Post by thread:[OT:] DS18S20 Questions

'[OT:] DS18S20 Questions'

        Is there a good description of how the 1-wire bus protocol

works in text, some where?

        I think these DS18S20's are just the thing for a project I

have in mind, but I should first learn how to speak their language.

Thank you.

Martin McCormick WB5AGZ  Stillwater, OK 

OSU Information Technology Division Network Operations Group

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

There's lots of stuff on http://www.maxim-ic.com :-)

It depends what you mean by a "good description"? Are you looking for 

code, pseudo-code, executive summary, engineer's AC timing plots?

-marc

On Tue, 7 Sep 2004, Martin McCormick wrote:

{Quote hidden}>         Is there a good description of how the 1-wire bus protocol

> works in text, some where?

>

>         I think these DS18S20's are just the thing for a project I

> have in mind, but I should first learn how to speak their language.

>

> Thank you.

>

> Martin McCormick WB5AGZ  Stillwater, OK

> OSU Information Technology Division Network Operations Group

> _______________________________________________

> http://www.piclist.com

> View/change your membership options at

> http://mailman.mit.edu/mailman/listinfo/piclist

>

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

"Ruben_J=F6nsson?=" writes:

>though). The trick here is for the master to keep the line low not too short 

>and not too long, which depends on the RC time for the line which in turn 

>depends on the length and the number of devices on the bus. 

        Thanks for all that information.  You have already saved me

from one bad design.  I had this image of stringing telephone wire all

over the house and placing sensors here and there to read

temperatures.  I am sure this can be done, but I think each sensor

will need a buffer or maybe even a PIC to

interface between the DS18S20 and the long wire.  It might even be a

good application for some sort of wireless technique instead of

stringing wire.  In that case, each sensor would definitely need a PIC

to handle the data and communication.  A 12F675/29 or even a 12C508

could probably do that very nicely.

Martin McCormick

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

> "Ruben_J=F6nsson?=" writes:

> >though). The trick here is for the master to keep the line low not too short 

> >and not too long, which depends on the RC time for the line which in turn 

> >depends on the length and the number of devices on the bus. 

> 

>         Thanks for all that information.  You have already saved me

> from one bad design.  I had this image of stringing telephone wire all

> over the house and placing sensors here and there to read

> temperatures.  I am sure this can be done, but I think each sensor

> will need a buffer or maybe even a PIC to

> interface between the DS18S20 and the long wire.  It might even be a

> good application for some sort of wireless technique instead of

> stringing wire.  In that case, each sensor would definitely need a PIC

> to handle the data and communication.  A 12F675/29 or even a 12C508

> could probably do that very nicely.

> 

> Martin McCormick

Hello Martin,

I didn't mean to discourage you, just point out where you could run inte trouble.

Here at work I have a test setup with a 100m cable (standard figure 8 speaker cable) fitted with 32 sensors spread out evenly over the entire length of the cable and it works fine. This is not the maximum limit. This system is based on a (obsolete) SX18 board that has the interface to 32 one-wire sensors in one end and an interface to an RS485 network in the other end. The RS485 network allows me to connect several of these boards to a PC (or any type of master).

Also take a look at the DS2409, micro lan coupler, that basically is a switch which switches in or out single one-wire trunks (with for example some DS18x20 sensors) connected to a main one-wire cable in order to limit the total load as seen by the master for larger networks.

The system I described above is the first generation of a temperture measurement system for grain storage silos. I am now working on the second generation where one RS485 node will have 4 separate outputs driven by 2 12F629s to offload the main processor from the one-wire tasks. Each of these outputs can be connected to 5 lan coupler boards in series. Each board having 4 lan couplers (conncetions to one-wire trunks). Each lan coupler can be connected to 48 sensors (theoretically 4*5*4*48=3840 sensors per RS485 node).

Regards / Ruben

==============================

Ruben Jönsson

AB Liros Electronic

Box 9124, 200 39 Malmö, Sweden

TEL INT +46 40142078

FAX INT +46 40947388

spam_OUTrubenTakeThisOuTpp.sbbs.se

==============================

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

On Thu, 9 Sep 2004, Martin McCormick wrote:

> "Ruben_J=F6nsson?=" writes:

>> though). The trick here is for the master to keep the line low not too short

>> and not too long, which depends on the RC time for the line which in turn

>> depends on the length and the number of devices on the bus.

>

>         Thanks for all that information.  You have already saved me

> from one bad design.  I had this image of stringing telephone wire all

> over the house and placing sensors here and there to read

> temperatures.  I am sure this can be done, but I think each sensor

> will need a buffer or maybe even a PIC to

> interface between the DS18S20 and the long wire.  It might even be a

> good application for some sort of wireless technique instead of

> stringing wire.  In that case, each sensor would definitely need a PIC

> to handle the data and communication.  A 12F675/29 or even a 12C508

> could probably do that very nicely.

Unrelated, but you might want to know that there is a profusion of Chinese 

weather 'computers' on the market, using wireless 433MHz external sensors. 

These could be used as is, or dyi using a temperature sensor, a PIC coder 

and a 433 MHz tx at each location (including outside), using some form of 

time spread transmission (aloha etc).

Peter

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

        Many thanks to you and to Peter L. Peres for your comments.

I am going to send a second message here with a different OT tag

because it is so far off this topic that it needs it's own.

        I think that for special cases such as reading the temperature

of the water pipes under the house, a DS18x20 seems like the best

solution, but the discussion of a remote weather station raises an

interesting sub topic.

Martin McCormick

"=?ISO-8859-1?Q?Ruben_J=F6nsson?=" writes:

>I didn't mean to discourage you, just point out where you could run inte 

>trouble.

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist