Comments on BCS 4.0 public beta 3

Things to come.
Post Reply
Timbrew
Posts: 34
Joined: Sat Jun 29, 2013 2:40 pm
Bot?: No
Location: Kent, UK

Comments on BCS 4.0 public beta 3

Post by Timbrew »

I’ve been enjoying 4.0 beta 3 for a few weeks during the commissioning of my new-build electric HERMS. This has a BCS-462 running a semi-automated three-keg system (3 Chuggers, HLT and BK heating and several 12 volt ball and solenoid valves; I have 4 floats controlling levels for 5 and 10 gallon batches). My early trials have thrown up a few questions / issues. Maybe these are naïve or already covered in the forum - so regrets for any cross-posts. Overall, the BCS is a great development; IMHO 4.0 adds much to the BCS’s ease of use. I particularly like the fact that the system is both intuitive and flexible, lending itself to skinning the cat in ever more intricate ways once you get into the system.

My comments/issues (in no particular order and mainly minor niggles under the circs) :
1. I can’t see how to set up an overall system timer which runs across all processes from start to finish.
2. Programming exit conditions in some processes seems to have glitches which I think have already been reported. In my case, process 2 state 1 (and some others) sometimes has blanks entries and/or will not accept a next state. I think I saw a work-around (by flipping into timers and then out again ??) but assume this is something you’re working on for the final version.
3. I’d like to be able to reorder states and move them ‘up’ or ‘down’ within processes, and processes within the overall system. This view is a function of how I approached programming. Now I’ve worked up some nice 5 and 10 gallon variants, many of my states are confusingly shown in the wrong order. Either perhaps they should reorder themselves depending on mutual exit condition parameters or there could be some drag and drop facility to move them up or down in the order shown.
4. I have several floats (as DINs) which don’t reset themselves following a system stop which has resulted in a couple of floods argh! The only way to deal with this as far as I can see is to manually flip each float before starting.
5. The title of the system or recipe could be a usual addition to the main screen so you know (or can simply check) which one you’re in.
6. Is there any demand of off-line programming? On many occasions when my BCS is switched off or remote, I’d have liked to sit in front of my laptop and get on with designs…
7. I have a few points on the Recipe Utility. I can’t find much in the forum on this or on how it’s supposed to work. However, although clunky it’s obviously potentially useful for storing sets of variables parameters say for 5 gallon and 10 gallon variants or for single and multi-step mashes, etc. Some issues: maybe I’m using this incorrectly (and it’s difficult to systematically document repeatable glitches…so I haven’t!!), but I’m generally finding that backing up one recipe messes up the variable setup in the other (e.g. blanks, temp changed to times). A further hard to explain issue is that under certain circumstances, I can only change [Celcius] values to integers under Recipe Values - it won’t accept any decimal points. The integer values are then reflected in the BCS program (e.g. 66.7 input in the Recipe Values truncates to 66 which renders as 66 in the BCS). This is weird because if I go into the BCS process editor and change (say) all instances of mash-temp to 66.7 C, the Recipe Value does then show the correct value (66.7 C). Do input values in the BCS determine the Recipe Value or vice versa? If the latter, how can I turn it off?
8. Maybe related to this (??) is the fact that when I edit some of the temps in the BCS process editor I can sometimes see either a red box or a blue box – what do these mean as I Can’t see a logical link with the recipe utility. OK, I guess this is still a work in progress but at least I’d like to understand some more about the recipe functionality.
9. Finally, when I switched on today my BCS had reverted to an unprogrammed status (Firmware 4 but no process/state defined and temps back to Fahrenheit, etc.). Restoring the files worked fine and all is working great, but that inexplicable loss was a bit worrying.

Apologies for all this, but hope it might be useful and I’d appreciate any feedback.
JonW
Site Admin
Posts: 1726
Joined: Sun Jul 18, 2010 7:51 am
Bot?: No
Location: Huntington Beach, CA
Contact:

Re: Comments on BCS 4.0 public beta 3

Post by JonW »

I'll try to address your questions/observations, etc.

