Volume Sensing - What are You using

[ryane]

Bit of background - Until recently Ive had a sort of modified Kal electric brewing set up, however I had my first child about a year ago now and time to brew is very little so I have began the process for getting together a completely automated brewing rig. This will allow me to still drink and formulate recipes, but skip on some of the mundane brewing tasks

Ive sourced most of my parts and pieces (valves, rims, elements, plc, hmi, etc, etc) and have access to the tools for custom machining many things I need. Yet one thing that I have yet to pull the trigger on is my method for volume/flow measurement. Ideally I would like to be able to measure volume for mash additions and sparging, and probably watch flow rates during my re-circulation.

When first thinking about this issue my thought was to use flow meters (aliexpress ~300$/ea), ultra sonic level meters(automation direct ~200$), or differential pressure transducers (ebay ~50-150$). In fact Ive found sources for all of said instruments for relatively cheaply, but each has its pros/cons.

At first I leaned towards the ultrasonic level sensor, but I never really could come up with a good way to mount them to the tops of keggles that also had lids (1"MPT) The thought of steam/condensation on the bottom of the unit also concerned me because I have had some recent issues with several similar units at work. I still like the idea because of the simplicity of it, but I'd need a good mounting method and a way to deal with the aforementioned issue.

Next I really started looking at flow meters, and its likely I'll add one in no matter what I do just to monitor mash recirc, but for volume control I'd be worried about the accuracy in the low range. Then theres the implementation issue, even though I found some cheap ones Im worried about documentation because they are Chinese and Ive had horrible luck in the past with this type of instrument. There are also somewhat inexpensive Bürkert or EL flow meters but they tend to be slightly larger than what I need thus reducing the accuracy

As it stands Im leaning towards using a differential pressure transducer in each of my brewing vessels (HLT, MT, BK) for volume measurement, and I can get them very cheap (~50-100/ea). You just have to account for the density of the wort in the kettle (estimated from recipe). While right now Im probably just going this route, the fact that I can be very anal about these types of details, this density issue slightly bothers me. If I could, I would prefer to have something in place that doesnt need an estimated adjustment factor.

With all that said, what is everyone else using? I would love to see pictures and or the specific models of instrumentation as well. Money isnt that big of a deal since Im blow it left and right on this build but some common sense is necessary because I got quite a nasty stink eye when my wife saw the bill for the butterfly valves I bought

 

[mliptack]

Subscribed - Looking for the same answer

 

[joeycmsu]

I don't have this on my homebrew setup, but have you considered continuous capacitive level transmitters? There are sanitary options, and the density and temperature of the liquid don't matter. They are expensive, so you'd have to find one on eBay, like this one:

http://pages.ebay.com/link/?nav=item.view&alt=web&id=301444239046

 

[Btaz]

I haven't done this yet, but will probably do something like this when I get far enough along in my build. Wort should be mildly conductive to allow a simple circuit to detect liquid levels. This link https://www.elprocus.com/simple-water-level-controller-using-microcontroller/ provides a good starting point

 

[ryane]

Ive seen the lumenite probes but I believe the working temp is only 160F??

 

[ryane]

I haven't done this yet, but will probably do something like this when I get far enough along in my build. Wort should be mildly conductive to allow a simple circuit to detect liquid levels. This link https://www.elprocus.com/simple-water-level-controller-using-microcontroller/ provides a good starting point

I need accurate volume measurement that can be variable, e.g. Recipe X needs 12.5qt for mash, Recipe Y needs 24qt, and Recipe Z needs 16qt

Now you could do something with a very fine scale like you suggested, but frankly I would prefer an off the shelf instrument at this point, Im all for DIY but not in this case

 

[trimixdiver1]

You could use a pressure transmitter to provide an analog output. You could measure the pressure of the water in the vessel. Look at ebay for rosemount pressure transmitter. Some are set up with a display too, but if you use the analog signal and a controller, you can scale it anyway you want.

