[Version 2 Release] RaspberryPints - Digital Taplist Solution

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.
That's what the other guy did with his uno. The uno doesn't show up on ttyusb0 it's ttyacm0. And he's having problems with it showing up. I'm working on getting mine up and running on a nano. I'll run through the setup and see if anything goes south for me.
 
Ok so I'm trying out the seeedstudio flowmeter. I've got it hooked up to small fountain pump and I'm doing milliliter tests using my flask. I've tried using the pulse count that was mentioned earlier for these meters. But I'm getting a reading of 4 oz per every real 600ml which is off by about 8 or 9 ounces. I'm using the pulse count setting in pours.php like day_tripper recommended. How should I proceed. Should I just keep lowering the number until I reach equilibrium, or is there a better way to do this?
 
No. There's math involved in the python script that reads the info from the flow meter.

I don't know where it is from memory, I'll check the code when the sun is up and let you know.
 
The flowmeter from seedstudio is the one that counts 450 pulses per liter? I just buy one from adafruit, and now I'm testing it. I've tried it with 3.3v and it works fine. The same with the sf800, it works fine at 3.3v too. Now I have to calibrate the one from adafruit. I think I have the code to test it... Let me check...
 
Ok. I've found a decent setting to get within 1oz accuracy per every 1000ml. I changed the pulse count in pours.php to 3000. Some more testing and tweaking and I could finalize that number. It's probably more or less within the ball park. 3 tests in now and I'm reading a constant 32 oz for approx 1000ml. Which is a bit off. But not too bad considering most pint glasses are not true pint glasses and are an ounce or 2 light.
 
I've used that code before. I don't really need to know what adafruit thinks a liter is. I need to know how to calibrate rpints effectively. So loading an entirely different sketch onto my arduino isn't going to calibrate rpints.
 
Ok so I've got the cheap flow monitor working on a cheap Chinese nano. It's registering pours and I'm close enough to accurate. Now I've been trying to get the whole thing working over bluetooth. I know it's possible because we've been doing it successfully using arduinos with brewpi installed. I am able to connect over Bluetooth without an issue. Starting flow_monitor.py actually connects faster than the Python scripts do for brewpi. The only issue I'm running into now seems to be similar to what I've seen before with mis-matched baud rates. When I run the script in terminal and simulate a pour, instead of getting pour info I get "unknown message:ááááááááááá"

I know no one else is doing this and I'm just shooting out in the dark looking for help. Just looking for ideas of where to go from here.
 
mcangeli said:
No. There's math involved in the python script that reads the info from the flow meter.

I don't know where it is from memory, I'll check the code when the sun is up and let you know.
Click to expand...

fwiw, there is some math coded in flow_monitor.py, but it was apparently only used for debugging and is commented-out.

The pour-size math actually used exists in pours.php, where the pulses per ounce (165 for the Swissflow meters) is translated into pulses per gallon (21120) to generate the fractional gallon pour value that ends up being posted to the database...

Cheers!
 
day_trippr said:
fwiw, there is some math coded in flow_monitor.py, but it was apparently only used for debugging and is commented-out.

The pour-size math actually used exists in pours.php, where the pulses per ounce (165 for the Swissflow meters) is translated into pulses per gallon (21120) to generate the fractional gallon pour value that ends up being posted to the database...

Cheers!
Click to expand...

Yea, I moved the math to the pours.php file so that it could in theory be updated via the website to handle changing calibration.
 
wbarber69 said:
[...]The only issue I'm running into now seems to be similar to what I've seen before with mis-matched baud rates. When I run the script in terminal and simulate a pour, instead of getting pour info I get "unknown message:ááááááááááá"

I know no one else is doing this and I'm just shooting out in the dark looking for help. Just looking for ideas of where to go from here.
Click to expand...

Certainly looks like a baud rate mismatch.

Just for context, when you do this do you compile the R'Pints sketch for a nano? Then hook up the BT/serial module to the nano and use a power-only cable on the microUSB receptacle?

