Automated volume sensing idea/help

Ok, the folks in here seem more whack than the other forums, and I'm counting on that for ideas/feedback for optical volume sensor.

Did a thread search and found a bunch of float switch setups, but don't like the idea of having extra stuff hanging in my keggles and was concerned about accuracy when moving between keggles. And proximity and other sensors I found were major $$$$$.

Working on my off-grid solar setup, I stumbled on this solar tracker module.
http://www.redrok.com/led3xassm.htm#paypalled3x
It uses the hysteresis or differential of LEDs to trigger a motor to move solar panels from east to west to follow the sun.

Got me to thinking that I could mount the two sensors on a chingadera clamped to sightglass that could be moved up and down to sense fill or regulate fly sparge flow. They would detect a buoyant cylinder or sphere in the sightglass.

For "fill" you would use a relay to bypass the "on" LED and rely on the "off" LED to stop filling at whatever level you set the clamp on sightglass.
For the sparge, the level would be set just above the grainbed and sensor would control pump or solenoid to maintain flow to match the output of MLT, or just have it on until level is reached/maintained.
Probably having a light source would add to reliability, but that should be pretty straightforward.
All feedback appreciated.
Thanks, Dale

I may be in over my head here, but why not premeasure the desired flow instead of measuring after it arrives in the pot?

http://www.buygpi.com/mechanical-batching.aspx

grathan said:

I may be in over my head here, but why not premeasure the desired flow instead of measuring after it arrives in the pot?

http://www.buygpi.com/mechanical-batching.aspx

Click to expand...

The best meter that place has only goes down to 5 gallon increments. Yeah, that's not going to work....

Why not use a pressure sensor like the brewtroller?

crazyirishman34 said:

Why not use a pressure sensor like the brewtroller?

Click to expand...

Not sure I am comfortable with the variables. Barometric pressure, temperature/viscosity. Seems like frequent tuning could be required.

Been viewing BT forums and don't recall anyone shouting they've seen Jesus. Do you have any positive experiences with BT bubbler?

Kind of like the idea of analog, yes/no measurement that can be easily set/adjusted.

I need to apologize for not stating intent. Building Brutus-like gas/electric hybrid/bi-sexual using BT or BCS for control and thinking BCS expansion module would handle a bunch of discrete inputs for volume measurement via optical sensors.

Will be posting up later on the peculiar build, but had the optical thing on my brain from the solar project.

Thanks for your replies, Dale

My set up uses the pressure sensor. It is pretty spot on for fill and sparge volumes. I am not using the bubbler setup I drilled a hole in the bottom of the pot and ran a silicone tube to the pressure sensor. I just calibrated it one time by adding water one gallon at a time till the pot was filled. The BT can extrapolate from there. In my experience it is accurate to a tenth of a quart.

There are plenty of devices to measure level along an analog scale.

They are just very expensive.

http://www.omega.com/toc_asp/sectionSC.asp?section=K&book=green&flag=1

crazyirishman34 said:

My set up uses the pressure sensor. It is pretty spot on for fill and sparge volumes. I am not using the bubbler setup I drilled a hole in the bottom of the pot and ran a silicone tube to the pressure sensor. I just calibrated it one time by adding water one gallon at a time till the pot was filled. The BT can extrapolate from there. In my experience it is accurate to a tenth of a quart.

Click to expand...

Interesting, can detect a double-jigger, that is awesome achievement that wasn't apparent to me from BT forums. That degree of resolution would seem a stretch via analog without a bunch of "tuning", now I gotta know more.

Do you get a display, and how is sensor used in your brew process?
Also, would your setup work with gas-fired keggles in the event I chose (or switched between) direct fired RIMS or gas HLT and HEX?

Maxkling said:

There are plenty of devices to measure level along an analog scale.

They are just very expensive.

http://www.omega.com/toc_asp/sectionSC.asp?section=K&book=green&flag=1

Click to expand...

Thanks, hadn't seen that source before, but just as expensive as those I did find. That's why the optical $24 kit and $39 assembled unit intrigued me. Seemed cheapest way I'd found with potential to control sparge input and figure out fill later.

Thank you both for your input.

I'm waiting to receive AD620's, I bought a cheap bathroom scale already and harvested the load sensors from it. I'll let you know, if it works reliable. https://www.sparkfun.com/products/10245

I wonder if you could set up the load sensors to output 0 to 5 volts?

Yes it can, that's what I'm planning to do. And then use that info with the Arduino. Try "Wheatstone bridge" in the Google.

I use load cells out of a cheap electronic bathroom scales and use a single supply instumentation amp



R3 thru to R6 represents the load cells, I have adjusted the gain to give me 5v at 102.3 kg which gives me 100mL resolution on a 10 bit ADC. I have been using it on my HLT for a couple of years now and has worked well, my theory is that if you know how much water you put in you will be able to work out how much you will get out because losses are fairly static.

cheers steve

I guess all of these load measurements are only good if you're dealing with straight water? Or if you know the SG of your wort....

Best way is to submerse 2 parallel electrodes in there and measure the resistance (conductivity) between them. The electrodes could be sank into the kettle, or in a sight-gauge-like assembly outside the main volume.

The electronics to monitor doesn't need to be complex; a simple voltage divider would work.

Since the resistivity of the wort will change with recipe, etc, another pair could be used to normalize the depth measurement. This pair would be small, i.e., 1/4", and would provide a constant resistance value. This pair goes to the bottom of the tank and is wetted as soon as wort is introduced.

Use the ratio of the two electrode pairs to determine depth in any conductive fluid.

That seams like the most straight forward way of doing things but the electronics are way above my head.

Vesku said:

Yes it can, that's what I'm planning to do. And then use that info with the Arduino. Try "Wheatstone bridge" in the Google.

Click to expand...

Wow, damn near ancient technology:rockin:

passedpawn said:

Best way is to submerse 2 parallel electrodes in there and measure the resistance (conductivity) between them. The electrodes could be sank into the kettle, or in a sight-gauge-like assembly outside the main volume.

The electronics to monitor doesn't need to be complex; a simple voltage divider would work.

Since the resistivity of the wort will change with recipe, etc, another pair could be used to normalize the depth measurement. This pair would be small, i.e., 1/4", and would provide a constant resistance value. This pair goes to the bottom of the tank and is wetted as soon as wort is introduced.

Use the ratio of the two electrode pairs to determine depth in any conductive fluid.

Click to expand...

I think I like this approach.

You could set the default resistance to the range of thermistors used by PIDs and calibrate/display the results as volume instead of temperature?

Are there modules/boards or any off-the-shelf solutions to bridge between electrodes and PID?

Guessing the electrodes would need to have some sort of native resistance for there to be an easily measurable change as level of submergence varies?

I'm wondering how much the linearity would be affected by differences in the conductivity of the fluid? Would it be tough to calibrate each use, especially as the chemistry of the wort changes during the mash, and whatever effect that has on conductivity?

Appreciate the input.

Dale

matho said:

I use load cells out of a cheap electronic bathroom scales and use a single supply instumentation amp

R3 thru to R6 represents the load cells, I have adjusted the gain to give me 5v at 102.3 kg which gives me 100mL resolution on a 10 bit ADC. I have been using it on my HLT for a couple of years now and has worked well, my theory is that if you know how much water you put in you will be able to work out how much you will get out because losses are fairly static.

cheers steve

Click to expand...

Interesting. Do you have any pics of the physical setup? What is max load for the load cells?

Dale

MrNatural said:

Wow, damn near ancient technology:rockin:


I think I like this approach.

You could set the default resistance to the range of thermistors used by PIDs and calibrate/display the results as volume instead of temperature?

Are there modules/boards or any off-the-shelf solutions to bridge between electrodes and PID?

Guessing the electrodes would need to have some sort of native resistance for there to be an easily measurable change as level of submergence varies?

I'm wondering how much the linearity would be affected by differences in the conductivity of the fluid? Would it be tough to calibrate each use, especially as the chemistry of the wort changes during the mash, and whatever effect that has on conductivity?

