Counter increments by 2 for every pulse #7488

robin13 · 2020-01-11T21:20:31Z

PROBLEM DESCRIPTION

I have been trying to get a tcrt5000 to work as a reader for the rotating disk on the household power meter. This has returns a high signal while most of the disk is rotating, and every time the red marked spot passes under the sensor it goes low.
Problem: for every pulse that the sensor sends, I get the counter incrementing by two.
Setup:
D4 / GPIO2 is configured as a Counter1
Counter relevant commands set:

Backlog CounterDebounce 100; CounterType 0; Counter1 0

I was wondering if this might be an issue with the TCRT5000 module, so I set up a simple push-button on a breadboard, and I can get the same effect with this as well.
BUT
With the manual button I can click significantly faster than the dial on the power meter passes (but less than the 100ms debounce time) - if I click very fast, I get the correct single increment of the counter, but if I click slowly, holding the button for (I guess) more than half a second I get two counts.
It feels to me as if counter works, as long as the pulses are very short, but not if they are a bit slower...

REQUESTED INFORMATION

Make sure your have performed every step and checked the applicable boxes before submitting your issue. Thank you!

22:14:51 MQT: stat/tasmota-test-01/RESULT = {"NAME":"D1 Mini","GPIO":[255,255,255,255,255,255,255,255,255,255,255,255,255],"FLAG":15,"BASE":18}
22:14:51 MQT: stat/tasmota-test-01/RESULT = {"Module":{"0":"D1 Mini"}}
22:14:52 MQT: stat/tasmota-test-01/RESULT = {"GPIO0":{"0":"None"},"GPIO1":{"0":"None"},"GPIO2":{"42":"Counter1"},"GPIO3":{"0":"None"},"GPIO4":{"0":"None"},"GPIO5":{"0":"None"},"GPIO9":{"0":"None"},"GPIO10":{"0":"None"},"GPIO12":{"0":"None"},"GPIO13":{"0":"None"},"GPIO14":{"0":"None"},"GPIO15":{"37":"PWM1"},"GPIO16":{"0":"None"}}

Provide the output of this command: Status 0:

22:15:22 MQT: stat/tasmota-test-01/STATUS = {"Status":{"Module":0,"FriendlyName":["tasmota-test-01"],"Topic":"tasmota-test-01","ButtonTopic":"0","Power":0,"PowerOnState":3,"LedState":1,"LedMask":"FFFF","SaveData":1,"SaveState":1,"SwitchTopic":"0","SwitchMode":[0,0,0,0,0,0,0,0],"ButtonRetain":0,"SwitchRetain":0,"SensorRetain":0,"PowerRetain":0}}
22:15:22 MQT: stat/tasmota-test-01/STATUS1 = {"StatusPRM":{"Baudrate":115200,"SerialConfig":"8N1","GroupTopic":"tasmotas","OtaUrl":"http://thehackbox.org/tasmota/release/tasmota.bin","RestartReason":"Software/System restart","Uptime":"0T00:00:48","StartupUTC":"2020-01-11T21:14:34","Sleep":50,"CfgHolder":4617,"BootCount":14,"SaveCount":163,"SaveAddress":"F9000"}}
22:15:22 MQT: stat/tasmota-test-01/STATUS2 = {"StatusFWR":{"Version":"8.1.0.3(tasmota)","BuildDateTime":"2020-01-10T16:42:41","Boot":31,"Core":"2_6_1","SDK":"2.2.2-dev(38a443e)","Hardware":"ESP8266EX","CR":"360/699"}}
22:15:22 MQT: stat/tasmota-test-01/STATUS3 = {"StatusLOG":{"SerialLog":2,"WebLog":2,"MqttLog":0,"SysLog":0,"LogHost":"","LogPort":514,"SSId":["cundr_iot",""],"TelePeriod":300,"Resolution":"558180C0","SetOption":["00008009","2805C8000100060000005A00000000000000","00000000","00000000"]}}
22:15:22 MQT: stat/tasmota-test-01/STATUS4 = {"StatusMEM":{"ProgramSize":568,"Free":432,"Heap":24,"ProgramFlashSize":1024,"FlashSize":4096,"FlashChipId":"1640EF","FlashMode":3,"Features":["00000809","8FDAE397","043683A1","22B617CD","01001BC0","00007881"],"Drivers":"1,2,3,4,5,6,7,8,9,10,12,16,18,19,20,21,22,24,26,29","Sensors":"1,2,3,4,5,6,7,8,9,10,14,15,17,18,20,22,26,34"}}
22:15:23 MQT: stat/tasmota-test-01/STATUS5 = {"StatusNET":{"Hostname":"tasmota-test-01-5541","IPAddress":"10.10.10.152","Gateway":"10.10.10.1","Subnetmask":"255.255.255.0","DNSServer":"10.10.10.1","Mac":"A4:CF:12:D5:D5:A5","Webserver":2,"WifiConfig":4,"WifiPower":17.0}}
22:15:23 MQT: stat/tasmota-test-01/STATUS6 = {"StatusMQT":{"MqttHost":"hassio.local","MqttPort":1883,"MqttClientMask":"DVES_%06X","MqttClient":"DVES_D5D5A5","MqttUser":"hassio","MqttCount":1,"MAX_PACKET_SIZE":1000,"KEEPALIVE":30}}
22:15:23 MQT: stat/tasmota-test-01/STATUS7 = {"StatusTIM":{"UTC":"Sat Jan 11 21:15:23 2020","Local":"Sat Jan 11 22:15:23 2020","StartDST":"Sun Mar 29 02:00:00 2020","EndDST":"Sun Oct 25 03:00:00 2020","Timezone":"+01:00","Sunrise":"08:41","Sunset":"17:14"}}
22:15:23 MQT: stat/tasmota-test-01/STATUS10 = {"StatusSNS":{"Time":"2020-01-11T22:15:23","COUNTER":{"C1":20}}}
22:15:23 MQT: stat/tasmota-test-01/STATUS11 = {"StatusSTS":{"Time":"2020-01-11T22:15:23","Uptime":"0T00:00:49","UptimeSec":49,"Heap":24,"SleepMode":"Dynamic","Sleep":50,"LoadAvg":19,"MqttCount":1,"POWER":"OFF","Dimmer":31,"Fade":"OFF","Speed":1,"LedTable":"ON","Wifi":{"AP":1,"SSId":"cundr_iot","BSSId":"02:9F:C2:F7:C1:63","Channel":1,"RSSI":64,"Signal":-68,"LinkCount":1,"Downtime":"0T00:00:07"}}}