 

[ryane]

You could use a pressure transmitter to provide an analog output. You could measure the pressure of the water in the vessel. Look at ebay for rosemount pressure transmitter. Some are set up with a display too, but if you use the analog signal and a controller, you can scale it anyway you want.

Thats what Ive been leaning towards, but the anal engineer in me prefers direct methods of measurement.....

I was perusing your build thread today and noticed an IFM flowmeter in one your pics. The temp rating on those units is pretty low, so Im curious what your using this for? Im pretty sure I'll scoop one up for metering water into my HLT and cooling, but it would be cool to use for sparge water as well

 

[Aerobrew]

This is purely speculative at this point since I have not had time to find just the right way to mount it but the idea behind it would work great and is tested in many industries. I had gone through the idea of flow meters, ultrasonic and other ideas in the past and each one either got really expensive, or had issues with temperature etc. The nice thing about water is since it has a set weight it should be able to just build it on a load sensor. The trick is making that load sensor dissipate any heat etc. I have done it on some smaller projects this way and can't see a reason why it wouldn't work, just have to find the right way to get it to work. This can even be done with the mash, since you can just zero the scale or add in the weight of the grain before the water. Just an idea to throw out there, it is on my list to experiment with the different load sensors and heat dissipation mounts for them, but haven't had the chance yet. Just wanted to add it as an option although I don't have a complete way to do it yet.

 

[trimixdiver1]

Thats what Ive been leaning towards, but the anal engineer in me prefers direct methods of measurement.....

I was perusing your build thread today and noticed an IFM flowmeter in one your pics. The temp rating on those units is pretty low, so Im curious what your using this for? Im pretty sure I'll scoop one up for metering water into my HLT and cooling, but it would be cool to use for sparge water as well

The max temp of the IFM is 176f, it's used on my HLT, so I'm good on temperature range.

 

[ryane]

The max temp of the IFM is 176f, it's used on my HLT, so I'm good on temperature range.

Just for filling the HLT?

Like I said I'll probably get one, but Im debating the need because if I measure the level in the HLT the flow meter is redundant

 

[trimixdiver1]

Just for filling the HLT?



Like I said I'll probably get one, but Im debating the need because if I measure the level in the HLT the flow meter is redundant

I have an analog GEMS float sensor for level, the IFM gives me temperature and a .1g per pulse for the flowmeter. You can use it for a batch fill application, I have the same float setup in the MLT too, but use the IFM to help give a redundant fill level.

 

[Scotty_g]

At first I leaned towards the ultrasonic level sensor, but I never really could come up with a good way to mount them to the tops of keggles that also had lids (1"MPT) The thought of steam/condensation on the bottom of the unit also concerned me because I have had some recent issues with several similar units at work. I still like the idea because of the simplicity of it, but I'd need a good mounting method and a way to deal with the aforementioned issue.

>skip<

As it stands Im leaning towards using a differential pressure transducer in each of my brewing vessels (HLT, MT, BK) for volume measurement, and I can get them very cheap (~50-100/ea). You just have to account for the density of the wort in the kettle (estimated from recipe). While right now Im probably just going this route, the fact that I can be very anal about these types of details, this density issue slightly bothers me. If I could, I would prefer to have something in place that doesnt need an estimated adjustment factor.

I have encountered ultrasonic level transmitters at work and we had a lot of trouble with them in steamy applications. There are new ones that work on higher frequencies that supposedly can see through the steam better, but those start at $4k.

You could use 2 dPT's (one with a fixed range to calculate density) to monitor your volume, but you are still going to run into accuracy issues. Vessels with a lot of turbulence (heavy mixing or vigorous boiling) don't always read density well.

You could use a dPT for filling the HLT, but for variable level measurements in a vessel full of wort, I'd recommend a sight glass and adjustable level switch (capacitance probe or float switch that adjusts up and down). If you want 16L (or whatever), set your switch to make/unmake at that level.

 

[Bill_in_VA]

Could someone enlighten me, why isn't just measuring the water volume on the cold side good enough? Vessel losses and grain absorption would be empirically known or calculated close enough, no?

