Automating A Brewery

[Wortgames]

Its all compatible with LabView software which I believe you can integrate into visual studio.

I think for $1200 it would want to clean up after the brew and fetch my slippers too


I love the idea of integrating some gadgets but I think 'planning' might be a bit extreme.

I guess ultimately it would be cool to have 6 or so temp probes, a pump, a few valves, an element or two and maybe a couple of motorised gadgets.

The first stage as I see it though is to get a couple of probes and use the computer to simply monitor temps at a couple of points and manually switch a couple of devices, then work towards automating those bits and expanding on it based on my programming proficiency.

I've downloaded VBExpress and can't make head or tail of it yet, but I suspect it will be a handy tool to become proficient with and I have some experience with BASIC and MS Access & Excel which might help to get me off the ground.

I also downloaded the trial version of ProfiLab Expert, and it looks fairly straightforward once you get to grips with it, but the ProfiLab / LabJack combination would obviously place all the load on the computer which might be a big ask.

Using a PIC / PLC / wotsit might help to take some of the strain off an old computer.

 

[chimera]

$1200? aah, you hadnt yet found VS Express versions yet.

To be honest, Im not much for VS myself, but most these sorts of DIO / PLC products either have their own propietry interface, or VS APIs (or both).

Its a learning curve to be sure, but think about the beer!

 

[ausdb]

you might want to have a look at the PLC's available from Onetemp (web site here).

Prices start @ $145 + gst for a 10 i/o unit The configuration software is windows based & is FREE! You can download the manual & softeare off the above web site to check out how tt use it.

One thing to watch for with the cheap PLC's is the input ytpes, if you are trying to monitor a variable like mash temps or hot liquor temps then you will need one with analogue inputs. I have been rolling this idea around in my head too and are not sure which way to go. Soldering things together is no problems but programming code is not my forte either

My ultimate dream a fully automated all in one system that automatically lifts the bucket out of the urn, boils, doe automated hop additions, then chills and fills the fermenter :excl:

 

[Wortgames]

I was actually referring to LabView, which seems to start at $1200?

http://www.ni.com/labview/

A bit rich for my blood!


So far, the software choice looks like being VB Express (free, but a steep learning curve) or ProfiLab Expert ($165, but probably easier to program).

Hardware-wise it's looking like a 'PLC' / 'PIC' / 'whatever the hell the others are called' is virtually essential (any advice on choosing between them?)

It would be interesting to know how well the LabJack option works - but my gut feeling is that it would deliver too much computational work back to the computer, as it appears to be a totally passive connection box. I'm sure this would be no problem on a newer computer but I'm begining to think it might be too much traffic for an old P166MMX running 98.

With a PIC or similar, mundane tasks such as temperature control could be performed independently without having to bother the computer, right? Then just use the computer to interject as necessary?

Can you interact with all of the devices (PLC / PICAXE / ATMega32 / any others) in real time from a PC?

Are there pros / cons to each type of device?

 

[sluggerdog]

sort of on topic, I had thought about something similar myself but start with the basics.

I was thinking a simple tap timer like: http://cgi.ebay.com.au/Gardena-Electronic-...1QQcmdZViewItem

to start and stop the hlt going into the mash tun however it would need to be ran per litre rather then time, does anyone know if there is such a thing available?

 

[chimera]

the PLCs do look rather good.

When (I say when, but I have no plans yet) I go down the path of automation i'd want a system that didnt require a PC link, and these PLCs appear to have this potential.

 

[DJR]

The NI boards if i remember correctly have a free development driver that can be used in VB, C++ etc called NI-DAQ. Hard to find on their site because they want to sell Labview but the NI-DAQ suite will work fine. Might just want to email them about the output capability though, we only ever used it for data acquisition purposes on an input-only board. Labview AFAIK is more of a program creation tool in itself.

 

[DJR]

sort of on topic, I had thought about something similar myself but start with the basics.

