NEW StrangeBrew Elsinore Thread

Yakima 728
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.
And the manual duty mode had some issues, all fixed (I think). There's now three different duty modes.
 
Last edited by a moderator:
jangevaa said:
Interesting... I haven't wired any extra ports for temp sensors (probably my largest oversight in my panel build), so I can only have three plugged in now. I think this would give me motivation roll up my sleeves and change that however.

It's not possible for me to fit in any extra relays though - so, you're saying there potentially would be a way of linking a single GPIO to two sensors/PIDs in a single configuration file, with having only one of those PIDs in command at a time.
Click to expand...


Just for you jangevaa, you can now hide devices/temps from the WebUI, even if they're not setup. When you're in edit mode you'll see all the devices, and when you click the title of a device to modify it, there'll be an extra button, "Hide" or "Show" to change the state. Click it, and when you've locked the UI the elements will be hidden.

In theory, you can run two separate elsinore instances on the same device, just don't cross the GPIO outputs, or things'll get weird.
 
For the pumps I would just use regular mechanical relays. SSRs are great for loads that need to be switched repeatedly but they get HOT and need to have heat sinks and thermal management planning. Standard mechanical relays do not have that limitation and are perfect for things like pumps or valves where you won't be using PWM/PID. You can use SSRs for pumps, but make sure you account for the heat. There probably won't be nearly as much heat with a 2A pump as there is for a 23A element, but it won't be none.

-Josh
 
The SainSmart 4-channel relay module on Amazon should do well for you. <$10 and leaves you some room for future expansion.
 
Last edited by a moderator:
That's the one, they must be monitoring this thread because it was just over $9 yesterday. For its current price, another $1.70 gets you the 8-channel board.
 
A quick(ish, it only took me one movie) present to y'all before I go on Vacation for a bit tomorrow.

You can now edit the status recorder (i.e. log recorder) settings from the Web UI.

Hit "Edit" and you'll have the ability to set whether the recorder is enabled or not, and the threshold values, and the minimum time between samplings (in ms).

There's some GUI improvements as well, and some more to come to tidy it up. But I'm away for just over a week and can't do any more testing.

Oh, I've also fixed a couple of migration bugs over the last couple of commits (mainly from when I enabled the cooling control in the Web UI).
 
petewa said:
the only way i could get that relay to work was to build this thing...

http://www.raspberrypi.org/forums/viewtopic.php?t=36225
Click to expand...

Ahhhh, this is why people want to use the inverted outputs mode on Elsinore!

The issue would be that using the inverted outputs option inverts ALL the outputs.

I'll add some changes in so you can select an inverted output for an individual pinout (I want to remove as many options as possible so everything can be done from the web interface)
 
Doug-
I have been following your threads for quite a while as I built my three keg super strut system over the last 15 months. I hacked Android onto an HP TouchPad and installed the server on an Rpi and the app on the Touchpad. I have also been building a TeensyPi and a TeensyNet on Jim Mayhugh's thread. Is there a way to Frankenstein-merge your software to that hardware? Just a thought, I am only good at following directions most of the time. Thanks.
 
purdman10 said:
Doug-
I have been following your threads for quite a while as I built my three keg super strut system over the last 15 months. I hacked Android onto an HP TouchPad and installed the server on an Rpi and the app on the Touchpad. I have also been building a TeensyPi and a TeensyNet on Jim Mayhugh's thread. Is there a way to Frankenstein-merge your software to that hardware? Just a thought, I am only good at following directions most of the time. Thanks.
Click to expand...

The teensypi looks like it already does the basics of what you need for temperature control.

But if you want Elsinore, all you need to make sure is that the /sys/devices/ has one wire on it.

Or you could just try to start Elsinore and see if it works :), I can't give direct support though.

As for android, the Elsinore App isn't supported anymore since Elsinore now has a mature web interface.
 
Thanks for the overview. I now understand the differences. Your work is great... I am glad you continued to improve it for us.
 
DougEdey said:
Ahhhh, this is why people want to use the inverted outputs mode on Elsinore!

The issue would be that using the inverted outputs option inverts ALL the outputs.

I'll add some changes in so you can select an inverted output for an individual pinout (I want to remove as many options as possible so everything can be done from the web interface)
Click to expand...

inverted on selected pins would be awesome. I love this program... you are doing a GREAT JOB.

would love to see timers count down not up.. hehe

how would i go about requesting different features?

thanks again for all your hard work!
 
petewa said:
inverted on selected pins would be awesome. I love this program... you are doing a GREAT JOB.

would love to see timers count down not up.. hehe

how would i go about requesting different features?

thanks again for all your hard work!
Click to expand...

If you've got a feature request the best way is to raise a bug on github for it.

The timer count down is something that's taking a lot of thought since I need to identify how to get it to work on the UI, since a new pop up library has come out which is nicer to use, I may revisit it. The other issue (which is still there with the timers) is how to mark the actual start/stop time, originally I did it with a timestamp on the Web UI being sent to the server but there could be a huge difference if you're not using NTP, this is why there's sometimes a couple of seconds between pressing start and the timer starting.
 
