BrewBlox Tilt Service

There was an issue with some of them that can occur if logging to the cloud rather than locally. They replaced mine with a new one with no quibbles, great service by the company.

1 Like

Thats some good advise, it is working now. Thanks!

I’ve got the brewblox-tilt service running successfully. Now I’m curious how / whether I can add some of its related metrics, like specific gravity and temperature, to the DisplaySettings widget (front panel display of the Spark3).

Is it only possible to display “native” metrics on the panel? (i.e., only metrics provided by the spark-one service)

Native blocks only, I’m afraid. The lcd display uses internal settings, not what we send to the history service on the Pi.

It would be possible if we create a mocked sensor on the spark that receives a value from the message exchange through devcon.

With that implementation, the tilt temp sensor could also be used as input. Although I still would not recommend that.

I’ve just pushed an update that should improve stability of the service. I’ve been testing it for a while now and it seems much better. Would be interested if those who were having bluetooth issues see any improvement.

@j616s,

Have just updated to the lates firmware and Tilt Serivce.

Was getting this before but thought I would wait for the update. Tilts are working but the data is only displayed on the Graph intermittently as you can see for both Tilts.

Not sure if this is a Tilt Service issue or a Brewblox Graph issue but my normal graphs for Temp PID PWM etc all show correctly.

https://termbin.com/r98e

I don’t have enough knowledge of how brewblox graphs work to know what might cause this. @Bob_Steers any thoughts?

It’s very strange.

As you can see from this screenshot the other lines plot perfectly, except the Tilt ones.

Looks like plotly doing its best to render relatively sparse data. Each column is 15 seconds.

@rbpalmer zoom in on the second graph until it has the same x resolution, I’d imagine you’d see something comparable.

1 Like

As an aside, putting the SG on the same axis as the actuator states might make the plot scale better vertically than with the temperatures.

@Bob_Steers

Is this what you meant? As you can see the normal Temp, Settings plots correctly but the Tilt is intermittent. Should plotty just not fill in the gaps between each data point?

@j616s Agree with your comment on the Y axis and will make that change.

I’m not surprised to see the data from the tilt being more sparse. Even with the service running reliably, the wireless connection itself will be less reliable than the wired connection to the other sensors. Even using the Tilt mobile app, updates aren’t regularly spaced and are only every ~3-15secs with a full battery. So I don’t think there’s a problem in the data or service here.

I wonder if there’s any tweaks that can be made to plotly to make it less picky about how close points need to be before they’re linked?

I can have a look at what’s going on with plotly.

@rbpalmer could you please export your graph to csv?

@Bob_Steers as requested the export.

graph-Cellar Graph-spark-one-downsample_1m.csv (2.6 KB)

Just noticed that the export doesn’t include the Tilt data. IS there a way to include this?

Export should have generated multiple files (one per measurement/service)

Ok got them now.

graph-Cellar Graph-tilt-downsample_1m.csv (2.6 KB) graph-Cellar Graph-spark-one-downsample_1m.csv (4.2 KB)

Thanks! I’ll squeeze in initial assessment when I’m done with current tasks.

1 Like

I ran this and my tilt integrated without any problem.

However, I was curious if anyone has any ideas on how to get the “corrected” SG data to import also. My TILT hydrometer displays both an “uncorrected” SG and one built off of a manual calibration. Brewblox as installed with above code only shows the “uncorrected”. I am guessing that this type of calibration is done by the TILT app not and as such is not part of the signal from the Hydrometer itself. I have reached out to TILT as well to see if they have any ideas.

@mmarshallod

Have you seen the Calibration section on @j616s GitHub>

Is that what you are interested in?