Litres of water are a good measure if you can accurarately measure it into buckets, don't forget to weigh the buckets of course, accurate enough for what you want.Originally Posted by Winter
Trainer wheels optional
Litres of water are a good measure if you can accurarately measure it into buckets, don't forget to weigh the buckets of course, accurate enough for what you want.
Full of Titanium go-fast bits!
ah yeap my 500ml pyrex should measure it out accuratley enough.
What about tools to set the zero and span settings?
"And, look, the luscious and fecund fronds of the Silver Fern has given brilliant birth to a stupendous fruit! A red Hondaberry, desposited by a lesser known species of Plonker Gittus Maximus Idiotus."
Likes a good argument...
You don't actually need to do any of this. Just scale it in the software.
Stick a full tank on it and whatever that number comes out as = full...
Stick an empty tank on it and whatever that number comes out as = empty...
Anything else beyond that is only needed if you want to know the 'actual' weight...
Full of Titanium go-fast bits!
I need to know the actual weight. The System knows we buy lpg at $2.22/kg. The flow meter on the shower knows when i start my showers, reads the bottle weights, and displays any changes in bottle weights in a $ value on the screen in the shower - you can see the cost of your shower as you shower - as a great incentive for me to stop those hour long showers.
"And, look, the luscious and fecund fronds of the Silver Fern has given brilliant birth to a stupendous fruit! A red Hondaberry, desposited by a lesser known species of Plonker Gittus Maximus Idiotus."
Likes a good argument...
How much LPG do you think you would use per shower? It might be getting down into the noise floor there....
Edit: Also, if you know how much LPG you've just purchased then you should have your calibration weight right there? If you've purchased 50kg of LPG then whatever the number changes by from full to empty is 50kg.
Full of Titanium go-fast bits!
0.41kg this morning. works out about 80cents.
"And, look, the luscious and fecund fronds of the Silver Fern has given brilliant birth to a stupendous fruit! A red Hondaberry, desposited by a lesser known species of Plonker Gittus Maximus Idiotus."
Full of Titanium go-fast bits!
I't will be good to be able to measure how much they give me. When I order 45kg do they really give me 40? who would know? ha! Now i can check it :P
"And, look, the luscious and fecund fronds of the Silver Fern has given brilliant birth to a stupendous fruit! A red Hondaberry, desposited by a lesser known species of Plonker Gittus Maximus Idiotus."
Likes a good argument...
.41kg on a total measured weight of 80kg (full LPG cylinder + platform) or so?
Assuming you want at least 10% of that as resolution (so measuring in 40g increments) then at 30mV = 100kg then you're trying to measure a change of 0.4% max range, or 120uV which at 10% steps is 12uV step size...
On the 300mV range ADC the LSB is 4.6uV so you're using about 14 bits worth of the ADC... That's going to be difficult seeing as they normally have +/- 1-4 LSB of repeatibility/non-linearity/noise... That coupled with the fact that you are WAY down on the input voltage, you might find that your results aren't particularly nice.
Are you 100% on using the ADAM module? Something with an adjustable reference would be better so you can make better use of the full range of the ADC... An instrumentation amp would be good alternative which would allow you to feed 0-1V or 0-5V into one of the ADAM modules, meaning the only difficulty was getting a good amp on it.
I can't say for certain what your outcome will be but you're seriously pushing all the tech you're using... anything beyond a 12 bits of accuracy can pose problems, and measuring below mV is difficult at best.
Edit: Yeah, I get what you mean about the delivery vs billing. One of the projects we did was for a similar situation, using a tank level monitor to make sure that the amount of diesel delivered was the same as being billed... We said stick a decent flow-rate meter in but they wanted to use the existing (and cheap) level gauges... The ultimate outcome was that the noise floor was way above whatever billing/delivery inaccuracy they were trying to detect...
Full of Titanium go-fast bits!
Only 100% on it in as much as its installed, and working.
I was looking at other units, but the price of a 2 channel ethernet loadcell controller is horrendous.
Makes me wonder what the setup at the local gas station is? They must be pretty accurate on their weigh systems to charge for it.. Maybe ill go visit in the middle of the night
"And, look, the luscious and fecund fronds of the Silver Fern has given brilliant birth to a stupendous fruit! A red Hondaberry, desposited by a lesser known species of Plonker Gittus Maximus Idiotus."
Likes a good argument...
Yeah, those systems are normally pretty hard-core, but then they have to be to get calibration certs attached... Then again, they have the added advantage of being able to use an appropriate instrumentation amplifier and appropriate ADC reference!Only 100% on it in as much as its installed, and working.
I was looking at other units, but the price of a 2 channel ethernet loadcell controller is horrendous.
Makes me wonder what the setup at the local gas station is? They must be pretty accurate on their weigh systems to charge for it.. Maybe ill go visit in the middle of the night
If it is installed and working then I guess just use it and see what happens, just don't trust the data you're getting too much. I wouldn't be surprised to see the data going all over the show... Building up a simple op-amp circuit could help out quite a lot, you have a lot more accuracy to be gained by using more range of the ADC than you are likely to lose in the amp.
Likes a good argument...
For future reference the general rules when measuring real world values are 1) use as much of the range of your ADC as possible and 2) use the highest signal levels possible. There is a reason most of the analogue SCADA stuff is done 0-10V
Differential signals are good, shielding on your cables is good, sticking any equipment in grounded metal enclosures is good, using short analogue cables and converting to digital asap is good.
Full of Titanium go-fast bits!
I'd like to learn more about constructing an amplifier, I don't know much about that sorta electronics. have soldering iron, will solder! (if I have enough instruction on what im building!)
"And, look, the luscious and fecund fronds of the Silver Fern has given brilliant birth to a stupendous fruit! A red Hondaberry, desposited by a lesser known species of Plonker Gittus Maximus Idiotus."
Likes a good argument...
Well a little dual op-amp on veroboard should be plenty good enough for this application providing it has a nice clean power supply I would think... If you configured the ADAM module to have +/- 5V input and gave the amplifier a gain of 150x then you would be at 4.5V input at 100kg...
That takes your output from being 2.7 counts per 40g change to 12.1 counts per 40g change which should clean up the results from the ADC a little bit.
Given that you've got a buffer there now you can also add a bit of filtering which will clean things up a bit too.
Edit: I can't figure out how to do it nicely so that you could use the full input range of the ADAM module though... My analogue design skills are pretty crap. Should be possible, will ask one of the PhDs that are hovering around the lab, surely they're good for something
Trainer wheels optional
We don't have any chemists or physicists on KB, I'd have bet this afternoon that a whole bunch would be debating over the specific gravity of water and the temperature it should be weighed at, and it's mass above and below 4 deg.
Likes a good argument...
lol yeah. The whole density vs temp thing is cool with water, it's density is highest at around 3 degrees so the bottoms of lakes can stay at 3 degrees if they're deep between surface temps of frozen up to quite a lot higher than 3!
There are currently 1 users browsing this thread. (0 members and 1 guests)
Posting Permissions
Reply With Quote
Bookmarks