precision scale for checking liquid amount from a peristaltic pump

So i am making a dispensing machine with a peristaltic pump. The problem with peristaltic pumps are that they need to be calibrated, usually involving a precision scale, were one would measure the weight of the fluid (this is common practice in laboratories etc.)

I have found a peristaltic pump with a stepper, but i can’t find any precision scales for arduino. Do you know were to get one? One that could send the measurement over serial would be preferable, because i don’t think that the arduino’s 1024 bit resolution for analog read will give me the precision i need, although i might be wrong here, so suggestions here are also welcome.

Other ideas for pumps that can dispense a precise amount of liquid are welcome as well. The liquid have charcoal in it, so it does contain small lumps (no problem for a peristaltic pump!).

Sincerely Kristoffer.

I should of course have clarified the precision of the scale. This will be a guestimate, but i think that the peristaltic pump will dispense around 2-5 cubic cm of liquid, so i will guess that the range will be from 1-7 g, preferably with a precision of 1/20 to 1/100 of a gram. A standard laboratory scale can do this.

Google ADS1231 for a 24 bit analog->digital + amplifier.

The load cell on the other hand is going to be more difficult to find (in a 1-7g range). Your 1/100th of a gram might be pushing it as well.

That would be one way to go, but since i am not an electrical engineer and would like to avoid building my own circuitry, i think i might have found a better way. Feel free to comment:

This is the scale i will be using: http://www.ebay.com/itm/150g-0-005g-Precision-Digital-Balance-Scale-Accurate-EY-/170549475001?pt=LH_DefaultDomain_77&hash=item27b58a76b9

(It was just the first i found, maybe they come cheaper that this, althoug this is acceptable)

Then use this item, to convert the RS232 to Seriel: http://www.ebay.com/itm/MAX232-Serial-TTL-To-232-Shield-Module-Arduino-Compatible-/280850832597?pt=LH_DefaultDomain_0&hash=item4164039cd5

Is there anything i am missing in transmitting the data to my arduino?

were one would measure the weight of the fluid

the arduino's 1024 bit resolution for analog read will give me the precision i need

Why not calibrate by measuring the volume, that can be done with a bucket :)

X pulses == 10 litres and do the maths from there.


Rob

Palmhoej: Is there anything i am missing in transmitting the data to my arduino?

Can you find the protocol details of the RS232 data sent to/from the scales? It will probably be easy to find/work out but may be proprietary and need help from the scale makers

I don’t know if this is wishful thinking, but i hoped the scale worked in a similar way to this:

Maybe not in the same way, but hopefully i can decipher the input signal to something meaningful. But maybe i should see if i could get the manual or buy one were i can get the manual.

Palmhoej: I don't know if this is wishful thinking, but i hoped the scale worked in a similar way to this:

http://learn.adafruit.com/digital-shipping-scales/using-a-7010sb-scale

Maybe not in the same way, but hopefully i can decipher the input signal to something meaningful. But maybe i should see if i could get the manual or buy one were i can get the manual.

I don't know if the scales work the same way but the manufacturer should give you the info you need or it may be in the manual. All I could find on G&G scales was this manual for the JJ-B series scales.

Graynomad:

were one would measure the weight of the fluid

the arduino’s 1024 bit resolution for analog read will give me the precision i need

Why not calibrate by measuring the volume, that can be done with a bucket :slight_smile:

X pulses == 10 litres and do the maths from there.


Rob

A peristaltic pump needs calibrating as the hose gets worn, the spring gets more lose or the viscosity of the liquid change. A bucket would be one way to go, but would not be as elegant as a scale. I am building this for a lab, so i would like to automate the calibrations as much as possible.

Riva:

Palmhoej: I don't know if this is wishful thinking, but i hoped the scale worked in a similar way to this:

http://learn.adafruit.com/digital-shipping-scales/using-a-7010sb-scale

Maybe not in the same way, but hopefully i can decipher the input signal to something meaningful. But maybe i should see if i could get the manual or buy one were i can get the manual.

I don't know if the scales work the same way but the manufacturer should give you the info you need or it may be in the manual. All I could find on G&G scales was this manual for the JJ-B series scales.

Thanks for the link.

3.8.6 Computer can control balance though RS232. The orders to balance are: 1BH+70H: sending data 1BH+71H: calibration (like “CAL”) 1BH+72H: Counting 1BH+73H: Units change 1BH+74H: Tare Sampling demo, connection method of balance and computer, please download from our website: http://www.gandg.de

I could not find anymore info on their website, but this maybe all i need, provided that this is the same on that scale. I would imagine that the data for weight just keeps flowing, and i dont need the arduino to ask for this.

Palmhoej: A bucket would be one way to go, but would not be as elegant as a scale. I am building this for a lab, so I would like to automate the calibrations as much as possible.

On the contrary, if you are measuring fluid, a "bucket" - or more correctly, a measuring cylinder with a level sensor - would enable the process to be completely automated.

Paul__B:

Palmhoej: A bucket would be one way to go, but would not be as elegant as a scale. I am building this for a lab, so I would like to automate the calibrations as much as possible.

On the contrary, if you are measuring fluid, a "bucket" - or more correctly, a measuring cylinder with a level sensor - would enable the process to be completely automated.

Using a balance for accurate measurements of weight is usually a better option for small volumes and high accuracy. In most lab settings its much easier to use a balance to measure a weight and calculate the total volume using the known density of the liquid. Plus every lab has a balance already, and anything fairly recent is likely to have serial output already to go.

Just keep in mind that your weight (or volume) measurements could vary quite a lot if you have a very inhomogeneous sample like the one you're describing. Any large chunks could create deviations from average pump performance, so just make sure your calibration time is long enough to average these out. Better yet, homogenize your sample some how (blender, shaking, stirring) to remove any of these problems.

The charcoal solution will be standing on a magnet stirrer to make sure that the solution is as homogenized as possible.

Thanks for your inputs :-)

(not to end this thread, i am all ears if anyone has more suggestions or problems)