Phase 2 brewery build progress shots

[kladue]

The chimmney tube design is a rough copy of the coil system in the larger boil kettles, the thermosiphon effect should give enough mixing to make it work. It will take quite a bit of steam to start but maintaining a boil should not be hard after wort temperature is near boiling. The 1/4" size was chosen to keep steam velocity up at low flow when steam is throttled during boil, backfeeding through drain might work but the need for a small surface area diffuser would make the large area drain inlet screen impractical for steam injection.

 

[WortMonger]

Ok, so disheartening, but expected information on using the kettle drain. Still, very good information on the ability of the kettle to maintain a boil. It is also good that thermosiphon turbulence would be high enough to make a good boil.

Would you suggest a insulated keggle and directly inputting steam through the 1/4" J, or would you leave the kettle naked? I ask with the provision that the naked kettle would have my steam boiler under it, helping to heat with the heat that doesn't turn the water in the flash boiler coil into steam entering the kettle on the inside? This latent heat may be negligible and so would definitely point me strongly in one direction over the other.

 

[kladue]

The lower the heat loss fom the sides the lower the steam demand for maintaining a boil, if you use the flash boiler and superheat the steam to 250+ degrees you would reduce the total steam requirements. Superheated steam carries more energy per pound than saturated steam from a pressure cooker, which means fewer pounds to get job done. If you insulate use the KAO wool ceramic fiber insulation as it has some impressive thermal properties and is easy to work with, and virtualy flame proof. I have taken a 6" X 6" X 1" square and have held it in my hand while heating opposite side to orange heat with a torch.

 

[WortMonger]

Wow, the kaowool sounds wonderful! Thanks for the heads up on that stuff. I may have to use that on my flash boiler outside as well.

Since the keggle will stay around boiling temperatures and not get much hotter, should I even worry about such expensive insulation? Would the way I made my HLT and MT work (with 3 layers of reflectix and then truck bed lining them)? I still want to track down some kaowool now for the flash boiler. :rockin:

 

[kladue]

Not sure what the kao wool costs as I got plenty for free, why not just use a 1/2" air gap and a second tube as a heat sheild on the boiler, then use the kao wool insulation on the boil vessel. For the Mash tun use the canned foam, spray it on and trim it to 1" thick, then put a wrap over the foam for protection. PM me if you would like to try some of the kao wool insulation on your boil kettle.

 

[WortMonger]

Sounds like a plan kladue. I have been checking out kao wool and it is impressive stuff. I have another concern that just came to mind. The quick disconnects that I use now for everything hot-side are rated to 280*F. Should I be looking for something a lot higher temperature for the flash boiler connections? Also, if the flash boiler was tap water fed, and had a flow meter, would it be alright to use and just take a time/gph measurements?

 

[kladue]

Although the swagelok compression fittings are not as convienient as quick disconnects the temperature rating would not be a concern on the hot side of the boiler. With new boiler I have seen temperatures 350-450 degrees during steam injection loop tuning, water flow adjustment reduced it to 220 degrees as expected. As to time/flow method, would work if you regulated water pressure into flow control as changes in water pressure will change flow rate. Look for a plastic preset regulator for drip irrigation ~$10 plumbed in ahead of flow meter on cold water inlet to boiler. Select a pressure setting that is no more than 50% available water pressure to prevent surprises from high demand uses in water system and allow enough pressure difference across regulator to hit peak flow through regulating valve. Might be able to use a cheap digital timer and solenoid valve to come fairly close to desired volume by calculated time/flow , although a bit under/over is not fatal in brewing. Lower the flow rate the smaller the error would be but if you are shooting for .5 - 1 GPM range then a couple trial runs with cold water will help dial in time/flow calcs.

 

[WortMonger]

Great news then!

 

[kladue]

Finally have taken the time to finish the control system schematics Picasa Web Albums - Kevin - Schematics, I am still working on the graphic portion of the control application and hope to test it on the hardware this weekend.

 

[jcdillin]

Very very impressive, the code behind all that must be crazy.

 

[kladue]

The software portion of the build is taking way more time than the construction did, but having nothing else to do nights and weekends it keeps me busy. The hardware and control sequence code is complete and functioning, still working on converting some of the older components to touch screen operation, database functions use mySql.
Basic structure is:
Hardware communication and initialzation with point data stored in database
Touch screens for analog point scaling and offsets for calibration, default flow settings for system, values stored in database
Touch screens for PID loop tuning, values are stored and retieved from database for use in automatic control sequence
Touch screen manual control for hardware testing
Touch screen input screens for batch setup, mashing, and hop addition parameters, values stored in database
Automatic calculation of water volumes and temperatures from batch information
Automatic control sequence with restart from last control function and elapsed time on power failure, values are stored and retrieved from database
Touch screen monitoring screens for process observation
Todo list:
Add graphing screen to system monitoring screens
Build touch screen import utility for use with other brewing software packages to import control schedule information into control sequence program

 

[IrregularPulse]

WOW! Planning on selling this to Budweiser? This is a crazy homebrew setup!

 

[Arkador]

The time i get to planning something like this is when SWMBO would pull the plug on my homebrewing and tell me "You are out of control!"

