Automated Rig Piping review

[PLOVE]

Something you might want to consider in your design - moving a valve (not three-way) to the outlet side of your pump, so you can feather the flow rate. If you want to keep from collapsing your grain bed during MT recirculation (especially with 1/2" all around and big grain loads), you want to be able to restrict the outlet pump flow. Restrict on the inlet side is a no-no with magnetic pumps, so I've heard. You want to run the impeller wet to keep from burning out the sleeve bearing.

 

[mattd2]

Something you might want to consider in your design - moving a valve (not three-way) to the outlet side of your pump, so you can feather the flow rate. If you want to keep from collapsing your grain bed during MT recirculation (especially with 1/2" all around and big grain loads), you want to be able to restrict the outlet pump flow. Restrict on the inlet side is a no-no with magnetic pumps, so I've heard. You want to run the impeller wet to keep from burning out the sleeve bearing.

There are manual valves after the pumps on all 3 flow paths, they are just after the auto 3-ways

 

[kickflip_mj]

One concern I am having is the manual valve that slows down the flow to the MLT might need to be adjusted between mashing and sparging. any ideas?

 

[mattd2]

One concern I am having is the manual valve that slows down the flow to the MLT might need to be adjusted between mashing and sparging. any ideas?

I'm guessing you are fly sparging. You could adjust the P2 valve to match whats coming in? If you think about it P1's valve should be set up at a good point to not result in a stuck sparge. So leaving that alone and adjusting P2's so that the water level in the MLT is stable.

 

[kickflip_mj]

Ok I figured that I could control the sparge using the volume sensing on the Brewtroller. the liquid is measured as it enters the BK

 

[marcb]

Ok I figured that I could control the sparge using the volume sensing on the Brewtroller. the liquid is measured as it enters the BK

Are you using a liquid/pressure sensor? I found a reasonably priced flow meter after checking out a brewmation system that uses pulse 5v dc output. I figured that flow volume was a more accurate way to measure and as a plus I'm going to make it tri clamp and portable for measuring flow rate into mash tun, hlt and boil kettle.

 

[kickflip_mj]

Can I see what you found?

 

[mattd2]

Also can you explain how you are taking the pulse output and converting it to a flowrate? I have seen a few examples on the Arduino pages but all of them seem to be poling (is that the right term) the digital pin every xxx ms and checking for a count. Seems like you could miss some data that way.
Edit: Ok I think I was reading that wrong - they seem to be polling the pin for 1 ms and counting the number of pulses - that makes much more sense

 

[marcb]

Can I see what you found?

Here you go. In the brewmation setup it feeds into the timer which also acts as a counter.

 

[marcb]

Also can you explain how you are taking the pulse output and converting it to a flowrate? I have seen a few examples on the Arduino pages but all of them seem to be poling (is that the right term) the digital pin every xxx ms and checking for a count. Seems like you could miss some data that way.

Heh. Not using it yet, could feed it into a timer unit as a flow counter as the brewmation guys do. I wanted to feed it into my BCS but it won't take pulsed input. I may just build a raspberry Pi or sanguino board and write some code to handle the input as I would like to measure flow to open and close valves at the start of the brew process.

 

[kickflip_mj]

You think that is practical over the brewtroller volume measurement?

 

[marcb]

You think that is practical over the brewtroller volume measurement?

Practical doesn't usually enter the equation for me : )

 

[kickflip_mj]

Haha that's the motto I live by. I just think what you plan on doing is going to be suck a pain in the A**!

 

[marcb]

Haha that's the motto I live by. I just think what you plan on doing is going to be suck a pain in the A**!

Yep, the problem that I am looking to solve is to have a portable sensor that I can plug into anywhere in the brew process where I want liquid volume measurement in gallons. Probably makes the most sense to just use the timer/counter approach.

 

[mattd2]

For connecting up to a Brewtroller it shouldn't be too bad... should it? Just trying to remember the two plateforms (Brewtroller & BCS), but IIRC the brewtroller is open sourced and base of the arduino development platform, there are plenty of examples/code to use the voltage pulse/hall effect flow sensors in the Arduino community. If I ever get this far into automation I probably would be looking to start fresh with a micro (probably the Arduino, at least at first) and build my system around what I need, not what a vendor offers. Saying that since the brewtroller is open sourced I would likely be taking a few leaves out of their book - of coarse any new developments would be feed back inot the community

 

[marcb]