I was thinking a simple tap timer like: http://cgi.ebay.com.au/Gardena-Electronic-...1QQcmdZViewItem

to start and stop the hlt going into the mash tun however it would need to be ran per litre rather then time, does anyone know if there is such a thing available?

A float switch to set the water level would be good, so that when the water level hits a certain amount the flow from the HLT stops, but then you'd have to use a actuated valve of some sort, Geordi wants $300 each for a actuated ball valve so i wonder if there's something more cost effective about?

Edit: something like this for a float switch: http://www.electusdistribution.com.au/prod...D&SUBCATID=

Another edit: Found this 1/2 BSP solenoid, rated to 90C with a 40% duty cycle, should be OK for most applications, for $30 who cares:
http://www.rsaustralia.com/cgi-bin/bv/rsww...heID=aunetscape

 

[Wortgames]

I had another look at crozdog's link, they look like a pretty good deal in comparison to the others: http://www.onetemp.com.au/PLC's.htm

$260 for a 12-input 8-output device, and free QuickII software (which I have downloaded and will play with later). The only other expense looks like a $35 PC link.

I have a couple of temperature sensors laying around that might do to get started with, I already have a pump and an element in the HLT, so it looks like I could begin automating for around $300.

That PLC is probably overkill but it would be good to have the options for future expansion. I'd like to end up with a couple of temp probes in each vessel, to get a visual idea of temp stratification, as well as sensing incoming water temps etc. Add a couple of water level sensors and I'll fill up 12 slots quite quickly I reckon!

Does this look like a good plan?

It may even be that I can use VBExpress at some point in the future to improve the graphical interface, it doesn't look like QuickII has much scope in this department.

 

[chimera]

Wortgames - had a play with that Quick II program. For the price, if those PLCs could be used to automate It'd be a top solution.

I cant say much about how to make them work with analog inputs, my experience of PLCs has been all digital I/O.

 

[Wortgames]

They appear to be the same device that Ocean Controls are selling: http://www.oceancontrols.com.au/plc/fab_plc.htm (thanks JasonY for the link) but OneTemp are slightly cheaper (it looks like Ocean Controls include the LCD panel as standard, but their prices are still a bit higher regardless).

There are a couple of different models available - some are digital only, and others have selectable digital / analog inputs (which is what I'd be getting I guess).



There are also some models there listed as "AC input" - I don't know what this means, maybe analog only? Anyone?

 

[Wortgames]

Hmm, could be back to the drawing board - I just spoke to a very helpful guy at OneTemp, and while he wasn't sure, he felt pretty confident that the PLCs don't offer real-time interaction with a PC. You can only use the PC to create your program, then you write it to the PLC. You can't tinker with anything during a running process and you can't read temps via the PC.

I also spoke to someone at Ocean Controls, who wasn't much more knowledgable, but he seemed pretty certain that you CAN monitor them in real time (ie, view temps etc) and that you can even adjust analog values (eg setpoints), but he didn't know if you could actually interact with them in terms of 'pausing' the program or manually overriding things. However we both decided that this might be possible by sacrificing one of the inputs as an 'override' switch, and getting the PLC to check the switch before performing each task.

This is going to be hard if even the guys selling these things aren't sure what they can and can't do!

It's looking as if something like the LabJack might be back in pole position, and finding (or making) the right software is the secret to this.

 

[ArnieW]

Another edit: Found this 1/2 BSP solenoid, rated to 90C with a 40% duty cycle, should be OK for most applications, for $30 who cares:
http://www.rsaustralia.com/cgi-bin/bv/rsww...heID=aunetscape

Hi DJR,

I tried to follow that RS link but didn't end up at a solenoid. But for $30 I'd be very surprised if it were suitable for brewing. Most cheap (ie. garden water system types) solenoids work with back pressure (like mains water pressure) and only have a pin hole diaphram - which means that the smallest bit of grain will clog it. This is why the watering system solenoids have an inbuilt filter.

A solenoid for brewing is more likely to be some kind of motorised ball valve, and these do not come cheap. I've been playing around with automating ball valves for about 4 years, and am close to a solution that I like. ONe of the issues is that as the taps get hot, they expand and get tigher, requiring more grunt to operate.