 

[atoughram]

We've used pressure transducers on 10,000 gallon anodize tanks at work - works well there but the fluid column is eight feet.

What about a load cell?

I was going to play around with a flowmeter - still have them but lost interest.

I also thought about a sliding light transmitter/receiver on the sight glass - set it to the height you want it to fill and let it go.

 

[ryane]

Could someone enlighten me, why isn't just measuring the water volume on the cold side good enough? Vessel losses and grain absorption would be empirically known or calculated close enough, no?

I need them to measure mash water volumes, monitor the sparge (stuck, rising, lowering level), verify the amount of liquid in the kettle, prevent dry firing elements, and the best reason of all, because I want them for my automated build

We've used pressure transducers on 10,000 gallon anodize tanks at work - works well there but the fluid column is eight feet.

What about a load cell?

I was going to play around with a flowmeter - still have them but lost interest.

I also thought about a sliding light transmitter/receiver on the sight glass - set it to the height you want it to fill and let it go.

Load cells are gonna be a lot of fiddling to get it to work right, and I want to have a spare sitting on the shelf in case something quits working on me. With DIY instrumentation this is tricky to do quickly

As to pressure transducers, the small water column we are talking about can be measured fairly easily with good selections in instrumentation and proper scaling. I actually just pulled the trigger (last night) on two pressure transducers for measuring the boil kettle and my mash tun, and two flow meters for mash volumes and metering into the HLT. All in all I got 2 x 1/2" ultrasonic flowmeters (new) and 2 x 2" sanitary pressure transmitters (surplus) for ~300$ + shipping - not a bad deal

That said because of how cheap I got everything I'm probably going to add some redundancy to my HLT for a more direct measurement of water than just influent flow, leaks do happen and I want extra assurances that I wont dry fire my HLT, and who knows maybe to the boil kettle as well. I do like the continuous floats that trimixdiver1 mentioned above and I may look into getting something similar

 

[thekraken]

You might be interested to read what this guy did: http://www.vandelogt.nl/htm/hardware_uk.htm

He has a good short write up on the different methods that he considered, he ended up going with pressure transducers. The one's he used are only about $11 on digikey.

Like Bill_in_VA suggested, I will probably only be measuring my fluids on the cold side so that gives me the option of the eTape as well. I'm investigating non-invasive flow switch ideas to prevent dry fires in the case of stuck sparge, pump failure, etc.

I looked into the load cells too, I figured that with your typical load cell that there would be too much drift with time and temp over the course of a brew day that your readings could be off by a meaningful margin. I have no experience with these so someone please correct me if that is wrong.

Here is some more reading material if anyone is interested: http://www.sensorsmag.com/sensors/leak-level/a-dozen-ways-measure-fluid-level-and-how-they-work-1067

 

[thekraken]

You could use a dPT for filling the HLT, but for variable level measurements in a vessel full of wort, I'd recommend a sight glass and adjustable level switch (capacitance probe or float switch that adjusts up and down). If you want 16L (or whatever), set your switch to make/unmake at that level.

There is an idea, this approach would seem simple, cheap, food safe, etc. How about a sight glass with an adjustable IRLED/phototransistor that you just slide up and down the sight glass to set your level? (edit) Same thing atoughram said.

Anyway, it sounds like your going with the pressure transducers, I'm very interested in what kind of accuracy and resolution you are able to achieve. Do you have links to the products you purchased?

 

[trimixdiver1]

You might be interested to read what this guy did: http://www.vandelogt.nl/htm/hardware_uk.htm

He has a good short write up on the different methods that he considered, he ended up going with pressure transducers. The one's he used are only about $11 on digikey.

Like Bill_in_VA suggested, I will probably only be measuring my fluids on the cold side so that gives me the option of the eTape as well. I'm investigating non-invasive flow switch ideas to prevent dry fires in the case of stuck sparge, pump failure, etc.

I looked into the load cells too, I figured that with your typical load cell that there would be too much drift with time and temp over the course of a brew day that your readings could be off by a meaningful margin. I have no experience with these so someone please correct me if that is wrong.

