For what it's worth, you can actually get a free sample of almost all of these Op Amp and Instrumentation Amp chip directly from TI.com I wish I would have realized that before i bought them off ebay, but now I have about 10 more chips headed my way from TI for FREE. Check it out... http://www.ti.com/general/docs/buy.tsp?DCMP=TIHeaderTracking&HQS=Other+OT+hdr_b_buy
Weighing Sensor for WebControl
- Thread starter ChubRock1
- Start date
apostolakisl
Senior Member
Glass syringes are very friction free if you have ever handled one. But the tolerance would have everything to do with which pressure transducer you were using and the diameter of the syringe.
Frankly, I think using a syringe and a gas pressure transducer is a bit of an oddball scale and I would rather use a load cell if I were doing it. Like I said earlier, I started to investigate making a CAI scale and just couldn't come up with all the parts and a way to put them together reasonably.
But, for fun, I may build one anyway. I think I have some glass syringes sitting around and I have a bunch of the pressure transducers.
I finally have all the parts and have been playing around the last couple of days. In general it works well. I have the loadcell mounted to a wooden crossbeam and the meat hook hangs overtop of the loadcell providing the down force. The load cell then runs to my (currently) breadboard where I have the INA125 wired up and the analog out running to the Webcontrol A1. I know there is some noise on the analogs as they do bounce around and at times it is a bit high (like +or- 10) but I will use an AVG calculation in Excel based on the past 10 or so readings so I should get a decent average number to convert to weight.
At this point there is really only one thing I am trying to overcome before I put it into production which is, the load cell(I have three and it is the same on all three) does not register a consistent voltage change until about 100 grams are added, from that point it seems to be a linear scale up to the max weight. I think it must have something to do with offset V being amplified and until it covers that offset range it does not work correctly but I can't seem to figure out how to reduce or remove the lag at low weights.
One work around would be to come up with something that is about 100g to mount on the load cell, then when I hang my meat from that load cell it will really be counting up from the 100g start and I can get voltage changes as small at .01V per gram added or removed. So I know it can be sensitive, but for some reason I can't get it to be that sensitive from 0-1gram.
As always I am open to suggestions, but at least I am very close at this point.
At this point there is really only one thing I am trying to overcome before I put it into production which is, the load cell(I have three and it is the same on all three) does not register a consistent voltage change until about 100 grams are added, from that point it seems to be a linear scale up to the max weight. I think it must have something to do with offset V being amplified and until it covers that offset range it does not work correctly but I can't seem to figure out how to reduce or remove the lag at low weights.
One work around would be to come up with something that is about 100g to mount on the load cell, then when I hang my meat from that load cell it will really be counting up from the 100g start and I can get voltage changes as small at .01V per gram added or removed. So I know it can be sensitive, but for some reason I can't get it to be that sensitive from 0-1gram.
As always I am open to suggestions, but at least I am very close at this point.
CAI_Support
Senior Member
Make sure you have nothing else but loadcell taking the weight. If you have small friction somewhere, that could cause the problem you see.
Put a 220uF to 470uF capacitor crossing the temperature sensor terminal 1 and 3, which is +5V and ground, will reduce the analog input noise to almost nothing. FWD03 wrote that result, too.
Put a 220uF to 470uF capacitor crossing the temperature sensor terminal 1 and 3, which is +5V and ground, will reduce the analog input noise to almost nothing. FWD03 wrote that result, too.
Thanks for the tip on the capacitor I will give that a try.
On the load cell, it is unloaded and friction should not be an issue. I have even tried it just loose not mounted to anything that might be pulling on it, and even then the same issue with lack of sensitivity for the first 100 grams or so. It seems that my unloaded V of .09 might be a tad high but I can't seem to figure out how to trim that if that would even help.
On the load cell, it is unloaded and friction should not be an issue. I have even tried it just loose not mounted to anything that might be pulling on it, and even then the same issue with lack of sensitivity for the first 100 grams or so. It seems that my unloaded V of .09 might be a tad high but I can't seem to figure out how to trim that if that would even help.
On the analog noise issue, I did try the capacitor trick, but that did not work for me. I posted about that on the "Analog Noise" thread so hopefully I can get that figured out.
After re-reading the INA125 data sheet it finally sunk in on how to get the offset adjusted on the chip. I just needed to put a resistor between Pin 5 and ground to, in my case, actually create more of a positive offset so the readings start within the sensitive range of the chip. The end result is that the solution works great. Using a 12V supply and the 10V Ref on the INA125 I am able to get very good resolution starting at 0 grams all the way past 5kg detecting as small as as 5 gram change. If I use a lower value gain resistor I lose the top end (the chip saturates at about 3kg) but I can sense a 2 gram change which is even more than I need for this.
All that's left is to get it all wired into my meat curing chamber and I think we can call this a success. Thanks for all the help.
After re-reading the INA125 data sheet it finally sunk in on how to get the offset adjusted on the chip. I just needed to put a resistor between Pin 5 and ground to, in my case, actually create more of a positive offset so the readings start within the sensitive range of the chip. The end result is that the solution works great. Using a 12V supply and the 10V Ref on the INA125 I am able to get very good resolution starting at 0 grams all the way past 5kg detecting as small as as 5 gram change. If I use a lower value gain resistor I lose the top end (the chip saturates at about 3kg) but I can sense a 2 gram change which is even more than I need for this.
All that's left is to get it all wired into my meat curing chamber and I think we can call this a success. Thanks for all the help.
That's just the beginning! Things to add for version 2, 2.1, 2.2 etc....
* Monitor weight lost each hour, flash a LED at differing rates to indicate how close to "finished" the process is.
* User button to indicate start-of-process. Let PLC weigh and estimate completion time based on temperature and moisture
* Add email reports so you get notice when it's nearly done
* "abnormal" detection - if it stops losing weight at the rate you expect, send a "I need checking" email.
etc etc etc!
CAI_Support
Senior Member
Great job, Chub!
Do you like to share you final schematics? That is a good reference design for many of us.
LarsK
Active Member
+1
Agree! Would be great since there is a lot of things this knowledge can be used for...