For connecting up to a Brewtroller it shouldn't be too bad... should it? Just trying to remember the two plateforms (Brewtroller & BCS), but IIRC the brewtroller is open sourced and base of the arduino development platform, there are plenty of examples/code to use the voltage pulse/hall effect flow sensors in the Arduino community. If I ever get this far into automation I probably would be looking to start fresh with a micro (probably the Arduino, at least at first) and build my system around what I need, not what a vendor offers. Saying that since the brewtroller is open sourced I would likely be taking a few leaves out of their book - of coarse any new developments would be feed back inot the community

I poked around a bit and couldn't find any prior art for the BT or other platforms for pulse 5VDC input for flow calculations so I assumed I'd have to write it myself. I like the Arduino/Sanguino boards for I/O and I like Rasberry Pi for more complete linux development environment.... it's all software so it's not a big deal and should be reusable by others after I'm done... unfortunately it's one of a gazillion projects so I'm not sure when it will become the highest order bit.

 

[mattd2]

Ok I figured that I could control the sparge using the volume sensing on the Brewtroller. the liquid is measured as it enters the BK

AFAIK the "cheap" auto valves have no feedback right? (although one wiring config I saw had switches for fully open/closed) but that might not matter.

So to control flowrate is your plan to fix the MLT valve position and vary the HLT valve to match?

Flowrate or level, either should work in this circumstance assuming a straight sided pot.

So you need to compare the 2 vessels emptying/filling. and adjust the HLT valve so that they match - thats the easy part right.

The hard part is controlling the HLT valve, made even harder by the fact there is no feedback so you don't "know" where the valve is - although that might not be an issue.
depending on the wiring of the valve will depend on how you control. Not really my realm of expertise but I'll throw ideas out to build on.

Force open/close (easier)- when an error signal between the rate changes is detected, the valve is switched for error*X seconds.

Force open / fail closed - could you set up a PWM at a high(?) frequency. Base output of this is 50%, this should keep the valve in the "last" position - i.e. it is opening/closing for half the cycle assuming the open/close rate are the same. Then based on the error adjust the PWM over/under 50% - output = 50% + 50%*error*proportional constant.

With error being (HLT change - BK change) / BK change -> making sure your +/- are the right way round for how you are measuring it.
As I said no even an entry level systems engineer but something to start throwing around

Also where are you getting and how much are the triclover auto valves?

 

[mattd2]

I poked around a bit and couldn't find any prior art for the BT or other platforms for pulse 5VDC input for flow calculations so I assumed I'd have to write it myself. I like the Arduino/Sanguino boards for I/O and I like Rasberry Pi for more complete linux development environment.... it's all software so it's not a big deal and should be reusable by others after I'm done... unfortunately it's one of a gazillion projects so I'm not sure when it will become the highest order bit.

I found some stuff yesterday when I was digging around - I'll post links when I get back on my laptop.
Rasberry Pi looks very interesting but way too complex for me to deal with, arduino and C++ will test me enough I am sure.
Can the Pi be setup to run just as a linux (ubuntu?) "computer" that you just need to plug a usb keyboard/mouse into and a HDMI screen (any semi modern TV)? I know it would be a waste of the platform but would make a nice second computer for my household.

 

[marcb]

AFAIK the "cheap" auto valves have no feedback right? (although one wiring config I saw had switches for fully open/closed) but that might not matter.

So to control flowrate is your plan to fix the MLT valve position and vary the HLT valve to match?

Flowrate or level, either should work in this circumstance assuming a straight sided pot.

So you need to compare the 2 vessels emptying/filling. and adjust the HLT valve so that they match - thats the easy part right.

The hard part is controlling the HLT valve, made even harder by the fact there is no feedback so you don't "know" where the valve is - although that might not be an issue.
depending on the wiring of the valve will depend on how you control. Not really my realm of expertise but I'll throw ideas out to build on.

Force open/close (easier)- when an error signal between the rate changes is detected, the valve is switched for error*X seconds.

Force open / fail closed - could you set up a PWM at a high(?) frequency. Base output of this is 50%, this should keep the valve in the "last" position - i.e. it is opening/closing for half the cycle assuming the open/close rate are the same. Then based on the error adjust the PWM over/under 50% - output = 50% + 50%*error*proportional constant.

With error being (HLT change - BK change) / BK change -> making sure your +/- are the right way round for how you are measuring it.
As I said no even an entry level systems engineer but something to start throwing around

Also where are you getting and how much are the triclover auto valves?

I haven't found any reasonable electric sanitary valves yet, my ideal would be 3 way electric 1" valves with 1.5" tri clamp fittings. I've got feelers on pneumatic valves and will then have to incorporate a series of gas solenoids to actuate.

