BruControl: Brewery control & automation software

wps-hbt
Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum

Help Support Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Hi @TenaCJed,

I would need to check with @augiedoggy and the manufacturer to see if the motor in the pump can be successfully mitigated via PWM (using the MOSFET). Let me see what I can find out...

Regarding the connections... I recommend using a screw shield (see the Order List on the website). That allows easy connection to the interface via screw terminals. These typically have a small prototyping area for you to add some components. So in the example of using 1-wire temperature sensors, as long as you are capable of soldering, it would be really easy to add a resistor. In the case where this is not an option, just add the resistor into the screw terminals in parallel with the VCC and Data lines. You could use a breadboard but I don't personally believe pins inserted into the breadboard and/or interface will have the durability that an industrial machine should. Many do it successfully so I don't want to say its not possible, but I personally like the security of the screw terminals.

In the case of thermistors and RTD's, yes some additional circuitry is needed to wire in a voltage divider & filter circuit for thermistors and RTD amplifiers for the RTD sensors. I have provided schematics for these on the website. But, as you noted, I have also prototyped some boards which I can provide if you would like. I haven't made them in quantity (yet?), but I do have some available at reasonable cost. Again, see the Order List for pictures/details.

The choice of probe is up to you. The 1-wire sensors are certainly easy and provide good (not perfect) accuracy. Much of the choice will come down to what you already have or what probes will physically work for your systems' ports etc. The 1-wire are a bit more limited due to the sensor's size inside the probe, so these tend to be short and stubby compared to RTD's and thermistors. The flowmeter is very easy to integrate. I included a schematic on the website.
 
Interestingly, the motors for similar models say 17V, but this one is 24. As I mentioned, @augiedoggy noted there is a regulator module onboard to reduce the 24V to 17V. The note for this model online states "DC brushless motor with energy efficiency technology use MPPT by micro processor, and variable speed & power consumption optional by dial control"

Also interestingly, (per the second comment), there is an optional model which allows variable control natively via a 0-5V input, which the MEGA can do with a simple resistor/capacitor circuit. I wonder if it is possible to upgrade your units?

Otherwise we need to see if this model can be controlled natively via PWM, or if that PWM will interfere with the internal "micro processor" mentioned.
 
I did not realize that the screw shield had a prototyping area! I have never gotten into Arduino before, or any electronics building/fixing. I would agree on not liking the idea of using a breadboard for production use. Any recommendation on a decent, but on the cheaper side soldering kit, this is something that I have somewhat looked into. I would not shy away from doing any soldering, just need to do a little practice!

So far all I have is the Topsflo TD5 pump. I plan on doing a pump in a toolbox setup as I do not have a permanent brew station. I just use a 4 foot folding table and my propane burner has extension legs. So I need to keep it portable and not permanent.

Thanks for the information and the help! I am looking forward to eventually getting this build moving forward! I know I could do this cheaper using other items, but I love the capabilities of this software and the expansion possibilities!
 
TenaCJed said:
I am interested in more information regarding the 24V Topsflo, as I have the TD5-A24-2505-P/US-M model, and was thinking about using the MOSFET board to control the output flow instead of using a controllable ball valve. I am still looking into a controllable open/close ball valve(s), just not a flow control ball valve if I can get away with it.

How are you connecting the addon components to the MEGA? It looks like even the standard temp probe DS18B20 requires at least a resistor to be installed. Are you making your own custom boards or using breadboards? This is the area that I need to do more research into.

I was looking at starting with 2 temp probes, one on the output of mash tun, and one on RIMS tube. I was also looking into the flow meter that you use as well.

Thanks!!!
Click to expand...
http://m.ebay.com/itm/1-2M-DS18B20-...7Cparentrq%3A0eecf63215c0ab1de6aa7e4affff5838 these work well for the one wire sensor I tried the resister and for some reason it wouldn't work for me. I use the rtd adapter boards as well as the customer brucontrol thermister board since I actually use all three different types of temp probes for different things.

As far as the TD5.. I'm pretty sure they all have the mppt buck/boost chip in them designed to take varying voltage from solar panels and buck it down while converting it to amperage or to take lower voltage and boost it... This unfortunately directly counteracts the pwm frequency making the speed control less linear and not work as well.
I have a TD5 clone and tried the pwm controllers and while it did work somewhat, it didn't work nearly as well as it does with the tan 24v food grade p38i pumps I use now.
 
