Yes, I have noticed that there are some problems on really long lines. I have one sensor connected running over about 110m of ethernet cable in quite noisy environment, and have about 15 sensors on other network, where some legs are about 20m long. Raspberry pi or ordinary PC with a TTL usb-serial adapter is used at host end, together with a two-way buffer/amplifier circuit, built on 74HC125. This way I get two independent channels for 1wire. On the far end, capacitors are used, and pull-up resistors too. It actually helps if just a smaller pull-up resistor is used at host end, like 2k or so, instead of standard 5.1k. Most probably it's out of recommended range, but I have not noticed any bad effects.
Now I'd like to have a more mobile solution, as power is available almost everywhere, but messing with data cables is not always a good option, especially for temporary setups. Sure it's possible to just preconfigure sensor id's, but it would be much easier if I could just plug necessary sensors and get data. There is another easy way to do this though, using a raspberry, or even WR703N as a mobile unit, with digitemp software. On Thu, Nov 27, 2014 at 8:54 AM, Vasile Surducan <[email protected]> wrote: > I have a system with three DS18B20 on one bus with separate power supply > wires running continuously from about three or four years. > You have to read separately the ID for each sensor (one sensor once on the > bus) , then to address the sensors based on the ID numbers. > I've noticed if you have a really long bus, that pull-up resistors and > decoupling capacitors near every sensor will help a lot. > > best wishes, > Vasile > > On Thu, Nov 27, 2014 at 8:42 AM, KPL <[email protected]> wrote: >> >> Thanks guys, I will check both of these. >> I need just that one type, ds18b20, for sure; >> Probably this might be a good reason to try pic18 finally, as I even >> have some of them in a drawer, just never had a real reason to use. >> >> Probably I could try to design or port some lib myself, but this seems >> to be quite difficult to debug, without proper tools. I have >> OpenSniffer logic analyzer, but it has difficult times to deal with >> those very long start/reset pulses. >> >> On Thu, Nov 27, 2014 at 2:42 AM, NoseHair <[email protected]> wrote: >> > >> > Hi KPL, >> > >> > Here's a JAL version of the original Dallas application code for >> > 8-bit >> > processors. It's >> > basically the Dallas 1-wire search routine with wrappers calling the >> > existing, JAL2 >> > 1-wire read and write functions. The only thing you need to do is to >> > add >> > handlers for >> > each type of 1-wire device that you're using (the DS18S20 thermometer >> > is >> > already >> > in there and I have some code for the DS1822 in another version >> > somewhere, if I >> > can find it). >> > >> > http://www.pucebaboon.com/JAL2/1-wire-search.tar >> > >> > The main problem with the search routine is that it uses a -lot- of >> > memory for those >> > massive serial/ID numbers. If you're only ever planning to use one >> > type >> > of device >> > and no CRC, then you could save a lot of memory and increase the >> > number >> > of >> > devices on your bus by just ignoring those parts of the serial >> > number. >> > Having said that, this >> > is never going to be viable for the smaller PICs (you might only be >> > able >> > to handle one >> > or two sensors). >> > >> > From the original directory I can see that this was last compiled on >> > Jan >> > 1st, which means it >> > was using this version of JAL2:- jallib-pack-bee-jalv24q-20131124. >> > >> > As always, improvements and bug reports welcomed. >> > >> > >> > -John Little- >> > >> > >> > >> > >> > >> > >> > On Wednesday, 26 November 2014 22:36:59 UTC+9, KPL wrote: >> >> >> >> Hi, >> >> >> >> I have used ds18b20 library from jallib many times, but now I would >> >> like to read multiple sensors, connected to the same bus. The only >> >> example I see in trunk is tst_ds18b20_multi_serial.jal, which just >> >> reads predefined sensors, by their ID's - but can not detect sensors >> >> and their ID's by itself. >> >> >> >> The only mentioning of search algorithm done in JAL i see here: >> >> http://mientki.ruhosting.nl/data_www/pic/libs_hw/dallas_1_wire.html >> >> but link to include file delay_20mc.jal returns 404, and procedure >> >> itself seems complicated, together with bank_arrays. >> >> Are there any other attempts at this? >> >> >> >> I am reading AN187, but... somehow... probably need more sleep or >> >> something. >> >> >> >> The general idea is to build a device with currently "hot" ESP8266 >> >> module, which would periodically check sensors connected to it, and >> >> send sensor id's together with corresponding temperatures to a web >> >> server. >> >> >> >> -- >> >> KPL >> > >> > -- >> > You received this message because you are subscribed to the Google >> > Groups >> > "jallib" group. >> > To unsubscribe from this group and stop receiving emails from it, send >> > an >> > email to [email protected]. >> > To post to this group, send email to [email protected]. >> > Visit this group at http://groups.google.com/group/jallib. >> > For more options, visit https://groups.google.com/d/optout. >> >> >> >> -- >> KPL >> >> -- >> You received this message because you are subscribed to the Google Groups >> "jallib" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> To post to this group, send email to [email protected]. >> Visit this group at http://groups.google.com/group/jallib. >> For more options, visit https://groups.google.com/d/optout. > > > -- > You received this message because you are subscribed to the Google Groups > "jallib" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at http://groups.google.com/group/jallib. > For more options, visit https://groups.google.com/d/optout. -- Kārlis -- You received this message because you are subscribed to the Google Groups "jallib" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/jallib. For more options, visit https://groups.google.com/d/optout.