The sketch uses 9600 baud, so you'd have to set the BT/serial module up for that or change the sketch; either one would be ok I guess. We were using 57600 for BrewPi...

Cheers!
 
Got the serial message to de-garble. I changed the ino file to 57600 baud. But now I'm still receiving unknown message errors only it's because the serial data is coming through all chopped up. ie:

"Unknown message: P"
"Unknown message: ;0;8;1786"

Where it should be reading:
P;0;8;1786
 
So far it's a no go. Besides all the trouble trying to program an hc-05. Running at 9600 baud is not better than 57600 in this case. I'm thinking of increasing everything to 115200 tomorrow and giving it a shot.
 
To close the loop on my FAR less complex issue - going to a quality 2+ amp charger with a quality low resistance cable, I haven't had any issues. Thanks all.

If anyone wants links to the specific products I used, I can send them to you.
 
Connecting the beer line to the flow meters this weekend...

Any tips? Anyone have problems with leaks?

Thanks,
 
hughbrew said:
Connecting the beer line to the flow meters this weekend...

Any tips? Anyone have problems with leaks?

Thanks,
Click to expand...

Two points of advice.

If you are using the refurbished Swiss flows, do not over-tighten the fittings on either side of the meter. The ones that convert the threaded to the john guest fittings. I have destroyed two of them when I realized I had one that was under-tightened and leaked out a whole keg.

Secondly. Pay attention where you put the meters. Too close to the taps and you can get some foaming though I didn't get all that much more when I made that mistake. Since my lines were old I replaced them and put the meters about .5-1 foot from the gas out and that seems to work pretty good.
 
I agree with the meter placement - as close to the keg post as possible without making maintenance a pita. I nestled the meters inside the keg rubbers then tie-wrapped the slack tubing in a coil on top. Changing a keg out is nbd.

Further, your dispensing quality should be totally sound before you add meters.
So if you have foamy or borderline quality pours, see the only beer line length calculator worth using to see what you should be using for beer line and lengths, and get that squared away.

fwiw, with 12' 3/16" ID Bevlex-200 (pvc) lines and a system pressure of 11 psi the pour quality remained excellent after adding the 6-pack of SF800 meters to my keezer...

Cheers!
 
I found good results from just removing the existing line from the liquid out connector and adding the meter there with 1.5-2' of new beer line. I figured that since I used lengths close to what the calculator day_tripper posted, that since I had decent pours before, why not just consider the meter the new line out connection.
 
so back to my dilemma trying to get rpints installed using a nano and bt. ive centered my bad messages down to something other than the bt. so i removed all the extra hardware and went back to scratch. i reverted everything back to stock. and now i cant get anything from the arduino.

i get this message when a pour tries to register:

pi@BrewPi /var/www/rpints/python $ sudo python flow_monitor.py
There are no windows in the stack
Invalid window '%1'
Usage: windowactivate [options] [window=%1]
--sync - only exit once the window is active (is visible + active)
If no window is given, %1 is used. See WINDOW STACK in xdotool(1)
^[[15;5-
 
Thanks to everyone for the advice, tips and patience. I'm not overly technical, so this was a big deal for me. But, I am happy to say my rpints is working, registering pours on both taps and the Wharton Brewing logo looks great! Special thanks to Day_Tripper and the whole rpints team. This is a really cool project and if I can do it, trust me anyone can!

Thanks to all!

And yes, I also made a donation to rpints via their paypal link. Thanks guys!

:rockin::D
 
wbarber69 said:
so back to my dilemma trying to get rpints installed using a nano and bt. ive centered my bad messages down to something other than the bt. so i removed all the extra hardware and went back to scratch. i reverted everything back to stock. and now i cant get anything from the arduino.

i get this message when a pour tries to register:

pi@BrewPi /var/www/rpints/python $ sudo python flow_monitor.py
There are no windows in the stack
Invalid window '%1'
Usage: windowactivate [options] [window=%1]
--sync - only exit once the window is active (is visible + active)
If no window is given, %1 is used. See WINDOW STACK in xdotool(1)
^[[15;5-
Click to expand...