Doug - great work on this! I'm very impressed by your work on Strangebrew Elsinore. I'm just wondering how folks handle multiple heater elements on a single controller, especially if power is limited to say 30A. On a typical rig, you might have an HLT and boil kettle with elements, and perhaps a RIMS tube to maintain mash temps. Would it be possible to have the software control output to the RIMS and HLT elements such that the load would never exceed 30A? During heating of the HLT, the MLT element would be off. But, to maintain temp of the HLT once the temp was reached, the element would not need to be on at 100%. During the off time, the MLT element could be on to maintain the mash temp. Once the mash was complete, the RIMS tube could increase to mashout/sparge temp, and then maintain sparge temp while activating the boil kettle element to bring it to a boil. In the best case scenario, we'd all have 60A 240 lines run to our brew stands, but this is rarely the case. Seems like it wouldn't be all that difficult to incorporate some prioritization code to intelligently split the current between heating elements.
 
I'm interested in this too but I'm also in the process of running a 60a circuit. :) I plan on using 4 elements though (2ea in hlt and bk) and then using a hardware pcb I'm still working on to guarantee only two could be fired simultaneously with priority of having one on in each if both are heating and then two on in one only if the other vessel isn't heating or is in a dwell period of the PID functions. It'd be easier in software, but I wanted a foolproof design in case a software glitch enabled more than two (or in your case more than 1) for whatever reason.

It gets tricky also because most ssrs are zero-crossing meaning they only switch states when the phase of the AC power crosses 0. This means that even if the controller only had one of your two outputs high at a time, you'd have to make sure one didn't activate within 1/120 of a second to ensure the other ssr had crossed zero and switched off.

Edit: forget that second paragraph. As I lay here thinking about it, as long as all of your ssrs are zero-crossing it shouldn't matter since the other wouldn't change state until crossing zero either. If you're mixing and matching ssr types, you would still have to take that into account. ORRR...Just assume that the sub-second nature of the times they may overlap would not be enough to trip the 30a breaker or heat up your wires substantially.

-Josh
 
I think managing based on the amount of current is introducing a lot of complexity for setting it up. And switching the elements so they aren't under complete control by a PID output could create software confusion. You'd need to identify how strong each element is, or how many elements to activate at once, then there's settings for how long reach element can be used, I would assume one duty cycle at a time, but that may not always be too useful for larger setups, so the alternative would be to use temperature difference, but then there could be too much droop and you'd be wasting energy
 
JayElDubya said:
I'm interested in this too but I'm also in the process of running a 60a circuit. :) I plan on using 4 elements though (2ea in hlt and bk) and then using a hardware pcb I'm still working on to guarantee only two could be fired simultaneously with priority of having one on in each if both are heating and then two on in one only if the other vessel isn't heating or is in a dwell period of the PID functions. It'd be easier in software, but I wanted a foolproof design in case a software glitch enabled more than two (or in your case more than 1) for whatever reason.

It gets tricky also because most ssrs are zero-crossing meaning they only switch states when the phase of the AC power crosses 0. This means that even if the controller only had one of your two outputs high at a time, you'd have to make sure one didn't activate within 1/120 of a second to ensure the other ssr had crossed zero and switched off.

Edit: forget that second paragraph. As I lay here thinking about it, as long as all of your ssrs are zero-crossing it shouldn't matter since the other wouldn't change state until crossing zero either. If you're mixing and matching ssr types, you would still have to take that into account. ORRR...Just assume that the sub-second nature of the times they may overlap would not be enough to trip the 30a breaker or heat up your wires substantially.

-Josh
Click to expand...


I have a 40 amp breaker... and i run both my 5500 HLT and my 3500 boil elements at the same time. No issues at all. one thing i want to figure out is how to wire in amp and voltage meters to both hot legs. just not sure how to do that.
 
JayElDubya said:
I'm interested in this too but I'm also in the process of running a 60a circuit. :) I plan on using 4 elements though (2ea in hlt and bk) and then using a hardware pcb I'm still working on to guarantee only two could be fired simultaneously with priority of having one on in each if both are heating and then two on in one only if the other vessel isn't heating or is in a dwell period of the PID functions. It'd be easier in software, but I wanted a foolproof design in case a software glitch enabled more than two (or in your case more than 1) for whatever reason.

It gets tricky also because most ssrs are zero-crossing meaning they only switch states when the phase of the AC power crosses 0. This means that even if the controller only had one of your two outputs high at a time, you'd have to make sure one didn't activate within 1/120 of a second to ensure the other ssr had crossed zero and switched off.

Edit: forget that second paragraph. As I lay here thinking about it, as long as all of your ssrs are zero-crossing it shouldn't matter since the other wouldn't change state until crossing zero either. If you're mixing and matching ssr types, you would still have to take that into account. ORRR...Just assume that the sub-second nature of the times they may overlap would not be enough to trip the 30a breaker or heat up your wires substantially.

-Josh
Click to expand...

