# Problem with testing of temperature sensor v1.1

Posted by davidgoodenough
 Problem with testing of temperature sensor v1.1 August 25, 2008 11:28AM Registered: 16 years ago Posts: 123
We are having a little trouble testing the temperature sensor v1.1

The sensor is showing a raw value of about 980 for room temperature, which seems to agree with the value shown on the monitor in the picture to the right of the Try it out section just below the code. The values shown in the monitor convert this to a value of about 20C.

But the converted value we get is 0C/32F, which is the overflow value.

We are using the RRRF supplied thermistor, and when measured with an ohm meter it has a resistance of about 100K, so it looks like we have have the right value.

Looking at the code I am not sure I understand what it is doing. The first column is marked adc, but it contains values between 0 and 300 which sounds like a temperature range. The other column goes up to from 1 to 680. Neither of these values go up to 980.

So was the picture generated with a different thermistor?

Anyone got any clues?

David
 Re: Problem with testing of temperature sensor v1.1 August 25, 2008 03:40PM Registered: 16 years ago Posts: 361
Have you read [reprap.org] (assuming you're using arduino electronics)?

Also, [hydraraptor.blogspot.com] might be useful, though I haven't peered at it closely enough to tell
 Re: Problem with testing of temperature sensor v1.1 August 25, 2008 06:06PM Registered: 16 years ago Posts: 123
My point was that I am using the RRRF thermistor, so the default table should be right, and the raw value agrees with the value shown on the monitor image. But I am not getting the right value, and reading the code I do not see how it would get the right value. The picture and the code are not consistant. Part of what I am trying to find out is whether they are supposed to be consistant.

David
 Re: Problem with testing of temperature sensor v1.1 August 29, 2008 08:35AM
The code snippet linked for testing uses the wrong thermistor table.

You need to cut and paste the one for the Epcos device if you are tesing one of the current RRRF kits ... there is a link on the make page to the RepRap thermistor page which has several tables.

Peter
 Re: Problem with testing of temperature sensor v1.1 August 29, 2008 09:41AM Registered: 16 years ago Posts: 123
Great, that fixed it. Perhaps a note to that effect on the temperature sensor page (i.e. get the Epcos data from the thermistor page for the current RRRF kit) would be nice.

David
 Re: Problem with testing of temperature sensor v1.1 August 29, 2008 10:10AM Admin Registered: 17 years ago Posts: 1,487
hey guys,

sorry about that. as pjr pointed out, the table in the example was wrong. i fixed the code in the wiki.

~Zach
 Re: Problem with testing of temperature sensor v1.1 January 28, 2009 03:19PM Registered: 15 years ago Posts: 2
Hi,

I had some trouble with my thermistor because I bought one from a local hardware store (Brazil) and it's different from those listed on the thermistor page. It had a lower beta value and Nophread's python scrypt did not calculate the correct converstion matrix. Obviously it got the temperature all wrong.

So I decided to start from scratch (almost) :-) and made a few changes to the thermistor test source code. In the end I believe the compiled code is smaller because I eliminated the celcius column in the matrix and kept only the rawdata. To do this, I made the table in a way that the celcius temperature is incremented in 25 degree steps, so I can simply increment the celcius variable in 25 every loop.

I made a spreadsheet to help me (and hopefully others) calculate the Beta, Rz, and the conversion matrix for any thermistor.

The spreadsheet and the thermistor test sourcecode are available at my RepRap journal:
[journals.fotki.com]

Regards,
Rod
Sorry, only registered users may post in this forum.