Replacement firmware for iSpindel (GravityMon)

[rkhanso]

My iSpindel got wet inside with beer on the last brew. Before the iSpindel got wet it worked OK with Fermentrack, but I'm getting ready to brew again and testing everything out first. The iSpindel didn't work for me the other day. Someone suggested Gravitymon, and I'm also trying Brewpiless instead of Fermentrack. When Gravitymon booted up, I saw it couldn't see the Dallas sensor, so I replaced it but later found out it wasn't bad. I also replaced the D1 Mini (with Gravitymon on it instead of iSpindel firmware), the Charging PCB, and added a jumper wire from the Dallas Sensor resistor to the D1 Mini (must have broken the trace at some point - the reason the Dallas sensor was not seen).

I now have Gravitymon running instead of iSpindel firmware. I can access the web interface of Gravitymon. I set up the URL for Brewpiless in the Gravitymon web interface. When I do a test push, the data shows up fine in Brewpiless. No trouble. As long as Gravitymon is in config mode, the updates get sent to and show up in Brewpiless every 10 seconds (for testing). As soon as I stand up the iSpindel off of horizontal, the blue LED stops flashing on the iSpindel and it won't update Brewpiless anymore. However, if I press the reset button on the D1 Mini, I'll see just one update in Brewpiless, but none after that.

I suspect it's a Gravitymon/iSpindel problem and not a Brewpiless problem since when sending the test push, data shows up. I can't figure out how to fix this.

 

[DuncB]

Have you calibrated the gravitymon for angle I notice this in the instructions.

Configuration - Device Settings - Gyro Calibration​

You need to place the device on a flat surface and then press the calibrate button. It will take a few seconds for this to complete and the angle should be close to 90 degrees. Without calibration the device will not go into gravity mode.

 

[rkhanso]

Digging in a little more, it's not waking up from deep sleep. It's set to update every 30 seconds, but after many minutes, here are the logs - not waking up.

Yes - I've calibrated. There are no yellow warnings at the top of the web page when using the battery. Now I have the battery removed and using the USB cable, and it does show a battery warning only.

I change from the flashing blue LED in the horizontal position to more vertical to get it in the running mode. It will update one time when pushing the reset button when in the upright position, but only that one time. It won't wake up again.
Serial monitor:

