuracy. I understand resolution as the lower identifiable value of the measurement; and, as so, half of it is the lower movement that you can see.

For example, the ds18b20 1-wire temperature sensor:

its 12bits AD resolution are analog to a 0.0625V value, and the typical error is given by:

ds18b20 typical error curve

so, suppose that I need to monitor a liquid that is between 30 and 70#186;C; that fantastic resolution a third than the mean error for that gap.

I tend to think that I could extrapolate the mean curve and, in function of the current point, adjust the resolution.

But, in other example, this is the specification values for the BME280 (t/p/h) temperature sensor:

bme280 temperature characteristics

In this case I have a 0.01#186;C resolution with a 0.5#186;C accuracy (at best); but no error distribution chart.

So.. What it is for!??

Thanks in advance.

Edited by - anniel on 01/08/2019 06:21:55
Reply #1

supervizeur

462 Posts

Posted - 07/09/2017 :  13:55:37  Show Profile
quote:
Originally posted by mozobata



Regards,

I've seen a few sensors which its datasheet claims to have a better resolution than the mean accuracy. I understand resolution as the lower identifiable value of the measurement; and, as so, half of it is the lower movement that you can see.

For example, the ds18b20 1-wire temperature sensor:

its 12bits AD resolution are analog to a 0.0625V value, and the typical error is given by:

ds18b20 typical error curve

so, suppose that I need to monitor a liquid that is between 30 and 70#186;C; that fantastic resolution a third than the mean error for that gap.

I tend to think that I could extrapolate the mean curve and, in function of the current point, adjust the resolution.

But, in other example, this is the specification values for the BME280 (t/p/h) temperature sensor:

bme280 temperature characteristics

In this case I have a 0.01#186;C resolution with a 0.5#186;C accuracy (at best); but no error distribution chart.

So.. What it is for!??

Thanks in advance.




h**p://kb.mccdaq.com/KnowledgebaseArticle50043.aspx

Edited by - anniel on 01/08/2019 06:25:48
Go to Top of Page
Reply #2

anniel

2572 Posts

Posted - 07/16/2017 :  04:33:50  Show Profile
quote:
Originally posted by supervizeur

quote:
Originally posted by mozobata



Regards,

I've seen a few sensors which its datasheet claims to have a better resolution than the mean accuracy. I understand resolution as the lower identifiable value of the measurement; and, as so, half of it is the lower movement that you can see.

For example, the ds18b20 1-wire temperature sensor:

its 12bits AD resolution are analog to a 0.0625V value, and the typical error is given by:

ds18b20 typical error curve

so, suppose that I need to monitor a liquid that is between 30 and 70#186;C; that fantastic resolution a third than the mean error for that gap.

I tend to think that I could extrapolate the mean curve and, in function of the current point, adjust the resolution.

But, in other example, this is the specification values for the BME280 (t/p/h) temperature sensor:

bme280 temperature characteristics

In this case I have a 0.01#186;C resolution with a 0.5#186;C accuracy (at best); but no error distribution chart.

So.. What it is for!??

Thanks in advance.




h**p://kb.mccdaq.com/KnowledgebaseArticle50043.aspx



Good explanation.

Edited by - anniel on 01/08/2019 06:26:44
Go to Top of Page
  Previous Topic Topic Next Topic  
 Forum Locked
 Printer Friendly
Jump To:
MCUmall EPROM BIOS Chip Burner Forum © Copyright 2003 - 2009 Mcumall Electronics Inc. Go To Top Of Page
Generated in 0.13 sec. Snitz Forums 2000