Appreciate the input.

Dale

Click to expand...

It would be linear as long as the two rods were spaced evenly apart.

The second set of electrodes automatically calibrates to the conductivity of the fluid, so it's not a factor. That's why the ouput from this is a ration of Electrodes1 / Electrodes 2, where Electrodes 2 are small and the depth of the fluid doesn't affect them.

MrNatural said:

Interesting. Do you have any pics of the physical setup? What is max load for the load cells?

Dale

Click to expand...

Do you see, there's that "Wheatstone Bridge" right there

Have anyone tried to build the 2 electrode system? I've seen some discussions about it, but they all measured capacitance rather than impedance. There was issues with heat affecting the readings. Also an AC current is needed for the measurement to avoid electrolytic corrosion.

If only a few different level indication are needed, here's a simple circuit that works: http://www.electronic-circuits-diagrams.com/alarmsimages/6.gif I didn't have LED's on mine, but send the info to the Arduino. I also had 3 CD4066's get more levels.

If you just want on off control you could rig up some type of magnetic float in your sight glass that would pick up a read switch mounted on the outside of the sight glass, just slide the switch up and down the sight glass to set your level. You could have multiple switches for different levels, hi, low, hi-hi, low-low.

Vesku said:

Do you see, there's that "Wheatstone Bridge" right there

Have anyone tried to build the 2 electrode system? I've seen some discussions about it, but they all measured capacitance rather than impedance. There was issues with heat affecting the readings. Also an AC current is needed for the measurement to avoid electrolytic corrosion.

If only a few different level indication are needed, here's a simple circuit that works: http://www.electronic-circuits-diagrams.com/alarmsimages/6.gif I didn't have LED's on mine, but send the info to the Arduino. I also had 3 CD4066's get more levels.

Click to expand...

Yeah, I was wondering about the electrochemistry of a 2 electrode setup too. You'd definitely want the waveform you use to measure the impedance to be charge balanced (waveform integrates to 0 - i.e. equal amounts of positive and negative current) to avoid redox reactions on the electrodes. Even then, the baseline impedance of the electrodes would likely drift a bit over time and require calibration every now and then, but with some silver/silver-chloride electrodes you'd probably be ok for a while.

Of course, since you may not care much about raw impedance versus just a change in impedance as the level rises it may not be too much of an issue.

MrNatural said:

Interesting. Do you have any pics of the physical setup? What is max load for the load cells?

Dale

Click to expand...

The Max load on the bathroom scales was 150Kg so physically they should be good for that but I have it setup to give me full scale at 102.3Kg

here are some more pics

the 4 load cells removed from the bathroom scales



a jiffy box connected to the underside of the scales containing the instrumentation amp



the complete unit



the unit being tested



and the results



I'm happy with it and its accurate for water in the Hlt and if I know the amount of water going in then I can work out the finished wort volume.

cheers steve

Thanks Steve. Pretty slick setup. My problem is I'm making a "bisexual" rig using electric or propane, or combination of both.

Dale

That seams like it would work for setting the strike water in your Mashtun also. It is too bad the Brewtroller can't make a weight correction for the expected gravity.

MrNatural said:

Thanks Steve. Pretty slick setup. My problem is I'm making a "bisexual" rig using electric or propane, or combination of both.

Dale

Click to expand...

there is no reason you can place the load cells under the burner ( a fair distance away to keep them sake from the heat) and zero the scales before you add the water.

crazyirishman34 said:

That seams like it would work for setting the strike water in your Mashtun also. It is too bad the Brewtroller can't make a weight correction for the expected gravity.

Click to expand...

yes you could weigh the mash tun as well, then you could tell when the tun is drained because the weight would stop dropping. I wonder how close you could estimate the gravity by ((weight of liquid out + estimated losses)/weight of water in). Deadspace of the mash tun is easy to know but the absorption rate of the grain would be a bit hit or miss. Maybe when I get some time I'll have a play.

cheers steve

Also I have had this page bookmarked for a long while but haven't followed up on it

http://www.automatedsonix.com/son40.htm

would be a good way to sense volume but I don't know how cheep it would be

Anybody tried one of these yet?
https://www.sparkfun.com/products/9072

matho said:

there is no reason you can place the load cells under the burner ( a fair distance away to keep them sake from the heat) and zero the scales before you add the water.
cheers steve

Click to expand...

