Add "flow rate" calculation by measuring what arrives in the cup #437
Conversation
FabianSperrle
commented
Jan 6, 2024
FabianSperrle
commented
- determine weight in the cup whenever we get a new weight measurement, divide by time since last measurement to get flow rate into the cup
- added a sliding window average over the last 10 measurements to get smoother output
- adjusted display template to show flow rate
|
In case this is well-received: would you also want flowrate (and weight) published via MQTT? |
|
I'm not into the scale, need the space to the portafilter for the cup, so I can't judge any scale related features. |
Until I buy a bottomless portafilter I just screwed off the spout to gain some space, otherwise it would be painful, yes... |
| weightBrew = weight - weightPreBrew; | ||
|
|
There was a problem hiding this comment.
I don't know why this was here before -- the weight is already calculated in brewvoid.h, why do it again here?
There was a problem hiding this comment.
Half done refactoring and moving stuff out of main?
There was a problem hiding this comment.
Unless somebody remebers/knows why it should be here I think removing it here is good. Now that I looked at it again I would guess that the entire shottimerscale() could be removed.
src/brewscaleini.h
Outdated
| float scaleDelayValue = 2.5; // value in gramm that takes still flows onto the scale after brew is stopped | ||
| bool scaleFailure = false; | ||
| const unsigned long intervalWeight = 200; // weight scale | ||
| const unsigned long intervalWeight = 50; // weight scale |
There was a problem hiding this comment.
Increased the scale frequency -- why was it so low before?
Potentially there could be a further change to only read the scale during shots to save some CPU cycles when idling?
FabianSperrle
force-pushed
the
flowrate-by-weight
branch
2 times, most recently
from
78685fe to
f2aa129
Compare
Switch to EMA smoothing Finalization: Lower EMA a to 0.05 for stable readings, increase scale frequency to get high resolution, and prevent shot weights <0g
FabianSperrle
force-pushed
the
flowrate-by-weight
branch
from
f2aa129 to
5f281e4
Compare
|
Superseded by #450 |
3 participants