Thanks @augiedoggy. I have emailed Topsflo to see their recommendations for modulating this pump. They do have a 0-5V regulating model, perhaps it is possible to just switch the onboard micro to enable this functionality? That would make it idiot simple.
 
@crane, here is a look at what the display might show. This pic was taken inside with a flash. I could not test it indoors, but I suspect it would be OK outdoors.
DisplayTest.jpg

Though this is a possible avenue, I want to reiterate that BruControl would be much better run with a legit monitor!
 
Yeah, I am happy with this one. I don't have an original Yun to compare to, but I think these are the same hardware, just a different branding in the Liunx distro.

I haven't stressed it yet, so I can't make any reliability claims yet. I would assume the extra software layer *could* introduce some communication hiccups, but I do see the value of the easier setup and the Ethernet plus Wi-Fi connectivity on one board that the Yun brings over the standard boards/shields. If price were not super-critical, I would probably recommend the Yun over the standard Wi-Fi or Ethernet shield assuming the reliability is the same.

The documentation on board versions is the really tricky part. There are multiple versions (see http://wiki.dragino.com/index.php?title=Yun_Shield#Difference_between_v1.x_and_v2.x). This one is the 2.4 version, but I suspect they all work likewise. The pricing varies about $10 between the 2.4 and older versions, so I opted for the 2.4.
 
For the MEGA, you cannot upload a sketch via the serial port. You either need to upload it via the Yun's interface or via the Arduino IDE's network port. Feel free to PM me if you have more questions/comments - will keep this thread cleaner.
 
BrunDog said:
The original project scope looked to have this compiled on a RPi, but we found it difficult/impossible to get it to run under Linux. We had been considering Win10 IOT, but that might require structural changes. Anyway, I know Windows has its issues but the RPi is way too buggy for automation control in our opinion.

That said, you can use the RPI to view/operate PC via RDP - this works well. The touchscreen KB stinks though (I used Florence and its OK at best). When the web interface is published, obviously you can use the RPi's browser.
Click to expand...

I take this to mean that BruControl can't run the Hosehead controllers since they use a RPI. Is this correct?

I can't imagine the amount of time and effort y'all put into this. And I hope your effort pays some really good dividends!
 
Yes, that's correct, currently. We understand there is interest in it running on RPi but it's a technical hurdle we would need to clear. It's not a priority for right now as we are ensuring base functionality, setup, etc. is solid. We will add web server next then will go from there.
 
BrunDog said:
Yes, that's correct, currently. We understand there is interest in it running on RPi but it's a technical hurdle we would need to clear. It's not a priority for right now as we are ensuring base functionality, setup, etc. is solid. We will add web server next then will go from there.
Click to expand...

Thanks for the clarification. I totally get the need for priorities, and don't expect you guys to overcome every technical hurdle. I'm still in awe that there are people like yourself that are smart enough to figure stuff like this out.

I know little too nothing about computer based control, but I wonder if it would be possible for Corey at Brewtronix to change over to an Andruino based system. I'll ping him and see if he has any thoughts.
 
BrunDog said:
I doubt they would want to share anything that would risk selling more controllers, and I would respect that. The LCD would be a better option IMO in that more information can be displayed.

However, I really need to reiterate how much is lost by not using a monitor / BC user interface. Even a cheap monitor and an RPi just for screen sharing would be a worthwhile option for not much $.
Click to expand...

I know you are very resistant to add support for these, but I remembered that brewtroller was open source, so here is the code for their 7 segment displays.

https://github.com/OSCSYS/brewtroller/blob/master/Com_BTPD.pde
 
BrunDog, Since we are on the topic of LCDs (sort of), I see in the latest wiring diagrams how to wire the LCD backpack, but I can't glean from the information I read anywhere on your site or adafruit's what the actual addresses used in the firmware. I have two of them in my cabinet, and I like using both a tablet for graphics and touchscreen control as well as the LCDs for a quick update. Are the addresses like 0x20 and 0x21 for example, so I can split data to both by using different addresses?
 
@crane: thank you, I will take a look at the code. I am busy pushing a version & firmware update, then will look after.

@purdman: the firmware expects the display to be at the default address of 0x20. You basically do not need to worry about the address since it is the default. I haven't multiple displays, so I would need to test that before confirming it can be done.