1. There is no overall system timer. Timers are only within a process/state.
2. Not sure if this is the same issue reported or not. Can you post pics and more details please.
3. You can move processes via the backup/restore utility. States cannot be reordered. You should design your states like a workflow where you progress through them. It may take you a few iterations, but once you get your workflow dialed in, you rarely change process/state programming.
4. I believe there is a bug entered for this. It only happens if you use the all stop function. All stop is usually used as the emergency stop and in normal operation you should stop your processes directly.
5. The recipe utility writes values to your configuration - it's doesn't work like Excel/Word, etc. that open a file and know what file is open.
6. Offline programming could be accomplished via a program that saved values and wrote them to the BCS later via the API. However, that is not anything that is on the roadmap to provide.
7. The recipe utility is merely an interface to quickly update values of timers, temps, etc. within your programming without having to go to different process/state screens directly. It isn't meant as a means of saving individual recipes. Instead you create a mapping and then just update the values you want the BCS to use when you need to modify for a different recipe - you do not redo the mappings though, you just enter the new values for things like mash temp, mash time, boil time, etc.
8. Not sure off hand on the coloring...
9. Sounds like a system reset was performed. There have been no reports of units spontaneously clearing their configuration.
Timbrew
Posts: 34
Joined: Sat Jun 29, 2013 2:40 pm
Bot?: No
Location: Kent, UK

Re: Comments on BCS 4.0 public beta 3

Post by Timbrew »

Thanks for this. Some more detail on my point 8 in first post. And incidentally, I wasn’t able to repeat issue 2 (exit condition editing problems), although it has definitely occurred from time to time over the last few weeks – I’ve always worked around by fiddling the drop downs (someone else mentioned this I think). On point 8:

1. OK, I get the mapping function of the recipe utility. However, I’m still experiencing dropping decimal points and strange behaviors. I have repeated the following to ensure I’m not imagining it:

2. I’m using MacBook and Chrome. I have a mash-temp variable mapped to my mash process states A “HERMS” (HLT PID set point temp) and B “Mash temp” (HLT exit condition temp) respectively. I open a tab for recipe utility and another for BCS (on process editor) and hit refresh when going from one to another. The routine is as follows:

3. Open recipe utility and change mash-temp variable from 66.8 to say 72.8 (I tried a range of non-integer nos with same result). The number truncates to 72 and won’t allow any decimals.

4. Open BCS, refresh, find that both A and B have changed to 72, then the interesting bit.

5. I change the temp value directly in BCS under B to 72.8 (n.b. I get a blue then red outline when I add the decimal; also tried with other non-integers). Reverting to the recipe utility and refreshing, the value has changed to 72.8.

6. However, if I repeat the above (5.) but instead change the value directly in A, (n.b. I get a blue outline), reverting to the recipe utility and refreshing, the value has NOT changed.

7. My conclusion from this half-hour fiddle seems to be that there’s some glitch in the BCS programming which results in truncating the decimal point under certain conditions. What this means in practice is that I open my recipe utility, change my mash-temp to say 68.2, and this renders the incorrect value 68 in the BCS. The (my) only way around this is to manually correct the numbers in the BCS which defeats the aim of the recipe utility.

Finally, reordering states and offline programming would seem to me to be potentially very useful adjuncts especially with new users (which assume you’re trying to attract) who will –like me- adopt a hit and miss approach until they fully understand the system, and despite having mapped out the unit operations beforehand... Reentering 5 processes by 8 states with many variables means you’ll probably need to resort to clunky <print screen> for each state followed by copying into final locations. OK, appreciate that once programmed in you may not need to change much, but already I’m finding I’m playing by shifting states around – e.g. chilling and recirculation and tricks to adjust temps efficiently.

Hope this helps!
Timbrew
Posts: 34
Joined: Sat Jun 29, 2013 2:40 pm
Bot?: No
Location: Kent, UK

Re: Comments on BCS 4.0 public beta 3

Post by Timbrew »

2. Not sure if this is the same issue reported or not. Can you post pics and more details please.
Not sure if this helps, but I was able to repeat this error during one recent session, but then not on another, so not sure what's going on. However, it might be a bug worth checking out.

First image (actually the second of my uploads below) shows that selecting exit condition <digital input> results in an empty selection box, which also repeats in all four exit conditions.

Second image (...the first of my uploads below) shows that refreshing the screen does not clear the problem as well as rendering a strange screen layout.

The only work around it seems is to select timers....the digital inputs then reappear. Think someone else has reported this but can't find the reference.

This issue only crops up in certain States in more than one Process - I've counted about three instances where this happens, but don't have the details with me right now.

