Okay , so we have 47kg bottles - now, lp gas cannot be measured using pressure
https://www.cnet.com/uk/how-to/ways-to-estimate-how-much-propane-you-have-left/
So I'm going to use 2 load cells in bridge to get the mass of a bottle. Each cell is good for 50KB, but in a bridge, and add 2 pairs (4 cells) it really gives you 100Kg, since weight is distributed across the cells - you just have to average (add) them.
http://www.ebay.co.uk/itm/like/272478605518?lpid=122&chn=ps&adgroupid=40208487538&rlsatarget=pla-279605276345&adtype=pla&poi=&googleloc=1006598&device=c&campaignid=738085096&crdt=0
https://www.cnet.com/uk/how-to/ways-to-estimate-how-much-propane-you-have-left/
So I'm going to use 2 load cells in bridge to get the mass of a bottle. Each cell is good for 50KB, but in a bridge, and add 2 pairs (4 cells) it really gives you 100Kg, since weight is distributed across the cells - you just have to average (add) them.
http://www.ebay.co.uk/itm/like/272478605518?lpid=122&chn=ps&adgroupid=40208487538&rlsatarget=pla-279605276345&adtype=pla&poi=&googleloc=1006598&device=c&campaignid=738085096&crdt=0
Vbus and 485
The other part besides a load of sensors to gather and log data for is the water heating system. It's Resol VBUS based.
- There is a google group apparently .groups.google.com/group/resol-vbus
- A protocol document https://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&cad=rja&uact=8&ved=0ahUKEwi1sd-crdvSAhUhqVQKHd52DnQQFgghMAE&url=http%3A%2F%2Fforums.parallax.com%2Fdiscussion%2Fdownload%2F60126%2FVBus_Protokol_en_20071218.pdf&usg=AFQjCNHpwhs6jqfw8jY7kaD5AJQHkf4dbA&sig2=HMneGk3YWP9eWLLXufSS4w&bvm=bv.149760088,d.cGw
- Resol Protocol spec (includes electrical and protocol framing) http://danielwippermann.github.io/resol-vbus/vbus-specification.html
A Circuit diagram
A full converter with usb is available here: http://www.resol.de/index/produktdetail/id/13/sprache/en/ price unknown. I sent off a query to get a price anyway.
One thing is that the solar controller unit is obviously not isolated from the power supply, something you will want to do for any equipment which is connected (at least by proximity) to water pipes, and to your roof in event of god ever getting unhappy with you and sending a bolt down. Nearby strikes also cause problems for piping anyway. So with a bit of searching, it appears that the bus supplies 35 ma (or 65 on newer devices) constant current. They appear to use this to drive a LCD remote display in one instance I believe, but not sure. anyways, it got me thinking that an opto isolator (which needs about 10mA) will work here if we power our converter from the line current. The LM339 uses almost no juice nor does our circuit. And the 5.5V signal sent on the master drops to around 4.5V due to our rectifier. which will still run our converter circuit, but may require a transistor on the Raspberry Pi side to clean up if needed, although I'm going to try without a transistor first.
Aha - just read the spec doc, the end of the doc contains a diagram supporting my assertion that you can use an opto coupled circuit to get galvanic isolation powered off of the line. Using a 5.1v zener; just add an extra rectifier diode to pump the supply which is fed through a 330 ohm resistor to prevent the power supply killing the data line, and smoothed off into a 470uF cap. (100nf ripple ceramic needed as usual.)
One thing is that the solar controller unit is obviously not isolated from the power supply, something you will want to do for any equipment which is connected (at least by proximity) to water pipes, and to your roof in event of god ever getting unhappy with you and sending a bolt down. Nearby strikes also cause problems for piping anyway. So with a bit of searching, it appears that the bus supplies 35 ma (or 65 on newer devices) constant current. They appear to use this to drive a LCD remote display in one instance I believe, but not sure. anyways, it got me thinking that an opto isolator (which needs about 10mA) will work here if we power our converter from the line current. The LM339 uses almost no juice nor does our circuit. And the 5.5V signal sent on the master drops to around 4.5V due to our rectifier. which will still run our converter circuit, but may require a transistor on the Raspberry Pi side to clean up if needed, although I'm going to try without a transistor first.
Aha - just read the spec doc, the end of the doc contains a diagram supporting my assertion that you can use an opto coupled circuit to get galvanic isolation powered off of the line. Using a 5.1v zener; just add an extra rectifier diode to pump the supply which is fed through a 330 ohm resistor to prevent the power supply killing the data line, and smoothed off into a 470uF cap. (100nf ripple ceramic needed as usual.)
Just found, they are sold on amazon.
ReplyDeletehttps://www.amazon.co.uk/Resol-Kommunikationsmodul-KM2/dp/B01L98BJK6/ref=sr_1_59?ie=UTF8&qid=1493207651&sr=8-59&keywords=resol+solar