Here is the thing with "more" devices... in BruControl's implementation, each micro-controller interface runs autonomously. And timing is important. When you say you want a duty cycle of 60%... then change it to 61%... as far as I am concerned, it had better dang be 61%. When you add auxiliary devices which run on buses, sometimes these timings are affected. For example, the "universal" display I wanted to use was the 3" TFT... and I tested it but it was so slow to update it would have been impossible to implement. I have seen similar impacts with the Yun shield. The Linux overhead affects the real-time nature of the micro-controller.

So while I want to give you all the bells as whistles, I don't want to sacrifice core functionality. Automation can be buggy by nature, and I want to deliver a quality product. So please bear with me as I evaluate your requests and don't commit anything until I can confidently tell you it's good. Thanks!!
 
Hi,

FYI, I just posted updates to the software application (v1.0 build 28079), firmware (v41), Interface Wiring Maps, and User Manual. This is a summary of the changes:

1. Updated Hysteresis Device Elements to include an Active Low setting (for active low relay boards) and an offset model. The offset model permits the Target to be adjusted while keeping hysteresis window the same.

2. Added Yun Shield support. The Yun shield facilitates Wi-Fi or Ethernet configuration. Note we have noticed this shield may impart some timing delays (likely due to the onboard Linux OS), so we suggest it not be used time-specific / critical applications. Fermenter control, etc. is fine, but discuss your application with us before implementing.

3. Added ESP8266 integration (except RTD support). NodeMCU and Wemos D1 are supported, and most other ESP8266 configurations should work out of the box.

4. Updated local LCD control. One 20x4 LCD can be added for interface-local data presentation (per discussion above). This was previously implemented in the software but we made some enhancements for speed and usability.

NOTE: BruControl build 28079 and firmware v41 must be used together! To update, close BruControl. Update the interface firmware, then start the new version of BruControl. Note the Hysteresis device elements must be updated to set the proper On Offset (Hysteresis window).
 
BrunDog said:
Thanks @augiedoggy. I have emailed Topsflo to see their recommendations for modulating this pump. They do have a 0-5V regulating model, perhaps it is possible to just switch the onboard micro to enable this functionality? That would make it idiot simple.
Click to expand...

I e-mailed them as well, they basically said that the pump cannot be modified after the fact and I would need to purchase the pump that has the control built in.

It took several back and forth e-mails due to a slight language barrier!


So even though it would not work well to control the flow of the pump, would it work good if kept at 100% to turn the pump on and off? This seems like a cheaper option then SSR that is DC in and DC out.
 
Well, there is no reason to run it with PWM/Analog. Just use a digital on/off output from the controller. That doesn't need to power an SSR - it can use a regular relay as the pump is relatively low power. Or you can use a small low current SSR.

Remember SSR's, MOSFET's, transistors, electromechanical relays all do essentially the same thing with minor differences, acting to take a small power signal and switch high current loads. The pump is relatively low (compared to the elements) at around 2 amps at 120VAC.

I think your choice depends on what other things you want to control, as it is easier to buy, mount, and wire relays in banks rather than individuals. The only individuals should be the element SSRs in my opinion.
 
Quick update while I am here: we are working on some updates to v1.0 to help with disconnection recovery and international conventions. Thanks to users for reporting these and for your patience as we make these improvements!

We will then begin the updates for v1.1. I will provide a list of what will be included, but as promised web server interface will be the priority.
 
BrunDog said:
Well, there is no reason to run it with PWM/Analog. Just use a digital on/off output from the controller. That doesn't need to power an SSR - it can use a regular relay as the pump is relatively low power. Or you can use a small low current SSR.

Remember SSR's, MOSFET's, transistors, electromechanical relays all do essentially the same thing with minor differences, acting to take a small power signal and switch high current loads. The pump is relatively low (compared to the elements) at around 2 amps at 120VAC.

I think your choice depends on what other things you want to control, as it is easier to buy, mount, and wire relays in banks rather than individuals. The only individuals should be the element SSRs in my opinion.
Click to expand...

EDIT::: Never mind, I did not realize that the relay I was already looking at can switch AC and DC, so I am back on track!!! :::END EDIT

I was just thinking about using the MOSFET due to price, and having difficulty finding a relay for 24VDC and 2 amps, as that is what my pump is and did not plan on using the AC to DC converter that the pump came with.
 
Ya! Most electromechanical relay boards we use can handle 10A. 2A is a nothing-burger to those! These have the benefit of NO and NC contacts too, as electrical switches do not. Of course there are other pros and cons though.
 
