Problem during the digital number change

[MacSchlack]

Hello,
thank you for your project.
I have the problem that the Value is jumping a lot if during the digitla number ist changeing.

Now i work with allow negativ values but i will be better if the system check the analog values and only change if the analog1 change from 9 to 0.

[jomjol]

Did you switch on the parameter CheckDigitIncreaseConsistency by setting to true (Expert Mode). This should prevent an too early switch of the first digit.

[Voggl93]

Hello everyone,

I seem to have the Same Problem.
I've enabled the CkeckDigitIncreaseConsistency Thing but after that the readout is absolutely wrong.

If i've a look at the recognition its absolutely correct but i do get an Error Message saying the rate ist to high (Picture attached)

Is there Something to reset or to Change?

[haverland]

Hello everyone,

I seem to have the Same Problem. I've enabled the CkeckDigitIncreaseConsistency Thing but after that the readout is absolutely wrong.

Looks like you use the 11.3.1 version, not 12.0.1? It fixed in version 12.

[Voggl93]

Yes i actually am in 11.3.1.
Going to Upgrade to 12 asap

Thanks a lot

[mad2xlc]

Hello,
I have the same problem that the first number counts up too early. Between x.7 and x.8 the first digit increments and results in a "rate too high".
For example, a previous value of 2592.8786 becomes 2593.8839 in the next reading.

I tried the last versions beginning with 11.x, 12.x and currently with a "nightly" build. Which is way more stable than 12.0.1 :)

I changed the camera distance from 10cm to 5cm. Light from the side with ws2812b leds is ok. Normally I get good reading results.
CkeckDigitIncreaseConsistency is already set to True.

[mad2xlc]

Next 1000 liters, next rate too high
Any suggestions, please?