Here is some more reading material if anyone is interested: http://www.sensorsmag.com/sensors/leak-level/a-dozen-ways-measure-fluid-level-and-how-they-work-1067

I can tell you load cells will drift, especially under a load ( even the empty kettle)

I have a DP sensor I will be placing on the output of the MLT to add redundancy to the float and monitor grain bed compaction. I've been toying with the idea of the other side of the DP being open to atmosphere or placed in the MLT, but since the levels in the MLT will change with recipe, I'm leaning toward open to atmosphere.

 

[ryane]

I can tell you load cells will drift, especially under a load ( even the empty kettle)

I have a DP sensor I will be placing on the output of the MLT to add redundancy to the float and monitor grain bed compaction. I've been toying with the idea of the other side of the DP being open to atmosphere or placed in the MLT, but since the levels in the MLT will change with recipe, I'm leaning toward open to atmosphere.

With the MLT being essentially open to atmosphere could you just get away with a relative pressure sensor, that way you wouldnt have to worry about the vent?

 

[trimixdiver1]

With the MLT being essentially open to atmosphere could you just get away with a relative pressure sensor, that way you wouldnt have to worry about the vent?

Or just keep the other side sealed I guess

 

[ryane]

Or just keep the other side sealed I guess

Have you thought about placement of the pressure sensor in the MLT?

I dont want to mount it to the bottom of the keggle because you would never get a CIP to work, so I was thinking of using attaching an instrument T or a wye to the bottom drain and having the pressure sensor there. That way it would drain when emptying the system, but Im worried it would be too close to a valve/pump below it. I might be over thinking this, but I generally have not had to deal with such a small system before (volumes, tank sizes, etc), thoughts?

 

[trimixdiver1]

Have you thought about placement of the pressure sensor in the MLT?



I dont want to mount it to the bottom of the keggle because you would never get a CIP to work, so I was thinking of using attaching an instrument T or a wye to the bottom drain and having the pressure sensor there. That way it would drain when emptying the system, but Im worried it would be too close to a valve/pump below it. I might be over thinking this, but I generally have not had to deal with such a small system before (volumes, tank sizes, etc), thoughts?

That's what I'm doing, I will put an extra 1.5in tri clamp tee after the kettle drain isolation valve (always open when brewing) I have an accutech DP-250 sensor. I found the placement of the tee to be say 25" so I calibrate to zero at empty and 25" when full. That level will be one shoted to a +/- SP and alert or throttle back flow if say it pulls x inh2o negative.

 

[ryane]

There is an idea, this approach would seem simple, cheap, food safe, etc. How about a sight glass with an adjustable IRLED/phototransistor that you just slide up and down the sight glass to set your level? (edit) Same thing atoughram said.

Anyway, it sounds like your going with the pressure transducers, I'm very interested in what kind of accuracy and resolution you are able to achieve. Do you have links to the products you purchased?

Sorry I didnt see notice your post sooner.

I ended up getting 4 total pressure sensors that I will be playing with, whatever I dont used on the brew rig will be used on other homebrew projects.

I picked up 2 x Anderson SL5 hydrostatic pressure transmitters and 2 x IFM PI2098 pressure transmitters. I snagged the andersons for 75$/ea (NIB - 1.5" TC) and the IFM's for 45$/ea (surplus - 2" TC)

The andersons have a minimum water column reading of 0-30" and go up to 140", the IFM's have a slightly lower minimum of ~20" (50mBar) and go up to 100" of water with a step of 0.2mBar or ~8oz in a keggle

 

[ryane]

I just added a capacitance level sensor for my BK, picked up a new lumenite for ~200$

All thats left now to get is some pipe, triclamps, and a couple RTD's!

 

[muhteeus]

I have an analog GEMS float sensor for level, the IFM gives me temperature and a .1g per pulse for the flowmeter. You can use it for a batch fill application, I have the same float setup in the MLT too, but use the IFM to help give a redundant fill level.

Do you have a link for this sensor?