I want to further support my earlier call to enable states to be moved up or down - I spent half a day shifting them manually which resulted in several errors creeping in, one of which (2 valve controls switched around...) created an MLT overflow situation :(

I also further believe that offline programming with a facility for later online download would be really useful so I don't have to keep my system switched on.
Attachments
Screen Shot 2015-08-29 at 14.00.32.png
Screen Shot 2015-08-29 at 14.00.32.png (209.1 KiB) Viewed 8443 times
Screen Shot 2015-08-29 at 14.00.13.png
Screen Shot 2015-08-29 at 14.00.13.png (216.17 KiB) Viewed 8443 times
Timbrew
Posts: 34
Joined: Sat Jun 29, 2013 2:40 pm
Bot?: No
Location: Kent, UK

Re: Comments on BCS 4.0 public beta 3

Post by Timbrew »

Oh, and my other point about Celcius figures being truncated in the recipe facility remains an issue for me, to the extent that I can't use it to update a range of Celcius values with decimal points - e.g. 66.8C for my Mash-temp....
A further hard to explain issue is that under certain circumstances, I can only change [Celcius] values to integers under Recipe Values - it won’t accept any decimal points. The integer values are then reflected in the BCS program (e.g. 66.7 input in the Recipe Values truncates to 66 which renders as 66 in the BCS). This is weird because if I go into the BCS process editor and change (say) all instances of mash-temp to 66.7 C, the Recipe Value does then show the correct value (66.7 C).
BrunDog
Posts: 96
Joined: Wed Mar 25, 2015 6:56 am
Bot?: No

Re: Comments on BCS 4.0 public beta 3

Post by BrunDog »

I SOOOOOO agree with being able to move states up or down. I have found myself playing the copy game too many times in an effort to add or remove steps. A simple copy or arrow scheme would be great!!

-BD
JonW
Site Admin
Posts: 1726
Joined: Sun Jul 18, 2010 7:51 am
Bot?: No
Location: Huntington Beach, CA
Contact:

Re: Comments on BCS 4.0 public beta 3

Post by JonW »

BrunDog wrote:I SOOOOOO agree with being able to move states up or down. I have found myself playing the copy game too many times in an effort to add or remove steps. A simple copy or arrow scheme would be great!!

-BD
Exit conditions really throw a wrench making moves. If all you look at in the state are the outputs/timers/registers that you are controlling, then moving the state up or down is simple. However, when you take into account the exit conditions and how they tie to other states, it really complicates the logic to easily move states.

If you want a quick and dirty way to move what you're controlling in a state (disregarding exit conditions), I can give you a program I wrote for testing API function calls. You could make a call to read "/api/process/1/state/4" and then do a write of the config back to something like "/api/process/1/state/3". It would overwrite what was in the destination state though, not move it.
BrunDog
Posts: 96
Joined: Wed Mar 25, 2015 6:56 am
Bot?: No

Re: Comments on BCS 4.0 public beta 3

Post by BrunDog »

Understood on the exits, and it's a solid point. I guess it is better than having to manually copy states up and down, but this will not continuentonhappen ad infinitem, so not the biggest priority out there. Adjusting temp and duty cycle on the main screen is! Hahaha!!

-BD
JonW
Site Admin
Posts: 1726
Joined: Sun Jul 18, 2010 7:51 am
Bot?: No
Location: Huntington Beach, CA
Contact:

Re: Comments on BCS 4.0 public beta 3

Post by JonW »

BrunDog wrote:Understood on the exits, and it's a solid point. I guess it is better than having to manually copy states up and down, but this will not continuentonhappen ad infinitem, so not the biggest priority out there. Adjusting temp and duty cycle on the main screen is! Hahaha!!

-BD
Yeah, once you get your processes dial in, they will change very little. I also went through that stage where I was tweaking them after every brew day.

Adjusting temp & duty cycle on the fly is very high on the priority list right now. We're looking at what it will take to add some new options to the API to facilitate this not only for the built-in BCS interface (which is fully driven by the API), but for external programs like my own BCS-Desktop. Adjusting either temp or duty cycle is more complicated than it sounds since the current running values can be asserted in 64 different places (8 processes x 8 states). To make the adjustment, you have to query for running processes, then active states, then asserted outputs and then determine the highest numbered process who is controlling that output. Updating the value is then written back to that process/states stored setpoint value. Because of the overhead involved in doing this, we're looking to make it an API call so a single query will tell who is controlling a given output, and then the code can directly update the value there.
BrunDog
Posts: 96
Joined: Wed Mar 25, 2015 6:56 am
Bot?: No

Re: Comments on BCS 4.0 public beta 3

Post by BrunDog »

I guess that is more complicated than I thought... the guages already know the setpoint. Why can't that control have both an input and output?

Anyway... ignore me. I have no idea how it works, so I am talking out my arse.

-BD
Post Reply