Josh:

I'd be interested to see your hardware method. I agree that doing it in software might be better since it would be possible to incorporate this into the PID logic, associate priorities for different heating loads (e.g. lower priority for boil vs. mash, etc). I'm just really curious about how others do this when limited to say 30A circuits. I currently do BIAB with a Blichmann burner and propane, but am considering moving to eBIAB and would like to have a second vessel for parallel brewing, sharing a single 30A circuit for the two.
 
Once I get it figured out, I'll start a new thread. What I have in mind should work with either 30a or 60a setups.
-Josh
 
A few bug fixes for the mashcontrol code, so if you delete all the steps when the mash profile is activated the control turns off. As well as fixing the countdown timer when a step is active.

Also capped the duty so if you don't have a cooling output it'll be from 0 -> 100 and when you don't have a heating output it'll be -100 -> 0
 
I have never been able to update via the GUI... I always end up updating via command prompt. I just tried to do it again, and it seems to get stuck (see screenshot... http://imgur.com/HgDiQNU)

Not sure if I'm doing something wrong, or if this is a bug..
 
Swab said:
Not sure if I'm doing something wrong, or if this is a bug..
Click to expand...

It's not stuck, it's not detecting the update unless you restart elsinore after I've pushed an update. Trying to work out what's going on now.

EDIT: Fixed, you'll need to do a manual update first though, sorry
 
Swab said:
I have never been able to update via the GUI... I always end up updating via command prompt. I just tried to do it again, and it seems to get stuck (see screenshot... http://imgur.com/HgDiQNU)

Not sure if I'm doing something wrong, or if this is a bug..
Click to expand...

I have always had to perform the check twice and it works on the second time...
 
petewa said:
I have always had to perform the check twice and it works on the second time...
Click to expand...

That makes sense, I wasn't waiting for the initial update check to complete before comparing it to the current hash, so what Elsinore though was the remote hash was actually the current hash.
 
Just a quick note to say that Corey has started to sell Hosehead on his website, he's responsible for customer queries directly (he's got contact info on his site), but $425 for a two element controller is neat imo.
 
anyway to get this to run without have a temp probe installed yet? I want to get it up and running to see if i can get my relay to switch? If not i can just wait until i get my temp probes next week.

Guess i should add that i just wanted to add a couple of pumps and see if i could get those to switch on and off. I am testing the SainSmart relay.

thanks,
 
TechyDork said:
anyway to get this to run without have a temp probe installed yet? I want to get it up and running to see if i can get my relay to switch? If not i can just wait until i get my temp probes next week.

Guess i should add that i just wanted to add a couple of pumps and see if i could get those to switch on and off. I am testing the SainSmart relay.

thanks,
Click to expand...

No, sorry, there was one added in about a year ago but it's vanished and I can't work out where it went.
 
Merged the changes that prevent the calibration temperature from throwing an error and refactored all the divide calls to use a 2 decimal point math context.

I want to do some more work on the decimal rounding but it's a trade off between the floatingness off the java double and the accuracy of the BigDecimal
 
TechyDork said:
No problem. Thanks!

Sent from HBT app.
Click to expand...

The email I got was asking about turning the GPIO on and off from the command line.

If you're on the Beaglebone (or Linux Kernel 3.8+) you'll need to export the DTO first.

Then you can use echo commands to test the outputs themselves.

There's lots of guides to show you how to use many ways to test the outputs. Which I think is what you're looking for.
 
DougEdey said:
The email I got was asking about turning the GPIO on and off from the command line.



If you're on the Beaglebone (or Linux Kernel 3.8+) you'll need to export the DTO first.



Then you can use echo commands to test the outputs themselves.



There's lots of guides to show you how to use many ways to test the outputs. Which I think is what you're looking for.
Click to expand...


Thanks for the reply. I posted that then quickly told myself "just google" it.
 
Has anyone got any bugs or problems that they've noticed? I just want to make sure that there's nothing major before I start working on features.
 
Tried several times today to run program and keep getting this message. "Couldn't read the one wire devices directory!" I have one temp probe installed. Went through the forum several times trying to find answer. Any help would be appreciated. Thanks.


Starting Elsinore as pi
Nov 11, 2014 3:16:06 PM com.sb.elsinore.LaunchControl main
INFO: Running Brewery Controller.
CFG IS NULL
DOC IS NULL
Couldn't read the one wire devices directory!
Creating element of pumps
Creating on configDoc base
 
You'll need to give more information. What system are you using? Did you follow the instructions to install the one wire module or the one wire cape?
 

Similar threads

S
Mypin setup/configuration
Replies
2
Views
257
scubahiker
S
S
New to Electric brewing
Replies
14
Views
1K
Exception13
Exception13
rkhanso
For Sale Completed/tested BrewPiLess-32 Controllers for sale. $40, shipping included
Replies
9
Views
2K
rkhanso
rkhanso
ALAN C
Need help to confirm wiring .
Replies
17
Views
2K
ALAN C
ALAN C
B
For Sale 3-Vessel Single-Tier Direct-Fire Beer Brewstand Pilot System
Replies
0
Views
593
BrettinSD
B
Yakima 300

Latest posts

Back
Top