 

[trimixdiver1]

Do you have a link for this sensor?

http://www.gemssensors.com/Products/Level/Continuous-Transmitters/Float/Small-Size/XM-XT-800-Series

 

[muhteeus]

http://www.gemssensors.com/Products/Level/Continuous-Transmitters/Float/Small-Size/XM-XT-800-Series

Wow those are pricy.

 

[ryane]

Wow those are pricy.

Watch ebay and you can pick them up for 100 or less

Trimixdiver - what is the resolution on the unit you have?

 

[trimixdiver1]

Watch ebay and you can pick them up for 100 or less



Trimixdiver - what is the resolution on the unit you have?

It's 12bit analog, so 4095 units scaled to 15g. I get .1 g resolution, though I could get .01 if I wanted. Just isn't as stable.

I got mine for $75 each, though they didn't come with the floats, and you need GEMS floats, I tried some SS floats off ebay and they were erratic at best, so I spent another $200 on the OEM floats and it's really stable, I also have a 1s filter in the logic to smooth out the response.

 

[BillKlineVT]

I'm working on a DIY capacitive level sensor... Proof of concept worked so building up a more robust version to try in the boil kettle. Using the "concentric pipes" design with a 1/2" copper pipe and an 8 AWG copper wire that'll be electrically isolated by 1/8" ID silicone hose. Connecting both to an arduino and using the capacitive sensor library that is already available. Also planning on correcting for temperature using my existing temp probes in the vessels.

Will try to post an update once I get it all cleaned up.

 

[thekraken]

I'm working on a DIY capacitive level sensor... Proof of concept worked so building up a more robust version to try in the boil kettle. Using the "concentric pipes" design with a 1/2" copper pipe and an 8 AWG copper wire that'll be electrically isolated by 1/8" ID silicone hose. Connecting both to an arduino and using the capacitive sensor library that is already available. Also planning on correcting for temperature using my existing temp probes in the vessels.

Will try to post an update once I get it all cleaned up.

I'm interested in what you come up with. Maybe a build thread is in order?

 

[BillKlineVT]

I'm interested in what you come up with. Maybe a build thread is in order?

Will do... I'm building the first one to make sure it's actually going to be accurate enough, then when I make the second one ill photo document the build and post something in this sub forum.

 

[cank]

I just bought 2 of these for $75 ea. Hoping I can get them to work with my Pi.
http://www.ebay.com/itm/like/301490551324?lpid=82&chn=ps

 

[trimixdiver1]

I just bought 2 of these for $75 ea. Hoping I can get them to work with my Pi.
http://www.ebay.com/itm/like/301490551324?lpid=82&chn=ps

That's the guy I get most of my stuff from. You will need the cable.

Good choice. Don't know if your controller does 4-20ma analog though.

 

[cank]

I'll probably have to make something for the pi to able to read it. I was also thinking I could make my own cable. Do you know what kind of connections these use?

 

[orangehero]

Looks like it has an M12 male connection. Measure the diameter to be sure.

 

[BillKlineVT]

So I have some good news on the DIY Capacitive Liquid Level Sensor... I got it working. Check my blog post here https://brewberrypi.wordpress.com/2015/02/13/level-sensor-working/

This was my first crack at building a sensor of this type, so I will photo document the build of the next one I make (for my HLT).

As far as how to process the raw sensor data, that's probably a whole separate thread, as you need to interpolate data based on liquid temperature.

 

[BillKlineVT]

Did some more calibration today using my DIY capacitive level sensor. I think I'm going to bail on this and go with the tried and true Brewtroller-style "Bubbler" system.

Here are some charts from my calibration of the capacitive sensor showing how variable it is with temperature. I built in a lookup table and interpolated between known points, but it still wasn't accurate enough for me (+/- 0.5 gal). I'm sure the industrial versions are much better, but wasn't looking to spend 100+ for each sensor...

 

[thekraken]

Interseting. Yeah, I don't think I would be happy with +- 1/2 gallon either. Thank you for testing it and posting your results.