Homebuilt Rov Forum

Exploring the Hobby of Building your own ROV - Imagine, Create, Inspire.
It is currently Feb 23rd, 2018, 10:14 pm

All times are UTC - 5 hours [ DST ]




Post new topic Reply to topic  [ 5 posts ] 
Author Message
PostPosted: Mar 7th, 2016, 2:52 pm 
Offline

Joined: Jan 31st, 2015, 4:12 pm
Posts: 18
The humble DS18B20 is a go-to for lots of people because they are cheap and fairly robust. But we needed to improve on the factory accuracy of ±0.5°C, and I think I finally have a process that delivers better than ±0.1°C from them:

http://edwardmallon.wordpress.com/2016/03/05/ds18b20-calibration-we-finally-nailed-it/

The water bath is just passively cooling, so you could do the process in a simple bucket. And if you went with the precision thermistor as your reference you could bring the whole setup in under $100.

Hope that info helps someone with their project.


Top
 Profile  
 
PostPosted: Mar 8th, 2016, 6:59 am 
Offline
User avatar

Joined: Oct 1st, 2013, 7:18 am
Posts: 77
@EKMallon - thanks for posting your results. It is great to see real research and results which are meaningful and useful.

Your research prompted me to look through the DS18b20 specification again and was surprised to see the error graph showing +- 3s.

If I am reading it right, outside this error lurks the > +-3s devices - presumably those you threw away. Your tests highlight that these devices should be tested before relying on them being OK out of the box! I will be doing this when I get to use them, thanks. :)


Top
 Profile  
 
PostPosted: Mar 8th, 2016, 7:20 pm 
Offline

Joined: Jan 31st, 2015, 4:12 pm
Posts: 18
I often get my hands on a sensor that behaves fine from 20-40C, but wont read below 10C. So I figure that all the cheap ones on eBay were probably production run rejects to start with. I've seen so many rumors about these out of spec sensors resurfacing in the grey market.


Top
 Profile  
 
PostPosted: Mar 9th, 2016, 4:51 am 
Offline
User avatar

Joined: Oct 1st, 2013, 7:18 am
Posts: 77
There have been other devices that I have used that were 'out of spec' - they worked for the most part but not 100%, and it was usually the part that didn't work that I needed.
If a part is low cost in the "too good to be true arena" then it probably is too good to be true. A few years ago there was an article saying that a significant proportion of military electronic parts were out of spec rejects, bought by some well meaning buyer who was trying to save tax payers' money. Frightening - especially as some of these parts could find their way into medical or aircraft applications.
I recently bought a smart-watch = £7, inc p&p - because I wanted the case, strap and some of the other internal parts. But am under no illusions that it will function as described.


Top
 Profile  
 
PostPosted: Mar 11th, 2016, 2:43 pm 
Offline

Joined: Jan 31st, 2015, 4:12 pm
Posts: 18
I usually use the cheap stuff because I need to noodle around for a few builds before I really get the physical parts in the right place. Trustworthy vendors sell waterproofed DS18Bs for $10 each, while they are $1.50 on eBay. Order of magnitude ratios like that mean I usually work out my prototypes with the dodgy parts, and move up to known good parts later.

Once and a while I get pleasantly surprised and find that the cheap parts match the performance of the ones from Digikey, etc. So far that's what happened with the DS3231 RTCs:

https://edwardmallon.wordpress.com/2014/05/21/using-a-cheap-3-ds3231-rtc-at24c32-eeprom-from-ebay/.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 5 posts ] 

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group