rll▒r$▒n
▒
l▒
b|▒▒▒rb▒
b▒nnlnob
bpp▒$blrlrll▒r$▒n
▒
l▒
b|▒▒▒rb▒
b▒nnlnnb
bp▒$blrlp▒n
▒
l
▒
b
n▒n▒
▒
b▒▒nn'l▒l`▒nn
l`nr▒▒▒n
bl▒`p▒n
r▒b
b
n▒n
c▒▒nn'
l`▒nn
l`nr▒▒▒n
b▒`r▒▒n
b▒`▒
b▒
l`▒n▒`Serial connection established
75 I: SDBG: Serial logging started at 115200.
76 I: Main: Started setup for dafed8.
76 I: Build options: 1.3.0 (..b68835) LOGLEVEL 4 ESP8266
92 I: CFG : Filesystem mounted.
96 I: WIFI: Current reset counter 0.
112 I: HELP: Last reset cause 0
127 I: CFG : Size of configuration file=1065 bytes.
142 I: CFG : Configuration file /gravitymon.json loaded.
146 W: CFG : Configuration file does not exist /hardware.json.
1040 I: Main: Battery 14.23 V, Gyro=86.49, Run-mode=1.
fpm close 3
mode : sta(58:bf:25:da:fe:d8)
add if0
1145 I: WIFI: Connecting to wifi (0) using stored settings myssid.
.........................scandone
state: 0 -> 2 (b0)
.state: 2 -> 3 (0)
state: 3 -> 5 (10)
add 0
aid 2
cnt
connected with myssid, channel 1
dhcp client start...
ip:192.168.2.32,mask:255.255.255.0,gw:192.168.2.1
6803 I: WIFI: Connected to wifi myssid ip=192.168.2.32.
6804 I: WIFI: Using mDNS name iSpindel1.
6839 I: TSEN: Found 1 temperature sensor(s). Using 9 bit
6839 I: Main: Activating web server.
6840 E: WIFI: OTA error checking version.json, response=-1
6843 I: WEB : Configuring web server.
6863 I: WEB : File system Total=1024000, Used=57344.
6875 I: WEB : File=error.log, 2122 bytes
6880 I: WEB : File=error2.log, 4005 bytes
6884 I: WEB : File=gravitymon.json, 1065 bytes
6887 I: WEB : File=reset.dat, 1 bytes
6887 I: WEB : Setting up handlers for web server.
6939 I: WEB : Web server started.
6939 I: Main: Setup completed.
pm open,type:2 0
25942 I: WEB : webServer callback for /api/status(get).
30953 I: WEB : webServer callback for /api/status(get).
35965 I: WEB : webServer callback for /api/status(get).
37098 I: PUSH: Free-heap 22 kb, Heap-rag 3 %, Max-block 21 kb Stack=2460 b.
37115 I: TPL : Transformed template 318 chars to 240 chars
37115 I: PUSH: Adding header 'Content-Type': 'application/json'
38251 I: PUSH: HTTP post successful, response=200
41158 I: WEB : webServer callback for /api/status(get).
41240 I: WEB : webServer callback for /api/test/push.
41254 I: TPL : Transformed template 318 chars to 239 chars
41255 I: PUSH: Adding header 'Content-Type': 'application/json'
41324 I: PUSH: HTTP post successful, response=200
42217 I: WEB : webServer callback for /api/test/push.
42237 I: WEB : webServer callback for /api/test/push.
42256 I: WEB : webServer callback for /api/test/push.
42275 I: WEB : webServer callback for /api/test/push.
56373 I: WEB : webServer callback for /api/config(get).
56455 I: WEB : webServer callback for /api/config/advanced(get).
67799 I: PUSH: Free-heap 21 kb, Heap-rag 16 %, Max-block 18 kb Stack=1168 b.
67807 I: TPL : Transformed template 318 chars to 240 chars
67808 I: PUSH: Adding header 'Content-Type': 'application/json'
67869 I: PUSH: HTTP post successful, response=200
97974 I: PUSH: Free-heap 22 kb, Heap-rag 11 %, Max-block 20 kb Stack=1168 b.
97985 I: TPL : Transformed template 318 chars to 240 chars
97986 I: PUSH: Adding header 'Content-Type': 'application/json'
98014 I: PUSH: HTTP post successful, response=200
128120 I: PUSH: Free-heap 22 kb, Heap-rag 3 %, Max-block 22 kb Stack=1168 b.
128133 I: TPL : Transformed template 318 chars to 241 chars
128134 I: PUSH: Adding header 'Content-Type': 'application/json'
128160 I: PUSH: HTTP post successful, response=200
130040 I: GYRO: Movement detected (500)      6251    211     37.
131046 I: GYRO: Movement detected (500) 3779    1323    644.
132057 I: PUSH: Free-heap 22 kb, Heap-rag 8 %, Max-block 20 kb Stack=1168 b.
132071 I: TPL : Transformed template 318 chars to 240 chars
132071 I: PUSH: Adding header 'Content-Type': 'application/json'
132136 I: PUSH: HTTP post successful, response=200
132142 I: WIFI: Stop double reset detection.
132162 I: MAIN: Entering deep sleep for 30s, run time 132.09s, battery=3.56V.
                                                state: 5 -> 0 (0)
                                                                 rm 0
                                                                     pm close 7
                                                                               del if0
                                                                                      usl
                                                                                         enter deep sleep

 

[DuncB]

Your battery says 5%
New feature in 1.3 is that it will go to sleep and only send every hour, does the battery setting need calibrating?

• Added battery saving mode. When battery drops below 20% the sleep interval will be changed to 1 hour. Can be disabled in advanced configuration.

 

[rkhanso]

Your battery says 5%
New feature in 1.3 is that it will go to sleep and only send every hour, does the battery setting need calibrating?

• Added battery saving mode. When battery drops below 20% the sleep interval will be changed to 1 hour. Can be disabled in advanced configuration.

I didn't have the battery in when I captured that serial output. Just connected to USB port on PC.
I changed the advance settings to disable that feature and it did something different, but wouldn't power up and send data. I only got a little jibberish. Maybe some of those other settings need to also be checked.

woke up from sleep, but serial output only had this:
l▒
b|▒▒▒rblb▒nnlnnb
bp▒$blrl▒r▒nnnbp

 

[rkhanso]

I recalculated the battery level/config with the USB cable plugged in. It's now reading 100% battery and I shut off the advanced features.
I'm still getting only jibberish when it's time to come out of deep sleep. And when the next interval is supposed to happen, nothing. Not even the jibberish. Though in the serial output, it shows battery is only at 2.46V. But in the GUI, it said

Measurement​

Gravity:
1.0472 SG
Temperature:
74.78 F
Angle/Tilt:
50.641
Battery:
4.92 V (100%)
Average runtime:
0 seconds

113836 I: MAIN: Entering deep sleep for 30s, run time 113.76s, battery=2.46V.
rm 0
pm close 7
del if0
usl
enter deep sleep▒怘▒▒▒~▒f▒▒x▒▒x▒▒▒~▒▒ff~fx▒▒▒▒▒▒▒▒`▒▒▒`fx▒▒▒x▒▒▒▒▒怘▒f~▒f▒ff▒▒▒

 

[rkhanso]

I found the problem. In addition to the trace from the Dallas sensor back to the ESP being open and needing a jumper wire, both traces from the diode were bad back to the ESP. I added 2 jumper wires from each side of the diode back to RST and D0 and it now wakes up and updates Brewpiless.

 

[Anatropia]

Plizz/ tell me how to change the Wi-Fi network without overwriting everything (ESP32Lite (Floaty))

 

[kdw2pd]

I'm trying to configure an iSpindel with GravityMon 1.3 to cooperate with Spindelmate without success. iSpindel --> Spindelmate setup details at: iSPINDEL support for SPINDELMATE – iBLOPPER – iot fermentation logger!

Two questions:
1.) Is the Device Name in the GravityMon configuration the same as the Chip ID in the original iSpindel firmware? If not, where can I find the ESP8266 Chip ID?
2.) Trying to figure out the push configuration to send data from the iSpindel to Spindelmate. In the original firmware (shown in the link), you specified the server address, port, and path separately. To send data to the Spindelmate port 80 with a path of /, would it just be http://192.168.X.XXX:80/ for HTTP1 Post? I think I tried that, and it wasn't working.

 

[fuzzybee]

I searched through the thread, but possibly not good enough for my answer. If I switch from the iSpindel firmware to GravityMon, do I need to go through recalibration again, or can I input my polynomial from the iSpindel software?

 

[mper]

I searched through the thread, but possibly not good enough for my answer. If I switch from the iSpindel firmware to GravityMon, do I need to go through recalibration again, or can I input my polynomial from the iSpindel software?

You should be able to use the same polynomial, but do a check/verification to make sure. We use the same software libraries for calculation and gyro reading. Let me know how it works out and i can update the docs.

 

[DuncB]

I searched through the thread, but possibly not good enough for my answer. If I switch from the iSpindel firmware to GravityMon, do I need to go through recalibration again, or can I input my polynomial from the iSpindel software?

I certainly used my Ispindel polynomial , it was a bit off but that was due to the fact the angles had changed a bit. But I just factored in that change of gravity.

 

[mper]

I certainly used my Ispindel polynomial , it was a bit off but that was due to the fact the angles had changed a bit. But I just factored in that change of gravity.

That is probably due to the zero calibration that will not be 100% the same as the ispindle. Wonder if it could be possible to convert the configuration from an ispindle to gravitymon and get better accuracy. Would be an interesting feature to implement.

 

[DuncB]

That is probably due to the zero calibration that will not be 100% the same as the ispindle. Wonder if it could be possible to convert the configuration from an ispindle to gravitymon and get better accuracy. Would be an interesting feature to implement.

All of the angles were off but you get some " drift " with any movement of the sled.
It would be good if the polynomial could be reset on the fly with any new zero and the starting original gravity.
This might require all of the calibration angles to be recorded on the ispindel or a utility off the ispindel to work on the numbers making an adjustment producing a new formula to input.
I've found if I can get the zeroing and in water angles very close to the calibration angles originally making the formula that it becomes accurate again.

 

[Level 3 Brewing]

Glad I found this thread! Great Work Magnus very impressed with what you have done!

I have two Ispindels that I built and have been successfully running stock Ispindel firmware on them over the last 4 years, but the battery and temperatures were always off. Your software makes this issue very easy to correct. So I upgraded my first Ispindel to version 1.3.0. Everything worked flawlessly. Thank you! A few days later when my second Ispindel became available I decided to upgrade it as well, but when I go to calibrate the device it locks up or loops. Everything else works in the configuration, just not able to calibrate the device. I believe I saw this issue in this thread earlier, but was unable to find it now. In the bug fixes it looks like something was done in version 1.3.0 to address this issue.

When I connect the device to Putty to peek at what is going on I get the following; "WEB : webServer callback for /api/calibrate." then a bunch of " ***********************************".

Any Ideas?

 

[mper]

Glad I found this thread! Great Work Magnus very impressed with what you have done!

I have two Ispindels that I built and have been successfully running stock Ispindel firmware on them over the last 4 years, but the battery and temperatures were always off. Your software makes this issue very easy to correct. So I upgraded my first Ispindel to version 1.3.0. Everything worked flawlessly. Thank you! A few days later when my second Ispindel became available I decided to upgrade it as well, but when I go to calibrate the device it locks up or loops. Everything else works in the configuration, just not able to calibrate the device. I believe I saw this issue in this thread earlier, but was unable to find it now. In the bug fixes it looks like something was done in version 1.3.0 to address this issue.

When I connect the device to Putty to peek at what is going on I get the following; "WEB : webServer callback for /api/calibrate." then a bunch of " ***********************************".

Any Ideas?

There was a bug that was fixed but that was more connected to when a gyro was not present. The correct printout should look something like this in the serial log;

13888 I: WEB : webServer callback for /api/calibrate.
>......>......526.00000, 628.00000, 732.00000, 94.00000, -62.00000, 3.00000

The calibration is done in a library that I use and when I looked at the code it will print * when there is a problem with the values from the gyro.

I dont think this is something I have seen before but, if it has worked before it should not be an faulty gyro.

It might be a good idea to have a hardware testing feature in the software to check for any issues that might occur. I have been working on a feature for 1.4 that allows the user to copy the formula and gyro calibration from the ispindel configuration file and skip the calibration steps. This might be an option that can be tested. I still have some more work to do on that feature before I can publish a beta version.

 

[mper]

Input wanted!

I'm working a a feature in the next version 1.4 that allows a user to copy formula and gyro calibration data from an old ispindel configuration file. Are there any other parameters that might we worth to copy ? Like mdns name, interval etc. I think the integration settings will be a little hard to migrate but other data should be fine.

 

[mper]

Glad I found this thread! Great Work Magnus very impressed with what you have done!

I have two Ispindels that I built and have been successfully running stock Ispindel firmware on them over the last 4 years, but the battery and temperatures were always off. Your software makes this issue very easy to correct. So I upgraded my first Ispindel to version 1.3.0. Everything worked flawlessly. Thank you! A few days later when my second Ispindel became available I decided to upgrade it as well, but when I go to calibrate the device it locks up or loops. Everything else works in the configuration, just not able to calibrate the device. I believe I saw this issue in this thread earlier, but was unable to find it now. In the bug fixes it looks like something was done in version 1.3.0 to address this issue.

When I connect the device to Putty to peek at what is going on I get the following; "WEB : webServer callback for /api/calibrate." then a bunch of " ***********************************".

Any Ideas?

I did some simple tests and you will get this behavior if the sensor is moving or giving inconsistent values when read. make sure the spindel is not moving when doing the calibration.

The calibration routine will run a PID algorithm to find a stable level and if the values are way off this will make the calibration fail and you will get * in the printout instead of . which is the sign for a good value.

 

[mper]

New beta out, new features:

* Easy migration from iSpidle
* Edit wifi settings in web ui

https://github.com/mp-se/gravitymon/releases/tag/v1.4.0-beta1

 

[bscuderi]

I had an ispindel that must have gotten a little beer or moisture in it cause it wouldn't switch on and had some signs of corrosion. I cleaned it up with some rubbing alcohol and poked around with a multimeter. I determined it had a bad switch so I had to unsolder the d1 mini on there to change out the switch. While I had that off I figured it was the perfect time to upgrade the chip to an s2 mini I had laying around and experiment with this firmware instead of ispindel. I have to say great work! It really is way more user friendly the configuration mode and software is everything ispindel needed to be a better piece of hardware so thank you! This all had me thinking has anyone found a better substitute for the tubes instead of the soda preforms? I hate how the bottom has all that rough knob I feel like overall it has lots of crevices for yeast and bacteria to hitch a ride and they never have had the most reliable seal it seems I've had several ispindels that get a little leaked in not alot but enough to gunk stuff up and cut the lifespan more short. It just seems the soda preforms are the weak link in an otherwise great design. I've even tried adding little gaskets in the lid as insurance.

 

[DuncB]

I have been using an oring that fits in the cap of the petling its' internal diameter is a little bigger than the outer edge of the petling edge. A tiny bit of silicone on the top edge of the petling and then a part tighten and loosen then retighten to full has made them much less permeable.
I too am finding that the switch is a weakpoint in the setup. Tempted to change to a reed switch and then just turn it off with a magnet when I take it out or no switch and just use the gravity mon software settings to turn it off.

I think the best container would be the rapt pill containers which are smoother and double o ring but this would require a redesigned motherboard due to the size difference.

I do pressure ferment and find the oring has been reliable.

Just took Ipindel out of a dessert wine in secondary for the last ten months and no leak or damage, remarkably still sending with a voltage that had dropped to 2.95.

 

[bscuderi]

I have been using an oring that fits in the cap of the petling its' internal diameter is a little bigger than the outer edge of the petling edge. A tiny bit of silicone on the top edge of the petling and then a part tighten and loosen then retighten to full has made them much less permeable.
I too am finding that the switch is a weakpoint in the setup. Tempted to change to a reed switch and then just turn it off with a magnet when I take it out or no switch and just use the gravity mon software settings to turn it off.

I think the best container would be the rapt pill containers which are smoother and double o ring but this would require a redesigned motherboard due to the size difference.

I do pressure ferment and find the oring has been reliable.

Just took Ipindel out of a dessert wine in secondary for the last ten months and no leak or damage, remarkably still sending with a voltage that had dropped to 2.95.

Yeah I saw those rapt pill containers. I'm using the jeffrey open source distilling pcb I'm sure it wont fit. Has anyone tackled designing a pcb for the rapt pill? I have done some custom pcb design and could consider trying it if nobody else has but I'm sure it would take me some trial and error to get a good design and alot of back and fourth orders from china to get it to fit and float right lol so i definitely don't want to reinvent the wheel if someone already has. I have an o ring in my lids which I think it helped alot but its still not perfect by any means. I just feel with a better container this project could be top knotch. Dang man 10 months and it was still reporting that's incredible was it a d1 mini? lLongest I've left one of mine in was a month or two based on D1 minis but I imagine the esp32 stuff should be way more battery friendly.

 

[DuncB]

I've had a good look at the rapt case and also use the Jeffrey PCB. Pretty confident it wouldn't fit.
In fact just checked data on that ispindel using D1 mini started every half hour 6th November 2022and it came out 2 days ago got recharged and into next wine.

I might get a rapt case and see what dimensions need changing on the PCB to get right angle and fit.

 

[mper]

I had an ispindel that must have gotten a little beer or moisture in it cause it wouldn't switch on and had some signs of corrosion. I cleaned it up with some rubbing alcohol and poked around with a multimeter. I determined it had a bad switch so I had to unsolder the d1 mini on there to change out the switch. While I had that off I figured it was the perfect time to upgrade the chip to an s2 mini I had laying around and experiment with this firmware instead of ispindel. I have to say great work! It really is way more user friendly the configuration mode and software is everything ispindel needed to be a better piece of hardware so thank you! This all had me thinking has anyone found a better substitute for the tubes instead of the soda preforms? I hate how the bottom has all that rough knob I feel like overall it has lots of crevices for yeast and bacteria to hitch a ride and they never have had the most reliable seal it seems I've had several ispindels that get a little leaked in not alot but enough to gunk stuff up and cut the lifespan more short. It just seems the soda preforms are the weak link in an otherwise great design. I've even tried adding little gaskets in the lid as insurance.

Thanks for the great feedback.

I havent found any better option for the case. On the leaking i ususally put some keg grease on the cap that seams to work fine and i havent had any problems with leaks

 

[mper]

Yeah I saw those rapt pill containers. I'm using the jeffrey open source distilling pcb I'm sure it wont fit. Has anyone tackled designing a pcb for the rapt pill? I have done some custom pcb design and could consider trying it if nobody else has but I'm sure it would take me some trial and error to get a good design and alot of back and fourth orders from china to get it to fit and float right lol so i definitely don't want to reinvent the wheel if someone already has. I have an o ring in my lids which I think it helped alot but its still not perfect by any means. I just feel with a better container this project could be top knotch. Dang man 10 months and it was still reporting that's incredible was it a d1 mini? lLongest I've left one of mine in was a month or two based on D1 minis but I imagine the esp32 stuff should be way more battery friendly.

On paper the esp32 should be more battery friendly but my testing has shown that the wifi connection takes much longer on an esp32 which has a huge impact on power. An esp32c3 with only ble enabled is the best option with really long batterylife.

Im planning to look into that part to see if i can add more data to the ble transmission and not just gravity and temp

 

[bscuderi]

Thanks for the great feedback.

I havent found any better option for the case. On the leaking i ususally put some keg grease on the cap that seams to work fine and i havent had any problems with leaks

great idea ill lube up my o-rings next time

 

[bscuderi]

On paper the esp32 should be more battery friendly but my testing has shown that the wifi connection takes much longer on an esp32 which has a huge impact on power. An esp32c3 with only ble enabled is the best option with really long batterylife.

Im planning to look into that part to see if i can add more data to the ble transmission and not just gravity and temp

interesting... I've made other battery powered sensors for some smart home stuff some from particle photons which were horrible on the battery. And most recently been using s2 minis and I've got a couple of sensors in my cigar humidor I made with s2 minis and an 18650 battery that have been going over a year since I've charged them reporting temp humidity and battery via mqtt every 10 min. I'm guessing mqtt is of a lighter wifi use. On that note I did add my ispindels into home assistant which is awesome great way to get alerts set up on your fermentation. I didn't try the auto discovery and I noticed the default mqtt message sends every sensor reading as its own message. I felt that might strain the battery so I set it up like my other DIY mqtt sensors for familiarity and in case this helps anyone my theory its one message sent vs spamming multiple message sends is probably better on the battery and server I send one message with a json string of all the values I want. So I Send something like this {"Temp": ${temp}, "Tilt": ${tilt}, "RSSI": ${rssi}, "Gravity": ${corr-gravity-sg}} Then on the home assistant side I do this for each sensor the value template json grabs the applicable value from the JSON string

mqtt:
sensor:
- name: "Beer Hydrometer 1 Temp"
state_topic: "ispindel/gravitymon1"
unit_of_measurement: '°F'
value_template: "{{ value_json.Temp }}"
force_update: true
icon: mdi:coolant-temperature

 

[bscuderi]

Deleted. posted to wrong thread!

 

[DuncB]

A link to this arrived yesterday.
Container looks very promising

https://williamswarn.co.nz/products/wifi-hydrometer

 

[gromitdj]

$189 for something I built for $30.... I don't know what the NZ to dollars conversion is, but no thanks. At that price, I'll just get another Tilt for less.

The container does look good. I wonder where they sourced it, or if it's a custom job. Or if the DIY ispindle parts will fit in it for that matter.

 

[DuncB]

$189 for something I built for $30.... I don't know what the NZ to dollars conversion is, but no thanks. At that price, I'll just get another Tilt for less.

The container does look good. I wonder where they sourced it, or if it's a custom job. Or if the DIY ispindle parts will fit in it for that matter.

Agreed grossly overpriced, custom built board as well.

I'm interested in the container only. Hoping they make them available like they have done for the rapt pill.

100 us dollars is 170 NZ dollars.

 

[mper]

interesting... I've made other battery powered sensors for some smart home stuff some from particle photons which were horrible on the battery. And most recently been using s2 minis and I've got a couple of sensors in my cigar humidor I made with s2 minis and an 18650 battery that have been going over a year since I've charged them reporting temp humidity and battery via mqtt every 10 min. I'm guessing mqtt is of a lighter wifi use. On that note I did add my ispindels into home assistant which is awesome great way to get alerts set up on your fermentation. I didn't try the auto discovery and I noticed the default mqtt message sends every sensor reading as its own message. I felt that might strain the battery so I set it up like my other DIY mqtt sensors for familiarity and in case this helps anyone my theory its one message sent vs spamming multiple message sends is probably better on the battery and server I send one message with a json string of all the values I want. So I Send something like this {"Temp": ${temp}, "Tilt": ${tilt}, "RSSI": ${rssi}, "Gravity": ${corr-gravity-sg}} Then on the home assistant side I do this for each sensor the value template json grabs the applicable value from the JSON string

mqtt:
sensor:
- name: "Beer Hydrometer 1 Temp"
state_topic: "ispindel/gravitymon1"
unit_of_measurement: '°F'
value_template: "{{ value_json.Temp }}"
force_update: true
icon: mdi:coolant-temperature

According to the datasheets for the ESP32 it will use around 50% of the power when BLE is used and WIFI is disabled so In theory this would double the life of the device. I have noticed before that WIFI can take some time to establish a connection which will draw much more power.

I've got a prototype working on the BLE sender and its possible to send 512 bytes of data that way, but I need to do some more tests to see the coverage etc.

 

[DuncB]

I had one ispindel send every half an hour for over 11 months. I can't imagine I'd need the BLE efficiency.

 

[kdw2pd]

Would it be possible to connect a C3 mini in Bluetooth mode to TiltBridge? Off the top of my head, it sounds like it should work, an earlier description of Bluetooth mode sounded like the iSpindel data could be read by the Tilt app. I have a couple of C3 Minis on the way, but wanted to save myself from disappointment if it won't work.

 

[Thorrak]

Would it be possible to connect a C3 mini in Bluetooth mode to TiltBridge? Off the top of my head, it sounds like it should work, an earlier description of Bluetooth mode sounded like the iSpindel data could be read by the Tilt app. I have a couple of C3 Minis on the way, but wanted to save myself from disappointment if it won't work.

To... TiltBridge? If GravityMon emulates a Tilt signal, then yeah, probably, but why would you want that, given iSpindel is already WiFi? You're just as well off going straight to whatever the iSpindel/GravityMon can target.

The Tilt app definitely does not work with iSpindel as Tilts output a Bluetooth signal, and iSpindels do not.

Now, if your question is whether or not a C3 mini can work with the TiltBridge firmware, the answer is also no. TiltBridge is built for ESP32, which is different from the ESP32-C3. If you want to compile TiltBridge for the ESP32-C3 you may be able to, but it is not currently an architecture I want to target.

 

[kdw2pd]

@Thorrak: Yes, using an iSpindel with an ESP32 C3 Mini to emulate a Bluetooth Tilt signal to send to a TiltBridge (GravityMon.com - open source for beer brewing). I've been trying to figure out a way, as a non-programmer, to be able to have more than 1 iSpindel repeater sending data to my Fermentrack setup on my regular WiFi network.

Here's my thinking. I occasionally have trouble getting iSpindel WiFi signals out of my SS fermenters. There's the off the shelf ESP8266 repeater, as well as iSpindHub, Spindelmate, and probably a couple of other WiFi-based repeaters that are designed to help with this. None of them currently offer Fermentrack as a push target. So instead, I'd have to connect my Pi to the AP of the repeater. Not the end of the world by any means, but a minor annoyance. The other issue is if I connect the Pi to one repeater AP, that means I can't use a repeater with the other iSpindel. I'm planning on doing a lot more split batches moving forward, so being able to track multiple ferments at the same time would be very nice.

Whereas, with two TiltBridges receiving BLE signals, I wouldn't have that issue.

 

[mper]

@Thorrak: Yes, using an iSpindel with an ESP32 C3 Mini to emulate a Bluetooth Tilt signal to send to a TiltBridge (GravityMon.com - open source for beer brewing). I've been trying to figure out a way, as a non-programmer, to be able to have more than 1 iSpindel repeater sending data to my Fermentrack setup on my regular WiFi network.

Here's my thinking. I occasionally have trouble getting iSpindel WiFi signals out of my SS fermenters. There's the off the shelf ESP8266 repeater, as well as iSpindHub, Spindelmate, and probably a couple of other WiFi-based repeaters that are designed to help with this. None of them currently offer Fermentrack as a push target. So instead, I'd have to connect my Pi to the AP of the repeater. Not the end of the world by any means, but a minor annoyance. The other issue is if I connect the Pi to one repeater AP, that means I can't use a repeater with the other iSpindel. I'm planning on doing a lot more split batches moving forward, so being able to track multiple ferments at the same time would be very nice.

Whereas, with two TiltBridges receiving BLE signals, I wouldn't have that issue.

Yes that should work. if you have the v1 of the c3 i would not do it due to the faulty transmitter (can handle full power)

 

[DuncB]

I don't know enough about fermentrack being able to monitor or action the signals from more than one ispindel.

A simple Martin ger WiFi repeater made with a D1 mini with a supplementary aerial soldered on or the D1 mini with external antenna is simple and can handle multiple WiFi inputs. The source beyond this you determine on the ispindel itself.
Would Bluetooth penetrate or exit SS fermenters better than WiFi?

 

[mper]

I don't know enough about fermentrack being able to monitor or action the signals from more than one ispindel.

A simple Martin ger WiFi repeater made with a D1 mini with a supplementary aerial soldered on or the D1 mini with external antenna is simple and can handle multiple WiFi inputs. The source beyond this you determine on the ispindel itself.
Would Bluetooth penetrate or exit SS fermenters better than WiFi?

Dont know if there is a difference in coverage but the ble signal is much simpler and quicker to send so its probably not that sensitive for interference. A simple wifi extender should do the same job as a d1 mini för relaying the wifi signal.

 

[larsh]

Dont know if there is a difference in coverage but the ble signal is much simpler and quicker to send so its probably not that sensitive for interference. A simple wifi extender should do the same job as a d1 mini för relaying the wifi signal.

If of any interest:
I use a tp-link EAP110-Outdoor AP.
High-power amplifier and good antennas.
The ESP32-s2 mini has never had problems connecting through the SS fermenter inside a fridge. Typically -60dBm.
The AP stands around half a meter from the fridge.
And the s2's battery seems to stay on 4.09V forever - well, at least 3-4 weeks