I've added a pic of my latest experiment. Basically it is a grunty geared down DC motor that drives a screw thread that then pushes or pulls the handle on the ball valve. There is a variable resistor mounted on teh top that will ultimately give the control part feedback on valve position, a bit like a servo motor.

cheers, Arnie

 

[DJR]

Most PLC's are pretty much set and forget, however you could get them to have a function that reports on the status of the various inputs, but i've only seen this for one particular PLC (a Koyo DL05, not sure if it could be done on others). If you're going to have a PC involved anyway, you might as well go the Labjack, get the drivers and write up a little VB Express program for it. You could also get a copy of VB6 from somebody who has an old copy as that's probably enough. But it will require a bit of coding to get it all going.

 

[DJR]

Hi DJR,

I tried to follow that RS link but didn't end up at a solenoid. But for $30 I'd be very surprised if it were suitable for brewing. Most cheap (ie. garden water system types) solenoids work with back pressure (like mains water pressure) and only have a pin hole diaphram - which means that the smallest bit of grain will clog it. This is why the watering system solenoids have an inbuilt filter.

A solenoid for brewing is more likely to be some kind of motorised ball valve, and these do not come cheap. I've been playing around with automating ball valves for about 4 years, and am close to a solution that I like. ONe of the issues is that as the taps get hot, they expand and get tigher, requiring more grunt to operate.

View attachment 9324

I've added a pic of my latest experiment. Basically it is a grunty geared down DC motor that drives a screw thread that then pushes or pulls the handle on the ball valve. There is a variable resistor mounted on teh top that will ultimately give the control part feedback on valve position, a bit like a servo motor.

cheers, Arnie

Nice work Arnie - the solenoid valve i put a link to (which didn't work, harumph) was just a $30 nylon solenoid that would be fine for controlling the output of a HLT - but as you say, when grain is involved for a RIMS/HERMS setup then you need a ball valve. The Geordi 3 pc ones with actuator go for $300 so i think if i was going to do it i'd just buy 2 or 3 of them - but for cheap HERMS/RIMS setups then you'd need something like what you've put together.

Edit: here's a link that works this time http://www.rsaustralia.com/cgi-bin/bv/rsww...amp;Ntt=342-023

 

[ArnieW]

Hardware-wise it's looking like a 'PLC' / 'PIC' / 'whatever the hell the others are called' is virtually essential (any advice on choosing between them?)

It would be interesting to know how well the LabJack option works - but my gut feeling is that it would deliver too much computational work back to the computer, as it appears to be a totally passive connection box. I'm sure this would be no problem on a newer computer but I'm begining to think it might be too much traffic for an old P166MMX running 98.

With a PIC or similar, mundane tasks such as temperature control could be performed independently without having to bother the computer, right? Then just use the computer to interject as necessary?

Can you interact with all of the devices (PLC / PICAXE / ATMega32 / any others) in real time from a PC?

Are there pros / cons to each type of device?

Hi Wortgames,

A PLC will be easier to start with, but may not be as flexible as you would like. A PIC is a microcontroller in a chip, so you can just about do any automated task with the right model. Some PICs are really difficult to program, others are much simpler. The reason the PICAXE is so popular is that the price is right and they are really simple to use because they are targetted at school classrooms.

With a PLC you usually get some basic interfacing/connections. You will still need to build your own to automate your rig. With a PIC you will need to build everything from the ground up. This is what I'm currently doing (actually making printed circuit boards today), so maybe we can talk a bit more about what is possible off-line.

You can certainly interact in real time with PIC based devices with a computer. Check out Hippy's Picaxe Q & A for more details on how you can do lots of stuff with a picaxe, including serial comms which means comms with a PC.

I can give you code examples in visual basic based on my own controller too.