For flow I'm looking for volume in for mash and out for boil kettle to tighten my consistency on brew day. In my setup I will be sparging with a tankless water heater and will use a float switch on my sparge assembly to keep the grain bed at the right level for sparging and manually control the liquid in valve on the brew kettle for sparge. Depending on tankless performance and lag time I may need an auto valve with temp control to ensure that the sparge water is at 180. Essentially the valve would divert until a temp probe hit 180 and then it would run into mash tun for sparge. I may be over complicating it though, I don't know how much temp loss in the lines I'll incur.

 

[marcb]

I found some stuff yesterday when I was digging around - I'll post links when I get back on my laptop.
Rasberry Pi looks very interesting but way too complex for me to deal with, arduino and C++ will test me enough I am sure.
Can the Pi be setup to run just as a linux (ubuntu?) "computer" that you just need to plug a usb keyboard/mouse into and a HDMI screen (any semi modern TV)? I know it would be a waste of the platform but would make a nice second computer for my household.

Definitely no video or hdmi out, most of these are serial console boards but may have VGA output. USB for keyboard and mouse should be there.

 

[mattd2]

Definitely no video or hdmi out, most of these are serial console boards but may have VGA output. USB for keyboard and mouse should be there.

Huh? Arn't you talking about this? http://www.raspberrypi.org/faqs
Both RCA video and HDMI on there? But no VGA.
Sorry for the off topic kickflip

 

[marcb]

Huh? Arn't you talking about this? http://www.raspberrypi.org/faqs
Both RCA video and HDMI on there? But no VGA.
Sorry for the off topic kickflip

Yep, dang - I had the specs mixed up with another board. The model b specs are fantastic, most importantly for me is the onboard Ethernet and new header pins as I will use these headless.... Although piping an xterm on hdmi to an external touch screen is introguing.

 

[mattd2]

Yep, dang - I had the specs mixed up with another board. The model b specs are fantastic, most importantly for me is the onboard Ethernet and new header pins as I will use these headless.... Although piping an xterm on hdmi to an external touch screen is introguing.

Yep... you lost me But it does look like such a neat concept that you can have a basic computer with you and all you need is basic gear that you can find in any house really. I sure you could do some amazing things with it too...
I want one just to have one... but the Arduino mega board @ $20 from dealextream is just more inline with my needs today

 

[kickflip_mj]

Ok you guys lost me.lol

Mattd2,
For the valves I was planning on taking 1/2 valves https://www.oscsys.com/store/product/291 drill the threads out to 7/8, cut it down and weld a 1 inch ti clover to it.

 

[mattd2]

Ok you guys lost me.lol

Mattd2,
For the valves I was planning on taking 1/2 valves https://www.oscsys.com/store/product/291 drill the threads out to 7/8, cut it down and weld a 1 inch ti clover to it.

Sweet as, yeah that was what I was sort of thinking of doing to... in a few years time! Just wanted to make sure there wasn't some secret stash hidden away somewhere on the internet
Also have a look here TF valves - http://www.tf-valvefittings.com/goods-13.html
No maual overide on that one though, the same product range the manual overide valve is about $43 so not much savings to justify. But they have another brand with manual overide for $23. Really I would be looking at whether you need the overide as you could be saving over $150 on your valve order
I think they are the same valves, but cheaper direct from the chinesse vendor. I just noticed that it actually references brewtroller on the page!
You can also choose the wiring type.
Lots of different valves on that site though.

 

[mattd2]

I found some stuff yesterday when I was digging around - I'll post links when I get back on my laptop...

Link for the methods of counting pulses - http://www.gammon.com.au/forum/?id=11504
I only skimmed through as too much inforamtion for my brain to retain until I need it!

 

[kickflip_mj]

Sorry I havnt wrote back in a while, I'm in Germany doing some work (and drinking the best beer in the world) and hopefully will have ths project in full swing when I get back this weekend.

 

[marcb]

Sorry I havnt wrote back in a while, I'm in Germany doing some work (and drinking the best beer in the world) and hopefully will have ths project in full swing when I get back this weekend.

Getting ready for my new build too so get your game on.....

 

[barryfine]

I apologize if this has been answered already, but what software are you using to make these great schematics?

 

[mattd2]

Kickflip / marcb - You guys have put the fire under me too. I am now looking at Arduino megas & TFT LCD screens to start building a better control system than my Sestos PID.
Current plan is:
Retrofit the new control system to my existing rig - Temperature monitoring/control only.
Build and implement flow monitoring
Sell existing rig (with old control panel)
Build nice new shiney SS rig with actuated valves/flow control/fancy graphical display/bells/whistle
...
enter dog box