2022-10-29T12:15:28 task_autodoFlow - next round - Round #26
2022-10-29T12:15:28 FlowControll.doFlow - ClassFlowMakeImage Heap: 3037526 Min Free: 1003222 larg. Block: 3014656 SPI Heap: 3029959 NOT_SPI Heap: 7567 Internal Heap: 7567 Internal Min Heap free: 1875
2022-10-29T12:15:36 FlowControll.doFlow - ClassFlowAlignment Heap: 3037510 Min Free: 1003222 larg. Block: 3014656 SPI Heap: 3029959 NOT_SPI Heap: 7551 Internal Heap: 7551 Internal Min Heap free: 1875
2022-10-29T12:15:49 FlowControll.doFlow - ClassFlowCNNGeneral Heap: 3037486 Min Free: 1003222 larg. Block: 3014656 SPI Heap: 3029959 NOT_SPI Heap: 7527 Internal Heap: 7527 Internal Min Heap free: 1875
2022-10-29T12:15:49 ClassFlowCNNGeneral::doFlow nach Alignment
2022-10-29T12:15:57 Nach Invoke
2022-10-29T12:15:58 _num (p, m): 2 3 1 _val (p, m): 0.997770 0.000148 0.002082 result: 1.997918 _fit: 0.999851
2022-10-29T12:16:03 Nach Invoke
2022-10-29T12:16:03 _num (p, m): 5 6 4 _val (p, m): 0.998630 0.000722 0.000637 result: 5.000722 _fit: 0.999352
2022-10-29T12:16:09 Nach Invoke
2022-10-29T12:16:09 _num (p, m): 9 0 8 _val (p, m): 0.988308 0.000013 0.011669 result: 8.988331 _fit: 0.999977
2022-10-29T12:16:15 Nach Invoke
2022-10-29T12:16:15 _num (p, m): 4 5 3 _val (p, m): 0.812871 0.000002 0.187126 result: 3.812873 _fit: 0.999998
2022-10-29T12:16:15 FlowControll.doFlow - ClassFlowCNNGeneral Heap: 3037490 Min Free: 1003222 larg. Block: 3014656 SPI Heap: 3029959 NOT_SPI Heap: 7531 Internal Heap: 7531 Internal Min Heap free: 1875
2022-10-29T12:16:15 ClassFlowCNNGeneral::doFlow nach Alignment
2022-10-29T12:16:17 Nach Invoke
2022-10-29T12:16:17 Nach Invoke
2022-10-29T12:16:18 Nach Invoke
2022-10-29T12:16:19 Nach Invoke
2022-10-29T12:16:19 FlowControll.doFlow - ClassFlowPostProcessing Heap: 3037490 Min Free: 1003222 larg. Block: 3014656 SPI Heap: 3029959 NOT_SPI Heap: 7531 Internal Heap: 7531 Internal Min Heap free: 1875
2022-10-29T12:16:19 ClassFlowCNNGeneral::getReadout _analog=0, _extendedResolution=0, prev=-1
2022-10-29T12:16:19 ClassFlowCNNGeneral::ZeigerEvalAnalogNeu - kein Vorgänger - Ergebnis = 4 zahl: 4.425382 ziffer_vorgaenger = -1 AnalogFehler = 3.000000
2022-10-29T12:16:19 ClassFlowCNNGeneral::ZeigerEvalAnalogNeu - Zahl eindeutig, keine Korrektur notwendig - Ergebnis = 4 zahl: 4.502664 ziffer_vorgaenger = 4 AnalogFehler = 3.000000
2022-10-29T12:16:19 ClassFlowCNNGeneral::ZeigerEvalAnalogNeu - Zahl eindeutig, keine Korrektur notwendig - Ergebnis = 5 zahl: 5.342636 ziffer_vorgaenger = 4 AnalogFehler = 3.000000
2022-10-29T12:16:19 ClassFlowCNNGeneral::ZeigerEvalAnalogNeu - Zahl eindeutig, keine Korrektur notwendig - Ergebnis = 8 zahl: 8.324225 ziffer_vorgaenger = 5 AnalogFehler = 3.000000
2022-10-29T12:16:19 ClassFlowCNNGeneral::getReadout _analog=0, _extendedResolution=0, prev=8
2022-10-29T12:16:19 ClassFlowCNNGeneral::ZeigerEvalAnalogToDigitNeu - digitaleUnschaerfe - Ergebnis = 4 zahl: 3.812873 ziffer_vorgaenger: 8.324225 erg_vorkomma: 4 erg_nachkomma: 0
2022-10-29T12:16:19 ClassFlowCNNGeneral::ZeigerEvalHybridNeu - Analoger Vorgänger, Bewertung über ZeigerEvalAnalogNeu = 4 zahl: 3.812873 zahl_vorgaenger = 8.324225 eval_vorgaenger = 8 DigitalUnschaerfe = 0.200000
2022-10-29T12:16:19 ClassFlowCNNGeneral::getReadout(dig100) prev=4
2022-10-29T12:16:19 ClassFlowCNNGeneral::ZeigerEvalHybridNeu - KEIN Analoger Vorgänger, kein Ziffernwechsel, da Vorkomma weit genug weg = 9 zahl: 8.988331 zahl_vorgaenger = 3.812873 eval_vorgaenger = 4 DigitalUnschaerfe = 0.200000
2022-10-29T12:16:19 ClassFlowCNNGeneral::getReadout#ZeigerEvalHybridNeu()= 9
2022-10-29T12:16:19 ClassFlowCNNGeneral::getReadout#result= 94
2022-10-29T12:16:19 ClassFlowCNNGeneral::ZeigerEvalHybridNeu - KEIN Analoger Vorgänger, kein Ziffernwechsel, da Vorkomma weit genug weg = 5 zahl: 5.000722 zahl_vorgaenger = 8.988331 eval_vorgaenger = 9 DigitalUnschaerfe = 0.200000
2022-10-29T12:16:19 ClassFlowCNNGeneral::getReadout#ZeigerEvalHybridNeu()= 5
2022-10-29T12:16:20 ClassFlowCNNGeneral::getReadout#result= 594
2022-10-29T12:16:20 ClassFlowCNNGeneral::ZeigerEvalHybridNeu - KEIN Analoger Vorgänger, kein Ziffernwechsel, da Vorkomma weit genug weg = 2 zahl: 1.997918 zahl_vorgaenger = 5.000722 eval_vorgaenger = 5 DigitalUnschaerfe = 0.200000
2022-10-29T12:16:20 ClassFlowCNNGeneral::getReadout#ZeigerEvalHybridNeu()= 2
2022-10-29T12:16:20 ClassFlowCNNGeneral::getReadout#result= 2594
2022-10-29T12:16:20 Data file writing temporarily disabled, see #1225
2022-10-29T12:16:20 FlowControll.doFlow - ClassFlowMQTT Heap: 3037458 Min Free: 1003222 larg. Block: 3014656 SPI Heap: 3029959 NOT_SPI Heap: 7499 Internal Heap: 7499 Internal Min Heap free: 1875
2022-10-29T12:16:20 MQTT - sent publish successful in MQTTPublish, msg_id=5856, wasserzaehler/connection, connected
2022-10-29T12:16:20 MQTT - sent publish successful in MQTTPublish, msg_id=65121, wasserzaehler/connection, connected
2022-10-29T12:16:20 MQTT - sent publish successful in MQTTPublish, msg_id=61521, wasserzaehler/uptime, 7334
2022-10-29T12:16:20 MQTT - sent publish successful in MQTTPublish, msg_id=38426, wasserzaehler/freeMem, 3034907
2022-10-29T12:16:20 MQTT - sent publish successful in MQTTPublish, msg_id=46539, wasserzaehler/wifiRSSI, -73
2022-10-29T12:16:20 MQTT - sent publish successful in MQTTPublish, msg_id=28326, wasserzaehler/CPUtemp, 47.777779
2022-10-29T12:16:20 MQTT - sent publish successful in MQTTPublish, msg_id=449, wasserzaehler/main/error, Rate too high - Read: 2594.8544 - Pre: 2593.8394
2022-10-29T12:16:20 MQTT - sent publish successful in MQTTPublish, msg_id=56135, wasserzaehler/main/changeabsolut, 0.0081
2022-10-29T12:16:20 MQTT - sent publish successful in MQTTPublish, msg_id=55041, wasserzaehler/main/raw, 2594.8544
2022-10-29T12:16:20 MQTT - sent publish successful in MQTTPublish, msg_id=39596, wasserzaehler/main/timestamp, 2022-10-29T12:10:45
2022-10-29T12:16:20 MQTT - sent publish successful in MQTTPublish, msg_id=31327, wasserzaehler/main/json, {"value":"","raw":"2594.8544","error":"Rate too high - Read: 2594.8544 - Pre: 2593.8394","rate":"","timestamp":"2022-10-29T12:10:45"}
2022-10-29T12:16:20 logfiles deleted: 0 files not deleted (incl. leer.txt): 8
2022-10-29T12:16:20 task_autodoFlow - round done
2022-10-29T12:16:20 CPU Temperature: 47.8

[mad2xlc]

With a value of 2593.9418 the CkeckDigitIncreaseConsistency setting works for me. But between x.7 and x.8 it does not work.

[haverland]

@mad2xlc CkeckDigitIncreaseConsistency is for dig-class11 only.
Your issue is solved in next release.

[mad2xlc]

@haverland is this already implemented in a current rolling?