Hi All,

FYI we just posted an updated version, 1.0 Build 28080, which includes the following changes:

• Changed text boxes to number boxes for calibration edit dialogs for locale support (fixes decimal conventions for EU users etc.)
• Timed auto-save for scripts (checks every 30 seconds).
• Fix for Hysteresis On Offset not allowing digits after the decimal point.
• Expanded serial port re-connection options.

Let me know if you have any questions or concerns!
 
Man! I've been working hard for three days to see if I can use a stripped down XP using either Bochs or Limbo on an android tablet and run BruControl on it! I forgot about the upcoming web server. I still wanna see if I can get it working.
 
Would be curious! However, I would be concerned about reliability. BruControl is meant to act as an control & automation "server", so the underlying hardware needs to be reliable. Please test it hard before committing it to an application where scripts are used.
 
So out of curiosity-- will the web server version allow BruControl to be OS agnostic or is it still bound to Windows?
 
No sir... BC needs to run on a Windows machine. But of course you would be able to access the interface from any browser.

I suppose by your question you don't like PC's / Windows. If this is a matter of cost compared to a RPI, mini-PC's can be had for less than a $100. Yes, more than a Pi but we sincerely believe the platform is more reliable for automation, and safety is very important when managing the type of hardware our rigs power.

If you don't like the platform, well I can appreciate your preference. I like it (it's come a long way in recent years) but respect those who don't!
 
augiedoggy said:
http://m.ebay.com/itm/1-2M-DS18B20-...7Cparentrq%3A0eecf63215c0ab1de6aa7e4affff5838 these work well for the one wire sensor I tried the resister and for some reason it wouldn't work for me. I use the rtd adapter boards as well as the customer brucontrol thermister board since I actually use all three different types of temp probes for different things.



As far as the TD5.. I'm pretty sure they all have the mppt buck/boost chip in them designed to take varying voltage from solar panels and buck it down while converting it to amperage or to take lower voltage and boost it... This unfortunately directly counteracts the pwm frequency making the speed control less linear and not work as well.

I have a TD5 clone and tried the pwm controllers and while it did work somewhat, it didn't work nearly as well as it does with the tan 24v food grade p38i pumps I use now.
Click to expand...


Could you use PID with the flow rate as your target to automate Pump in theory?
 
I would use a MOSFET board instead of an SSR. They are geared to switch faster. Otherwise, correct.

Also, depending on your application, it might be better to use a PWM Output rather than a PID Output to drive the pump. The PWM Output control could be varied using a script. I suggest this as I contemplate it because The reaction to PID output changes will occur very rapidly, and stability might be difficult to attain. The rate changes can be limited (aka damped) via the script whereas damping is trickier with PID (dependent on tuning).

In any event, the hardware is the same. You can try PID then switch to PWM if it doesn't work to your liking.
 
I was just starting to order components to build around a different system when I saw augiedoggy mention this in another thread and I am very glad I did. I like the flexibility and the ease of expanding the system.

I have a couple of quick questions while I continue learning about this since I was in the process of ordering a few parts.

Are the pumps run through an SSR?

I did not see any switches or contactors in your system. I have read that SSR's have a tendency to fail on, is this not a concern?

What style temp probes do you recommend?
 
Hi @shadow160,

You can run pumps through an SSR or through electromechanical relays. Standard March or Chuggers only draw about 2A @ 120VAC so an appropriately sized relay is all you need. My personal brewery has these driven by SSRs but if I were building again I might use EM's.

In my brewery, I do have contactors in front of my element SSRs. I recommend this is done for anyone's brewery. I switch them via my E-stop so that anytime they should not be on, the E-stop can be pushed.

Regarding probes, that is personal choice. 1-wire are easy but have limited physical options (mount, length, etc.). Thermistors work well but need supporting hardware. RTD's (Pt100) give you the best accuracy and offer the most mount options but need supporting hardware too.

Let me know how else I can help.
 

Similar threads

h22lude
My BruControl UniFlex review
Replies
3
Views
2K
h22lude
h22lude
duncan.brown
Brewery Dashboard
Replies
18
Views
2K
suprchunk
suprchunk
braumke
Automate your electric brewery - 4 Testers needed for open source project
Replies
29
Views
6K
estricklin
estricklin
RonBurgundy10
BruControl Uniflex: Automation For All
2
Replies
51
Views
8K
BrunDog
BrunDog
muhteeus
muh's BruControl Build
Replies
25
Views
6K
sensei247
sensei247
Yakima 300

Latest posts

Back
Top