And your 166Mhz Win 98 laptop will do surprisingly well. You don't need to read a temp every second. If you think about how slowly (compared to a computer processor) things change during a mash, you will realise that you actually don't need much processing grunt to drive a brewing machine. My first machine was on a Win 3.1 laptop (admittedly running VB3) and no problems with speed.

My backup laptop I think runs at 166MHz and was coded in VB6, again no performance issues.

Keep dreaming wortgames,

cheers, Arnie

 

[bigfridge]

Hmm, could be back to the drawing board - I just spoke to a very helpful guy at OneTemp, and while he wasn't sure, he felt pretty confident that the PLCs don't offer real-time interaction with a PC. You can only use the PC to create your program, then you write it to the PLC. You can't tinker with anything during a running process and you can't read temps via the PC.

I also spoke to someone at Ocean Controls, who wasn't much more knowledgable, but he seemed pretty certain that you CAN monitor them in real time (ie, view temps etc) and that you can even adjust analog values (eg setpoints), but he didn't know if you could actually interact with them in terms of 'pausing' the program or manually overriding things. However we both decided that this might be possible by sacrificing one of the inputs as an 'override' switch, and getting the PLC to check the switch before performing each task.

WortGames,

NNL Beer Supplies has a unit under development that is based on a standalone microprocessor with a built in web server. This lets you point your PC at it to read or change any values using just a web browser, but the unit runs indepandantly just like any other temperature conroller. The unit that we are using has heaps of digital and analogue I/O's so you can read and control temperatures, fluid levels, relays solenoids etc. It has a built in RTC (Real Time CLock) so that it can handle having your HLT up to temperature ready for you to start in the morning.

We will have the first one running in our brewery by the end of the year and then they will be available to buy for between $500 - $1000 for the basic model.

My aim is to make the brewery run just like a washing machine - stepping through the various 'cycles' and riging a bell when the wort is ready to pitch the yeast.

HTH,
Dave

 

[Wortgames]

Good on you Arnie, I have a feeling you are going to be my ticket to success here :beer:

The only thing that puts me off the picaxe approach is that it is starting a bit too far down the ladder for my liking - there already aren't enough hours in the day for me to do what I need to do anyway, so starting from scratch with an IC, a soldering iron and a book on electronics theory just isn't going to happen.

The PLC is probably a great idea if you can see yourself getting to a point where the whole thing is fully automated - but I think that will be out of my reach for a while, so I need a more hands-on, real-time solution.

I'm heartened to hear you say the laptop will be up to the task - at least that fits with what I originally had in mind, a computer program doing all the work, where I can click on things to make them happen and automate a few routines.

Do you think the LabJack is the best option for an interface or are there others?

I don't have USB on the laptop but I can probably get hold of a USB PCMCIA card.

 

[Wortgames]

My aim is to make the brewery run just like a washing machine - stepping through the various 'cycles' and riging a bell when the wort is ready to pitch the yeast.

An excellent analogy!

I suspect it will take me a long time before I'm at that stage though - I'm happy to turn the odd valve myself but if I can automate a few procedures (eg the mash) and monitor a few others (the boil, the chill) from the comfort of an armchair with a beer in one hand and a mouse in the other, then I'll be a very smug little brewer indeed...

:super:

 

[ArnieW]

Do you think the LabJack is the best option for an interface or are there others?

I don't have USB on the laptop but I can probably get hold of a USB PCMCIA card.

I don't really know, to be honest. It depends a bit on how much work it is getting access to real time information off the labjack via VB.

I am close to testing a Tun controller if that helps. This is what I'm currently building circuits for.

It is picaxe based with serial comms to PC. With the PC connected you have PID control of a 20A solid state relay. If no PC is connected, you can still vary temperature setpoint and power %. The unit includes level sensors to make sure you can't turn a heater on when a tun is dry.

As you say though, there are too many things to do. I'm on leave at the moment and unlikely to get the boards done before I'm back at work, but I am getting closer. The Picaxe coding is just about done, and the PC end of it should not be too difficult.

The only real problem now is that I've already had a glass of Helles and tomorrow is grand final day ...

cheers, Arnie :beerbang: