Feature request: HOBO bluetooth sensors

It really depends which 433 MHz protocol, assuming you mean RTL_$33, then no, this doesn’t currently work in parallel with Bluetooth.
Even with other 433 MHz protocols I would always suggest, and do it myself, to have separate gateways, just for the fact of RF/WiFi blackout times during BT scanning.

I am not that actively involved with the Theengs App development, so I cannot confirm or deny the addition, but as the Mopeka level sensor is included there might be a chance. Either way, for a full inclusion it would be nice to extend the HOBO decoder with at least the battery level and possibly the temperature, if these are definitely in the advertising data.

It seems that temperature is still important, I use this sensor in the winter and the temperature goes below zero and it’s bad when the sensor freezes over.
I tried to find the temperature relationship but I guess I’m doing something wrong :frowning:
maybe the hex values ​​are in fahrenheit scale not celsius,because it is made in the USA
did linear regression analyze to hex values and there is something in 3 groups
some hex values are ~4600 most are ~37000 and some are ~~53000

when filter out values ~37000 it is linear

I also tested the battery voltages and found the battery value on the package
4 first blue numbers are temp and last 4 blue numbers are battery values
c500073fd54f0106240000000800903c 0265 d078b6ba
or
6 first blue numbers are temp and last 2 blue numbers are battery values
c500073fd54f0106240000000800903c02 65 d078b6ba

I noticed that if the temp is over 51 then the blue 5-6 number is 00
if the temp is over 25 then the 5-6 number is 01
if the temp is under 25 then the 5-6 number is 02

c500073fd54f0106240000000800903c0265d078b6ba  temp 19,11   voltage 1,7v   bat indicator 0 logger stopped
c500073fd54f0106240000000800123c0268283dd1ba  temp 19,11   voltage 1,7v   bat indicator 0
c500073fd54f0106240000000800123c0268283dd1ba  temp 19,11   voltage 2,0v   bat indicator 1
c500073fd54f0106240200000800d23c027000f4aab9               voltage 2,2v   bat indicator 2
c500073fd54f0106240200000800523b027c0047f5b9  temp 19,21   voltage 2,4v   bat indicator 3
c500073fd54f0106240200000800d23b0288004e19b8  temp 19,21   voltage 2,6v   bat indicator 4
c500073fd54f0106240000000800523b0295c0135a39  temp 19,21   voltage 2,9v   bat indicator 5
c500073fd54f0106240200000800123b029960e3ec39  temp 19,21   voltage 3,0v   bat indicator 6
c500073fd54f0106240000000800123b02a1d0d310ba  temp 19,21   voltage 3,15v  bat indicator 6
c500073fd54f0106240200000800d24f02a2c0c0a6ba  temp 17,30   voltage 3,3v   bat indicator 6
c500073fd54f0106240000000800d27002a200cf1e39  temp 17,30   voltage 3,4v   bat indicator 6
c500073fd54f0106240000000800d27002a200cf1e39  temp 17,30   voltage 3,5v   bat indicator 6
c500073fd54f0106240200000800529102a2f03a45ba  temp 10,96   voltage 3,1v   bat indicator 6 

c500073fd54f0106240000000800d2330291b012ad3a  temp 19,88
c500073fd54f0106240200000800d2320291c0976fb9  temp 19,97
c500073fd54f0106240200000800d27602904cada23b  temp  13,57
c500073fd54f0106240200000800d28602903646883b  temp 12,03
c500073fd54f010624020000080092b70290787c553b  temp 8,39
c500073fd54f010624020000080092c60290cce3293b  temp 5,67
c500073fd54f010624020000080092cf029090401d3b  temp 4,74
c500073fd54f010624020000080092d60290b487093b  temp 4,01
c500073fd54f010624020000080092e102902044d73c  temp 1,01
c500073fd54f010624020000080092ab0290a5e0c4bc  temp 8,39
c500073fd54f0106240000000800923302901fabcb3d  temp 19,97
c500073fd54f0106240200000800923302901fabcb3d  temp 19,97
c500073fd54f0106240000000800926a0190e72e893c  temp 40,47
c500073fd54f010624020000080092110190a8bad03a  temp 51,80
c500073fd54f010624020000080092d4009048e33d3d  temp 57 58
c500073fd54f010624020000080092de009090e96d3d  temp 59,75
c500073fd54f010624000000080092330190b03a673d  temp 50,81
c500073fd54f010624020000080092420190f6f0623d  temp 46,53
c500073fd54f0106240200000800d2860190143862bc  temp 37,32
c500073fd54f0106240200000800d299019016105bbc  temp 35,27
c500073fd54f010624000000080012d701904a6d50bc  temp 32,13
c500073fd54f010624000000080012ea019060725fbc  temp 27,01
c500073fd54f010624000000080052f70190d2114ebc  temp 25,74
c500073fd54f010624020000080092010290e27f48bc  temp 24,77
c500073fd54f01062400000008009204029008794dbc  temp24,48
c500073fd54f0106240200000800120a029028c741bc  temp 23,90
c500073fd54f010624000000080012150290500836bc  temp 22,84
c500073fd54f0106240000000800122302900c9617bc  temp 21,50
c500073fd54f0106240200000800122b0290601e2ebc  temp 20,73
c500073fd54f0106240000000800924c0290a0d0fbbb  temp 17,59
c500073fd54f0106240200000800125902902029cdbb  temp 16,35
c500073fd54f0106240200000800125c029030acdbbb  temp 16,06
c500073fd54f0106240200000800525f029030c5e2bb  temp 15,78
c500073fd54f01062402000008001265029004a6a3bb  temp 15,20
c500073fd54f01062400000008001267028e245dc8bb  temp 15,01
c500073fd54f01062402000008001269028e00b8aabb  temp 14,82
c500073fd54f0106240000000800526b028e9ce8c5bb  temp 14,63
c500073fd54f0106240200000800526f028e3c80b6bb  temp 14,34
c500073fd54f01062402000008001272028e08074dbb  temp 13,96
c500073fd54f01062400000008001273028e08f237bb  temp 13,86
c500073fd54f0106240000000800d273028fecaf52bb  temp 13,86
c500073fd54f0106240000000800d273028ffc8046bb  temp 13,86
c500073fd54f01062402000008009274028fec418abb  temp 13.77
c500073fd54f01062402000008009274028f105da8bb  temp 13.77
c500073fd54f01062402000008005269028f6879b5bb  temp 15,11
c500073fd54f01062400000008009269028f6879b5bb  temp 15,30
c500073fd54f0106240000000800d269028f6879b5bb  temp  15,59 
c500073fd54f0106240200000800d234028f48de70bb  temp 19,88
c500073fd54f01062400000008005235028900f6be38  temp 19,78

I’m afraid I haven’t had much time to look into this, but the temperature still eludes me so far.

If anyone else sees it encoded in the manufacturerdata, feel free to chip ion :slight_smile:

I gave AI and got an answer. most are OK, maybe some numbers were my own fault

Okay, let me start by understanding the user's request. They want a table that includes the hex code, their provided temperature, the calculated temperature using the formula, and the error between the two. The formula given is Temperature = -0.09 Γ— hex_value + 70.55, where hex_value is derived from the second and third bytes in little-endian.

First, I need to process each hex string. For example, take d23302. The second and third bytes are 33 and 02. In little-endian, that's 0x0233, which is 563 in decimal. Applying the formula: -0.09*563 +70.55 = 19.88, which matches the user's value. That's good.

Next, some hex codes start with 92, like 92b702. The bytes here are B7 and 02, which is 0x02B7 = 695. Using the formula: -0.09*695 +70.55 = 7.0, but the user's value is 8.39. There's a discrepancy here. Wait, maybe the formula isn't the same for all prefixes. The user mentioned that some entries, like those starting with 92, might use a different scaling. For example, 92de00: bytes DE and 00 make 0x00DE = 222. Using the formula gives 70.55 -0.09*222 = 50.57, but the actual temp is 59.75. That's a big error. So maybe these entries use a different formula, like direct scaling. For instance, 222 * 0.268 β‰ˆ 59.5, which is close. But the user wants to apply the original formula for all. Hmm.

Wait, the user provided some examples where the formula works for D2, 12, 52 prefixes but not for some 92. So maybe I should proceed with the formula as given, even if some entries have larger errors. The user wants to see the calculated temps and the errors, so I need to apply the formula to all and note discrepancies.

I need to process each line, extract the second and third bytes, convert to little-endian, compute the value, apply the formula, then compare with the user's temp. Let's start listing them one by one.

For instance, d27602: bytes 76 and 02. 0x0276 = 630. Formula: -0.09*630 +70.55 = 70.55 -56.7 =13.85. But the user's value is 13.57. Error is about 0.28. That's okay.

Another example, 92ab02: bytes AB and 02. 0x02AB = 683. Formula: 70.55 -0.09*683 = 70.55 -61.47 =9.08. User's temp is 8.39. Error 0.69.

Wait, but the user's temp for 92ab02 is 8.39, but calculated is 9.08. So error is 0.69. That's a bit off. Maybe there's a rounding difference or the formula isn't perfect for these.

For 926a01: bytes 6A and 01. Little-endian is 0x016A = 362. Formula: 70.55 -32.58 =37.97. User has 40.47. Error 2.5. That's significant. Maybe the formula isn't right here.

But the user wants to see all calculated temps with the formula given, even if there are errors. So proceed.

Now, some entries have commas as decimal separators, others periods. Need to standardize for calculation. Also, note that 92d400 has two temps: 57 58. Assume it's 57.58. But written as 57 58. Maybe a typo. Let's check. Original user input: "92d400  temp 57 58". Probably a typo, maybe 57.58. So take that as 57.58.

Proceeding line by line. For each hex code:

1. Split the hex into second and third bytes. For example, d23302: bytes 2 and 3 are '33' and '02'.

2. Convert to little-endian: reverse the bytes to '0233' (but wait, the bytes are in order 33 02, so little-endian is 0x0233).

3. Convert that to decimal: 563.

4. Apply formula: -0.09 *563 +70.55 = 19.88 (matches).

5. Error is 0.

Now, for hex codes starting with 92, like 92b702: bytes B7 02 β†’ 02B7 = 695. Formula: -0.09*695= -62.55 +70.55=8.00. User's temp is 8.39. Error: 8.39 -8.00=0.39. Wait, but in the previous example, the user's temp was 8.39, calculated is 8.00. So error is +0.39.

Wait, but earlier when I calculated 92ab02 (683), the formula gave 9.08, but user's temp is 8.39. So error -0.69. Hmm, inconsistencies.

But regardless, proceed. The user wants all calculated, even with errors.

Now, compiling all into a table. Need to be careful with each hex code, extract the bytes correctly, convert to little-endian, apply formula, compute error. Let me list all entries and process them step by step.

Let me start with the first few to check:

d23302 β†’ bytes 33 02 β†’ 0x0233 =563 β†’ -0.09*563= -50.67 +70.55=19.88 β†’ matches. Error 0.

d23202 β†’ 32 02 β†’0x0232=562 β†’ -50.58 +70.55=19.97 β†’ matches. Error 0.

d27602 β†’76 02 β†’0x0276=630 β†’-56.7 +70.55=13.85 β†’ user 13.57 β†’ error -0.28.

d28602 β†’86 02 β†’0x0286=646 β†’-58.14 +70.55=12.41 β†’ user 12.03 β†’ error -0.38.

92b702 β†’B7 02 β†’0x02B7=695 β†’-62.55 +70.55=8.00 β†’ user 8.39 β†’ error +0.39.

92c602 β†’C6 02 β†’0x02C6=710 β†’-63.9 +70.55=6.65 β†’ user 5.67 β†’ error -0.98.

92cf02 β†’CF 02 β†’0x02CF=719 β†’-64.71 +70.55=5.84 β†’ user 4.74 β†’ error -1.10.

92d602 β†’D6 02 β†’0x02D6=726 β†’-65.34 +70.55=5.21 β†’ user 4.01 β†’ error -1.20.

92e102 β†’E1 02 β†’0x02E1=737 β†’-66.33 +70.55=4.22 β†’ user 1.01 β†’ error -3.21. Wow, big error here.

92ab02 β†’AB 02 β†’0x02AB=683 β†’-61.47 +70.55=9.08 β†’ user 8.39 β†’ error -0.69.

923302 β†’33 02 β†’ same as d23302? Wait, hex is 923302. Wait, bytes are 33 02? Wait, the hex is 92 33 02. So second and third bytes are 33 02 β†’ same as before, 0x0233=563 β†’ same calculation:19.88. User's temp is 19.97. Wait, no. Wait, hex 923302: bytes are 33 and 02. So 0x0233=563. So formula gives 19.88, but user's temp is 19.97. So error +0.09. Wait, but in the first entry, d23302 also has 19.88. So why here, same hex value but different prefix (92 vs d2) leading to same calculation but user's temp is different? That's confusing. But according to the formula, same result. So this entry has a discrepancy. Maybe a typo in the user's data?

Similarly, the next entry is 923302 again, same calculation, user temp 19.97. Error +0.09.

926a01 β†’6A 01 β†’0x016A=362 β†’-32.58 +70.55=37.97 β†’ user 40.47 β†’ error +2.50.

921101 β†’11 01 β†’0x0111=273 β†’-24.57 +70.55=45.98 β†’ user 51.80 β†’ error +5.82.

92d400 β†’D4 00 β†’0x00D4=212 β†’-19.08 +70.55=51.47 β†’ user 57.58 (assuming typo) β†’ error +6.11.

92de00 β†’DE 00 β†’0x00DE=222 β†’-19.98 +70.55=50.57 β†’ user 59.75 β†’ error +9.18.

923301 β†’33 01 β†’0x0133=307 β†’-27.63 +70.55=42.92 β†’ user 50.81 β†’ error +7.89.

924201 β†’42 01 β†’0x0142=322 β†’-28.98 +70.55=41.57 β†’ user 46.53 β†’ error +4.96.

d28601 β†’86 01 β†’0x0186=390 β†’-35.1 +70.55=35.45 β†’ user 37.32 β†’ error +1.87.

d29901 β†’99 01 β†’0x0199=409 β†’-36.81 +70.55=33.74 β†’ user 35.27 β†’ error +1.53.

12d701 β†’D7 01 β†’0x01D7=471 β†’-42.39 +70.55=28.16 β†’ user 32.13 β†’ error +3.97.

12ea01 β†’EA 01 β†’0x01EA=490 β†’-44.1 +70.55=26.45 β†’ user 27.01 β†’ error +0.56.

52f701 β†’F7 01 β†’0x01F7=503 β†’-45.27 +70.55=25.28 β†’ user 25.74 β†’ error +0.46.

920102 β†’01 02 β†’0x0201=513 β†’-46.17 +70.55=24.38 β†’ user 24.77 β†’ error +0.39.

920402 β†’04 02 β†’0x0204=516 β†’-46.44 +70.55=24.11 β†’ user 24.48 β†’ error +0.37.

120a02 β†’0A 02 β†’0x020A=522 β†’-46.98 +70.55=23.57 β†’ user 23.90 β†’ error +0.33.

121502 β†’15 02 β†’0x0215=533 β†’-47.97 +70.55=22.58 β†’ user 22.84 β†’ error +0.26.

122302 β†’23 02 β†’0x0223=547 β†’-49.23 +70.55=21.32 β†’ user 21.50 β†’ error +0.18.

122b02 β†’2B 02 β†’0x022B=555 β†’-49.95 +70.55=20.60 β†’ user 20.73 β†’ error +0.13.

924c02 β†’4C 02 β†’0x024C=588 β†’-52.92 +70.55=17.63 β†’ user 17.59 β†’ error -0.04.

125902 β†’59 02 β†’0x0259=601 β†’-54.09 +70.55=16.46 β†’ user 16.35 β†’ error -0.11.

125c02 β†’5C 02 β†’0x025C=604 β†’-54.36 +70.55=16.19 β†’ user 16.06 β†’ error -0.13.

525f02 β†’5F 02 β†’0x025F=607 β†’-54.63 +70.55=15.92 β†’ user 15.78 β†’ error -0.14.

126502 β†’65 02 β†’0x0265=613 β†’-55.17 +70.55=15.38 β†’ user 15.20 β†’ error -0.18.

126702 β†’67 02 β†’0x0267=615 β†’-55.35 +70.55=15.20 β†’ user 15.01 β†’ error -0.19.

126902 β†’69 02 β†’0x0269=617 β†’-55.53 +70.55=15.02 β†’ user 14.82 β†’ error -0.20.

526b02 β†’6B 02 β†’0x026B=619 β†’-55.71 +70.55=14.84 β†’ user 14.63 β†’ error -0.21.

526f02 β†’6F 02 β†’0x026F=623 β†’-56.07 +70.55=14.48 β†’ user 14.34 β†’ error -0.14.

127202 β†’72 02 β†’0x0272=626 β†’-56.34 +70.55=14.21 β†’ user 13.96 β†’ error -0.25.

127302 β†’73 02 β†’0x0273=627 β†’-56.43 +70.55=14.12 β†’ user 13.86 β†’ error -0.26.

d27302 β†’73 02 β†’0x0273=627 β†’ same as above β†’14.12 vs user 13.86 β†’ error -0.26.

Same for the next d27302 β†’ same result.

927402 β†’74 02 β†’0x0274=628 β†’-56.52 +70.55=14.03 β†’ user 13.77 β†’ error -0.26.

Another 927402 β†’ same.

526902 β†’69 02 β†’0x0269=617 β†’-55.53 +70.55=15.02 β†’ user 15.11 β†’ error +0.09.

926902 β†’69 02 β†’ same as above β†’15.02 vs 15.30 β†’ error +0.28.

d26902 β†’69 02 β†’ same β†’15.02 vs 15.59 β†’ error +0.57.

d23402 β†’34 02 β†’0x0234=564 β†’-50.76 +70.55=19.79 β†’ user 19.88 β†’ error +0.09.

523502 β†’35 02 β†’0x0235=565 β†’-50.85 +70.55=19.70 β†’ user 19.78 β†’ error +0.08.

Now compiling all these into a table. Need to ensure that each entry is processed correctly. Also, note that some entries have multiple instances (like 923302 appearing twice), but the calculated temp varies when the formula is applied, depending on the hex code's bytes. Wait, no, in the case of 923302, the bytes are 33 and 02, same as d23302. But since the prefix is 92 instead of d2, but the formula uses only the second and third bytes, the calculated temp is the same. However, the user's temp for 923302 is 19.97, while for d23302 it's 19.88. So same hex value but different reported temps. This suggests that maybe the first byte (prefix) affects the calculation, but according to the initial instructions, the formula only uses the second and third bytes. This inconsistency indicates possible errors in the user's data or that the formula might not be universally applicable.

But proceeding as per the user's request to apply the formula regardless. The table will show these discrepancies. The user can then see where the formula works and where it doesn't.

Now, assembling all these calculations into a table. Each row will have Hex, Actual Temp, Calculated Temp, Error. Error is Actual - Calculated. But need to check decimal separators. Some temps use commas, others periods. Assume all commas are decimal points. For example, 19,88 is 19.88. Similarly, 13.77 is already with a period. So convert all to decimal points.

Hi @mortenx

I somehow think that the AI deducted temperature encoding might not be quite right :wink:

  1. Apply formula: -0.09 *563 +70.55 = 19.88 (matches).
  2. Error is 0.

as -0.09 *563 +70.55 would be quite an awkward encoding, unless its sole purpose was to make it as difficult as possible to reverse engineer. But even more so because the subsequent calculation with the same formula give such massively diverging results from your observed temperatures, e. g.

92de00 β†’DE 00 β†’0x00DE=222 β†’-19.98 +70.55=50.57 β†’ user 59.75 β†’ error +9.18.

923301 β†’33 01 β†’0x0133=307 β†’-27.63 +70.55=42.92 β†’ user 50.81 β†’ error +7.89.

924201 β†’42 01 β†’0x0142=322 β†’-28.98 +70.55=41.57 β†’ user 46.53 β†’ error +4.96.

with errors of up 9Β°. So for a single and even maybe two or three I also found matching algorithms for decoding, but unfortunately none which worked for all the samples you gave - or at least most of them to account for possible few wrong readings.

Could you possibly convert all your temperatures in your above list from Celsius to Fahrenheit and throw that list at your AI again, which might hopefully give us a constant formula, applicable to all samples with the same accuracy?

Then it is like that?

captured some new packets:

c5000744d54f0106240000000800526a0291a261b83d  14,73
c5000744d54f0106240200000800526a0291a261b83d  13,28
c5000744d54f0106240000000800528502912fe0a73d  12,12
c5000744d54f0106240000000800d2890291232cac3d  11,74
c5000744d54f0106240000000800528c0291a479aa3d  11,44
c5000744d54f0106240000000800d28f029116fba93d  11,15
c5000744d54f0106240000000800d2930291e02aa63d  10,76
c5000744d54f0106240200000800d2d801916007383b  28,79
c5000744d54f010624000000080052d901918037773b  28,69
c5000744d54f0106240200000800d2da019100f4763b  28,59
c5000744d54f010624000000080052db0191001b373b  28,49
c5000744d54f0106240000000800d2dd019100b12c3b  28,29
c5000744d54f010624020000080052e90191c085fd3b  27,11
c5000744d54f0106240200000800d2f40191b064f53b  26,03
c5000744d54f010624000000080052fb019120a2363c  25,35
c5000744d54f0106240200000800d2ff019180393d3c  24,96
c5000744d54f0106240200000800520202918833413c  24,67
c5000744d54f0106240200000800d2030291e8a91c3c  24,57
c5000744d54f010624000000080052040291e0c12c3c  24,48
c5000744d54f0106240000000800d2050291d879413c  24,38
c5000744d54f0106240200000800d2f80191fbc0953d  25,64
c5000744d54f0106240000000800d2f80191fbc0953d  19,97
c5000744d54f0106240000000800d2050291d879413c  27,30
c5000744d54f0106240000000800d2050291d879413c  34,63
c5000744d54f0106240000000800529e0191e09010bd  34,74
c5000744d54f0106240200000800528701912504ef3c  37,21
c5000744d54f0106240000000800d2770191ac7ad93c  38,99
c5000744d54f0106240000000800526c0191f753e23c  40,24
c5000744d54f0106240000000800d2660191096ce03c  40,93
c5000744d54f010624000000080052620191a1a2e33c  41,63
c5000744d54f0106240000000800525f01919fc4f13c  41,75

and i gave it to AI
converted to F
and get results:

Step 2: Hex Decoding Approach

  1. Hex Handling:
  • Reverse hex bytes (little-endian to big-endian):
    Example: d801 β†’ 01d8 (hex) = 472 (decimal).
  • Split into two groups based on observed patterns:
    • Lower Hex Range (Hex ≀ 500): Higher temperatures (94.33Β°F–106.93Β°F).
    • Higher Hex Range (Hex > 500): Lower temperatures (66.92Β°F–94.53Β°F).
  1. Linear Formulas:
  • Lower Hex Range (Hex ≀ 500):
    Temp (Β°F)=βˆ’0.104Γ—Hex+145.6Temp (Β°F)=βˆ’0.104Γ—Hex+145.6
  • Higher Hex Range (Hex > 500):
    Temp (Β°F)=βˆ’0.078Γ—Hex+119.2Temp (Β°F)=βˆ’0.078Γ—Hex+119.2

And with Celsius

I also tried with copilot. a few points are wrong, but maybe that was my own mistake in measuring
Is it possible to add a rule to the decoder so that if the result is outside this line, it is not displayed?

with corrections

and with F is formula
Temperature (F)=βˆ’0.1805Γ—HexValue+169.64

I still find this found formula a bit off, from my past experience with encoded hex data, but your curve does show quite well fitting data points with it :slight_smile:

With your marked battery byte there, could you post one or two more sample messages with the battery level value as shown in the app? Is it shown as 0-100% level, or with an actual voltage?

I have then prepared a test decoder for you to try out once the battery question has been sorted.

The app only has a battery indicator with 6 levels as I described in a previous post. no % numbers or voltage numbers.
I tested it by putting different voltage into the sensor and measuring the voltage same time and getting the change in hex.
Sensor is working with 2x AA battery , so max voltage is about 3v i think

c500073fd54f0106240000000800903c0265d078b6ba  temp 19,11   voltage 1,7v   bat indicator 0 logger stopped
c500073fd54f0106240000000800123c0268283dd1ba  temp 19,11   voltage 1,7v   bat indicator 0
c500073fd54f0106240000000800123c0268283dd1ba  temp 19,11   voltage 2,0v   bat indicator 1
c500073fd54f0106240200000800d23c027000f4aab9               voltage 2,2v   bat indicator 2
c500073fd54f0106240200000800523b027c0047f5b9  temp 19,21   voltage 2,4v   bat indicator 3
c500073fd54f0106240200000800d23b0288004e19b8  temp 19,21   voltage 2,6v   bat indicator 4
c500073fd54f0106240000000800523b0295c0135a39  temp 19,21   voltage 2,9v   bat indicator 5
c500073fd54f0106240200000800123b029960e3ec39  temp 19,21   voltage 3,0v   bat indicator 6
c500073fd54f0106240000000800123b02a1d0d310ba  temp 19,21   voltage 3,15v  bat indicator 6
c500073fd54f0106240200000800d24f02a2c0c0a6ba  temp 17,30   voltage 3,3v   bat indicator 6
c500073fd54f0106240000000800d27002a200cf1e39  temp 17,30   voltage 3,4v   bat indicator 6
c500073fd54f0106240000000800d27002a200cf1e39  temp 17,30   voltage 3,5v   bat indicator 6
c500073fd54f0106240200000800529102a2f03a45ba  temp 10,96   voltage 3,1v   bat indicator 6 

I also noticed now when the first numbers that come after c500 change, it seems like they could be the sensor ID, I saw in the app that each sensor has a different ID, I have several sensors that I have tested here. haven’t paid attention to it before

c5000744d54f0106240200000800d247028f4b0c373e    sensor ID:22009156
c5000740d54f01062402000008001229029d827f963e    sensor ID:22009152

I checked some more sensor IDs, but there are probably too few to decode.

c5000744d54f0106240200000800d247028f4b0c373e    ID:22009156
c5000740d54f01062402000008001229029d827f963e    iD:22009152
c500075aa4410106240000000800d23f02a39252b63f    id:21079130
c5000746d54f0106240200000800526c02a2537c233f    id:22009158
c5000745d54f0106240200000800924002a02fd93e3e    id:22009157
c5000742d54f0106240200000800124b02a15a94ca3d    id:22009154

With these identical data sets, but with different battery indicators, I don’t think the battery is included in the advertising data.

Also with the temperature I’m still thinking that the deducted algorithm is still not quite right, as it seems very convoluted, while the initial level was very straight forward. Nonetheless I will publish a test decoder for you to try out when I get back home.