 

[Ol' Grog]

Sweet Mother Mary!!! That is one of the most beautiful things I've ever seen. If I had one of those, I'd be divorced, for sure. What has this cost so far?

 

[kladue]

I guess I am lucky to have a SWMBO who brews and enters competitions with me and indulges my out of control tinkering with this R&D unit. So far the building was $3600, utilities for building $700, system hardware $2000+ (Never added it up), too many manhours to estimate in assembling components, welding, plumbing, wiring, software construction, and testing of assembled system. Still have to insulate and finish interior of brewing building, and chiller/heater bands for the carboy fermenters. Software is still under development for use on a touch screen panel computer to be mounted on panel #2, and I have not decided on final functionality limits yet, always one more "what if" to do.

 

[Ol' Grog]

From what I've read so far, your going for a push button brew set up. Put all the ingredients in hoppers, think up a formula, punch them into the main computer and then the rest is done for you. That's just toooooo cool. This way, you could experiment with brews until the day you die. Could life get any sweeter for you??

 

[beerjunky828]

Very, very nice. That is even an understatement. Thank you for the post!!

 

[kladue]

Here is some of the control graphics for operating the phase 2 system, everything is built for touch screen operation Picasa Web Albums - Kevin - Phase 2 Brewe..., the graphics did not convert to JPG format very well but are somewhat legible. The graphics and supporting code were assembled using Netbeans and Java( a PITA at times), with MySQL taking care of the database fuctions recording the point and configuration data for reuse and trending. I am almost done with the transition to touch screen control, next up is the file import code to upload the batch control values from recipe programs to save the effort of manual entry.

 

[Rhoobarb]

This set up is just mind-boggling!:cross: Thanks for posting this and keep things updated, please. I'd like to follow along.

 

[kladue]

The panel PC is installed in the second panel now, going to try for a test run this weekend if possible.

 

[Araxi]

WOW, that's one cool setup. Now all you need to do is hire a few oompa loompa's.

 

[kladue]

Panel PC test was a bust, got the MySql going, Java 5.0 JRE installed, loaded brew system program, then found out class path method changed between runtime versions. Now it appears that I need to upgrade to Win XP and 256M memory in the Advantech panel Pc to run 6.0 JRE, it was either that or rework about 2.8K lines of code to resolve classpath issue. Next test window will be during July 4 R&R trip home.

 

[kladue]

Here is a couple shots of the panel pc and other panels for the brew system Picasa Web Albums - Kevin - Panels, touch screen PC needs operating system and memory upgrade to run brewing software under development for automatic operation.

 

[android]

damn... that is one sweet set up... you better get some brews crankin' out soon!

 

[kladue]

My son finally completed all the upgrades to panel PC and loaded all the necessary software, killed a couple last minute bugs and has the autocontrol software running on the panel PC and controlling the hardware now. He went through the manual control screen and confirmed the operation of the ignition,valves, and pumps. Then the input points were checked and are operating but some need scale values entered for levels and flows. Next up is loop tuning and calibration then first full automatic sequence run with malt this weekend during my next R&R run home.

 

[beerthirty]

Good luck with the test run this weekend. I will be rewiring the CB with cat7 SSTP including the LCD ribbon to knock out crosstalk and EMI. The cat5 UTP only helped a little bit. I wish I was just down to software issues

 

[Simphoto02]

I am not sure this is what Jimmy Carter had in mind when he legalized Home Brewing for his brother Billy!!!!!!!!

Holly crap dude, a Freakin Grain Elevator?

I am kneeling before that altar of brewing! (you can't see me but trust me I am)

Teach me oh wonderful wizard

 

[ek1050]

Once your automated brewery is complete, you'll have to work up an automated label maker that comes up with clever names and designs all by itself
Nice gear.

 

[kladue]

Sounds like one of the future projects, modifying one side of a dual sink to make a carboy and bottle washing system in one side that will run through cycle like automatic dishwasher and clean the glassware. Should not take that long to design and build, a circulation pump, a couple automatic valves, inline heating chamber, some liquid metering pumps for chemicals, capacitive level sensor, temperature sensor, some opto plc hardware, a day to build graphics, a weekend to write and test the code, a bit of wiring and plumbing. Not sure what to do after all the construction is over, break down and make beer again and torment the club members in the homebrew competitions again.

 

[kladue]

Test run on the panel pc was an educational event, the 500Mhz processor was bogged down when running the compiled application so i switched back to the laptop and CPU time was considerably more than when it was running in the Netbeans IDE. Looks like there needs to be more thread.sleep statements to lighten the load before next try at panel operation, and a few tweaks to restore sound effects on alarm conditions.
During PID tuning it became obvious that a valve dead angle (level of output that needs to be exceeded before flow begins) variable needs to be part of the hardware setup tables. This offset value will be applied to PID loop calculation so control loop will function better when it is trying to operate in the open range of the control valve instead of entire control range. Another PID loop problem was need for a multiplier value for loops where signal range is .13 - 1.3 GPM and control setpoint is .15 GPM, control range is so small and signal change is so small that PID values begin to look absurd in order to make loop function as needed.