|
StBarnabas
|
 |
« Reply #15 on: January 12, 2010, 07:25:17 PM » |
|
HalfBee is a name Wyleu has coined for our PIC based 1-wire slaves. I can't say I agree with him though. Our sensors will be flow meters, irradiance sensors and power meters none which need get too hot. Using HalfBees for temperature sensing though possible is not very sensible as DS18B20s are very reliable and cheap.
|
|
|
|
|
Logged
|
 Gestis Censere. 40x47mm DHW with TDC3. 3kW ASHP, 9kW GSHP, 3kW Navitron PV with Platinum 3100S GTI, 6.5kW WBS, 5 chickens. FMY 2009. |
|
|
|
KLD
|
|
« Reply #16 on: January 12, 2010, 08:51:24 PM » |
|
Well, the DS18S20 are spec'ed up to 125°C I believe. If it could be done at a reasonable effort why not device something with a high temp NTC (there are some spec'ed for up to 300°C) ? Would be nice to have as ET panel sensors, at least for those of us who don't have Wyleu's magic that protects standard 20s against thermal death ;-)
Klaus
|
|
|
|
|
Logged
|
|
|
|
|
wyleu
Guest
|
|
« Reply #17 on: January 12, 2010, 09:03:15 PM » |
|
All I can say is they got to 125 %c ( yes he still can't remember the degree symbol) and made sane readings afterwards.
I'm nasty to kit me...
I've kicked footballs at solar tubes!.
|
|
|
|
|
Logged
|
|
|
|
|
StBarnabas
|
|
« Reply #18 on: January 12, 2010, 09:16:33 PM » |
|
Hi Klaus
it should be possible to design a thermocouple based system or indeed an NTC which would go up to several hundred degrees; The PIC itself would need to be kept below 125 I think. The 12F683 on which they are based has an extended -40 to 125 degree temperature range if I am reading the data sheet correctly. If there is a major interest in a high temperature sensor we could design one.
Sean
|
|
|
|
|
Logged
|
Gestis Censere. 40x47mm DHW with TDC3. 3kW ASHP, 9kW GSHP, 3kW Navitron PV with Platinum 3100S GTI, 6.5kW WBS, 5 chickens. FMY 2009. |
|
|
|
wyleu
Guest
|
|
« Reply #19 on: January 12, 2010, 09:29:53 PM » |
|
It would be interesting to see how linear an DS1820 would be after a good cooking...
|
|
|
|
|
Logged
|
|
|
|
|
KLD
|
|
« Reply #20 on: January 12, 2010, 10:34:17 PM » |
|
Sean
I guess if you were to opt for a thermocouple then you'd also need a temperature compensation for the "reference" end? And that'd be another NTC? For solar thermal applications 300°C should be plenty, so a single HT NTC would do. The sensor could be mounted remote from the PIC, so that the PIC itself is kept in a more "normal" environment.
Klaus
|
|
|
|
|
Logged
|
|
|
|
|
wyleu
Guest
|
|
« Reply #21 on: January 12, 2010, 10:40:23 PM » |
|
Ultimately you'd sling a DS1820 adjacent to the remote reference. Jus to add to the overall confusion.
|
|
|
|
|
Logged
|
|
|
|
|
StBarnabas
|
|
« Reply #22 on: January 12, 2010, 11:10:52 PM » |
|
yes you would need Cold junction compensation for thermocouples but could go up to 1000 degrees+ ?not sure there is a need ?
|
|
|
|
|
Logged
|
Gestis Censere. 40x47mm DHW with TDC3. 3kW ASHP, 9kW GSHP, 3kW Navitron PV with Platinum 3100S GTI, 6.5kW WBS, 5 chickens. FMY 2009. |
|
|
|
wyleu
Guest
|
|
« Reply #23 on: January 13, 2010, 07:48:47 AM » |
|
Do we actually HAVE any solar furnace users?
|
|
|
|
|
Logged
|
|
|
|
|
ericw
|
|
« Reply #24 on: January 13, 2010, 09:52:58 AM » |
|
Although semiconductor manufacturers aren't likely to admit it, devices will withstand over 125 degrees for short periods. Life tests are routinely run for several thousand hours at higher temperatures than this.
Thermistors are frequently spec-ed to 300 degrees the main problem with them that they generally have wide tolerances and are very non linear. However in the these days of cheap micro processors this can be overcome in software as well as the traditional linearisation with a combination a series and shunt resistor.
As thermocouples have very low output voltages they need special interface circuitry and so probably would only be a sensable option to use for normal solar applications if nothing else was available.
|
|
|
|
|
Logged
|
|
|
|
|
KLD
|
|
« Reply #25 on: January 13, 2010, 10:20:42 PM » |
|
Do we actually HAVE any solar furnace users?
Might be interesting for people who make their own pottery? (The sensors, not the solar furnace.  ) |
|
|
|
|
Logged
|
|
|
|
|
hiccup
|
|
« Reply #26 on: January 14, 2010, 01:27:34 PM » |
|
Best not forget the Pt1000 type sensors that the controllers themselves use. Much simpler to interface than thermocouples and very tight tolerance. 0.1% is not too difficult to achieve, and temperature range is very wide.
For simple measurements a constant current source and precision resistor in series with the sensor will give a sensor voltage and reference voltage that can be in the normal measurement range of a PIC. Then a lookup table or piecewise linear approximation will give very good results.
(I used to do this stuff for a living (20 years ago!)
Hic!
|
|
|
|
|
Logged
|
16 x Sanyo HIT250E01 into SB4000TL inverter, 2 x 20 x 58mm Navi Tubes on 22deg roof facing SSE, Gledhill Torrent RE Solar 277litre Store, TDC4 Ether Controller, Xpelair Xcell400BP HRV, Stovax Riva 66 Wood Burner
|
|
|
|
StBarnabas
|
|
« Reply #27 on: January 14, 2010, 03:28:20 PM » |
|
Hi Hic
Pt 1000 sensors would also be very possible. Look up tables are very easy to implement on PICs and at present we are using only about 20% of the memory. It has been obvious from the quality of your posts that you know what you are talking about.
Sean
|
|
|
|
|
Logged
|
Gestis Censere. 40x47mm DHW with TDC3. 3kW ASHP, 9kW GSHP, 3kW Navitron PV with Platinum 3100S GTI, 6.5kW WBS, 5 chickens. FMY 2009. |
|
|
|
djh
|
|
« Reply #28 on: January 14, 2010, 11:23:11 PM » |
|
Might be interesting for people who make their own pottery? (The sensors, not the solar furnace. ) Earthenware is 900 degrees but stoneware is 1300 and porcelain up to 1500 degrees IIRC. Might be stretching the specs a bit! |
|
|
|
|
Logged
|
Cheers, Dave
|
|
|
|