What's "back to stock" in your case?

When you're running the python script in a terminal, did you first kill the same process being run automagically via the system service (flowmon)?

Cheers!
 
hughbrew said:
Thanks to everyone for the advice, tips and patience. I'm not overly technical, so this was a big deal for me. But, I am happy to say my rpints is working, registering pours on both taps and the Wharton Brewing logo looks great![...]
Click to expand...

Always happy to see another RaspberryPints tap list come on line!

And I'd love to take some credit for the development of RaspberryPints but the most I can claim is "helpful user"...

Cheers! :mug:
 
day_trippr said:
What's "back to stock" in your case?



When you're running the python script in a terminal, did you first kill the same process being run automagically via the system service (flowmon)?



Cheers!
Click to expand...


Maybe not

Edit: no I haven't set it to run at startup yet till I get it working properly. Just checked and double checked and it's not running nor is it set to run at startup. If I start flowmon and try to register a pour, it exits out of the Python script. If I stop flowmon then I get the error I posted above instead. It's weird because I had all this working before.
 
I can see the messages perfectly in serial monitor. I just can't get flow_monitor.py to stop giving me the error.
 
In the pours.php script comment out the part that calls for the refresh of the screen.

I think the error you're seeing is related to the fact that it's trying to refresh window 1 but you do not have a monitor with a browser open connected to it.

The auto refresh on a pour only works if you have a monitor connected.
 
The suggested root cause seemed easy enough to replicate on my dev system, so here's what I did:

Code: 
pi@rpints ~ $ flowstat
[ ok ] /var/www/python/flow_monitor.py is running.
pi@rpints ~ $ flowstop
[ ok ] Stopping system RPFlowMon daemon:.
pi@rpints ~ $ www
pi@rpints /var/www
pi@rpints /var/www $ cd python
pi@rpints /var/www/python $ ls
flowmon  flow_monitor.py
pi@rpints /var/www/python $ sudo python flow_monitor.py

Pour:
  - Addr : 0x0
  - Pin  : 2
  - Count: 2612

No protocol specified
Error: Can't open display: (null)
Failed creating new xdo instance
No protocol specified
Error: Can't open display: (null)
Failed creating new xdo instance
No protocol specified
Error: Can't open display: (null)
Failed creating new xdo instance
Keg Kicked:
  - Addr : 0x0
  - Pin  : 2

I did not have anything displaying the tap list.
So R'Pints does get cranky if there's no display to update on a pour.
But I get a different set of messages than what @wbarber69 is seeing.

As an aside, it's interesting that I got that Keg Kicked message...

Cheers!
 
Did a little monitor upgrade thanks to a freebie. Something to keep me busy while I wait for you guys to give me another reason to spend money. LolView attachment ImageUploadedByHome Brew1433380556.557954.jpgView attachment ImageUploadedByHome Brew1433380629.737134.jpg

Went from a 20 inch monitor sitting on the top to a 32 inch hanging on the brick. Sure makes changing kegs easier...

Edit: Excuse the horrible lighting situation. This room is my basement and even though the lights suck as they are, every bit of light I attempted (Even the camera flash) created a bad glare.
 

Similar threads

Tobor_8thMan
[Version 2 Release] RaspberryPints - Digital Taplist Solution - Tobor-8thMan Version
Replies
6
Views
4K
day_trippr
day_trippr
Fred S
RaspberryPints - Latest HW / SW as of Dec 2017 (PRi 3 B; Raspian Oct 2017)
Replies
21
Views
3K
day_trippr
day_trippr
OkieNotFromMuskogee
A Brewpi Remix Build
2
Replies
52
Views
10K
bloombrews
bloombrews
garzlok
My Fermentrack Build
2
Replies
56
Views
17K
garzlok
garzlok
B
DIY 5KW Induction Heater
Replies
2
Views
5K
Spivey24
S

Latest posts

Back
Top