DOH :smack:
Going HERMS, so could use something like this to maintain MLT weight during sparge.

matho said:

Also I have had this page bookmarked for a long while but haven't followed up on it

http://www.automatedsonix.com/son40.htm

would be a good way to sense volume but I don't know how cheep it would be

Click to expand...

Almost looks plugnplay, but don't see a price anywhere. Probably comes with industrial sized pricetag.

Gunfighter04 said:

Anybody tried one of these yet?
https://www.sparkfun.com/products/9072

Click to expand...

I'd like to know that too. Pretty cheap, but only rated to 50°C.
BCS would see that just like a 10K thermistor. Strap it to sightglass, drop in magnetic float and set up custom display for volume.

matho said:

Also I have had this page bookmarked for a long while but haven't followed up on it

http://www.automatedsonix.com/son40.htm

would be a good way to sense volume but I don't know how cheep it would be

Click to expand...

I managed to rig up an arduino with a HC-SR04 to do level measurement in a keg. It returned a pulse whose width was inversely proportional to the volume of fluid in the keg. Then i got married had a child bought a house and now have less time than marty mc fly

http://www.amazon.com/dp/B004U8TOE6/?tag=skimlinks_replacement-20

I'm using this ultrasonic sensor with my arduino build to tell my unit when to fire the heating element (safe volume level) and when to turn off the heating element if I'm in danger of a boil over. Can't wait to say goodbye to the chance of a boil over while I'm not watching the kettle. Gotta love automation!

BenS said:

http://www.amazon.com/dp/B004U8TOE6/?tag=skimlinks_replacement-20

I'm using this ultrasonic sensor with my arduino build to tell my unit when to fire the heating element (safe volume level) and when to turn off the heating element if I'm in danger of a boil over. Can't wait to say goodbye to the chance of a boil over while I'm not watching the kettle. Gotta love automation!

Click to expand...

That is way cool. What is the state the sensor uses to determine impending boil-over?

It's an if statement based on the distance the sensor pings. If the distance it sees to the liquid level is short enough, then it turns off the little output pin to the heating element.

Can you tell us what type of ultrasonic sensor you're using?

Brewer Gerard said:

Can you tell us what type of ultrasonic sensor you're using?

Click to expand...

http://www.amazon.com/dp/B004U8TOE6/?tag=skimlinks_replacement-20

same one that i setup. it worked over a keg of water at ambient temperature but i wonder how good it will be when mounted over a steamy hlt or aimed at a foamy potential boilover.

Stream does not (or should not) be seen by the sensor. I am more worried about debugging the sensor for the temp change and the nonreflective surface of the foam. Haven't gotten it yet...

It's just that i've done a very modest bit of research on this and i found even with far more sophisticated sensors there is a degree of tuning required based on the environment in question. That's not to say this approach can't work.

How about this thread be used to motivate us to be he first to find an actual practical use for this HC-SR04 module. Winner can prove that the module measures accurately enough to automatically deliver a specified volume of strike water to a mash tun. This means the system has to stop the HLT being filled at the desired volume, hit the desired strike temp and subsequently stop the pump to the mash tun when enough strike water has been transferred.

In case any one thinks i have a head start i still don't have power/water in my brewery

What about a liquid level sensor
https://www.sparkfun.com/products/10221

jrubins said:

What about a liquid level sensor
https://www.sparkfun.com/products/10221

Click to expand...

Pretty slick, but datasheet specs only rate it to 150°F, so would probably fail in brewing conditions.

MrNatural said:

Pretty slick, but datasheet specs only rate it to 150°F, so would probably fail in brewing conditions.

Click to expand...

Last I read, they are working on a higher temp one, but no idea when that will happen.