TO REPRODUCE

As described above

EXPECTED BEHAVIOUR

I would expect the counter to increment exactly +1 per pulse regardless of the length of the pulse

The text was updated successfully, but these errors were encountered:

to-scho · 2020-01-15T17:43:43Z

Seems that counter debounce time for less than 200ms does not work reliably. Have you tried larger values.

ascillato2 · 2020-01-22T14:07:29Z

Hi,

Any news on this? Have you tried a higher counterdebounce? Have you tried the gpio as counter1n?

robin13 · 2020-01-22T20:16:05Z

I have tried setting CounterDebounce to 200, but no change.
I have also tried Counter1n but this also has no effect.

to-scho · 2020-01-27T16:20:13Z

I had same 2x increment with on board push-button of a sonoff basic. At my place it was gone with CounterDebounce 200.

If you are willing to debug further:

-You need to build your own binary after altering some code
-In file tasmota/xsns_01_counter.ino add the following single line after line 55 of current development branch: AddLog_P2(LOG_LEVEL_INFO, PSTR("COUNTER: Update of counter %d @ %d us, last update @ %d us, debounce time is %d us"), index, time, Counter.timer[index], debounce_time);
-Get log from console when 2x inc. happens
-Log should state system timings in us when counter change. Any counter increment should only happen when debounce time in log message is greater than CounterDebounce*1000.
-From that I found that CounterDebounce 200 (in ms) solve my issue.

Here is how updated tasmota/xsns_01_counter.ino looks like around line 55:
void CounterUpdate(uint8_t index)
{
uint32_t time = micros();
uint32_t debounce_time = time - Counter.timer[index];
if (debounce_time > Settings.pulse_counter_debounce * 1000) {
AddLog_P2(LOG_LEVEL_INFO, PSTR("COUNTER: Update of counter %d @ %d us, last update @ %d us, debounce time is %d us"), index, time, Counter.timer[index], debounce_time);
Counter.timer[index] = time;
if (bitRead(Settings.pulse_counter_type, index)) {
...

robin13 · 2020-02-12T09:59:45Z

Thank you for the debugging help @to-scho

I tried what you said, and this is the log output resulting:

10:54:08 CMD: Counter1 0
10:54:08 MQT: stat/tasmota-test-01/RESULT = {"Counter1":0}
10:54:14 COUNTER: Update of counter 0 @ 38263316 us, last update @ 0 us, debounce time is 38263316 us
10:54:16 COUNTER: Update of counter 0 @ 39895150 us, last update @ 38263316 us, debounce time is 1631834 us
10:54:53 COUNTER: Update of counter 0 @ 76808967 us, last update @ 39895150 us, debounce time is 36913817 us
10:54:53 COUNTER: Update of counter 0 @ 77203754 us, last update @ 76808967 us, debounce time is 394787 us
10:55:16 CMD: CounterDebounce
10:55:16 MQT: stat/tasmota-test-01/RESULT = {"CounterDebounce":100}
10:56:44 COUNTER: Update of counter 0 @ 187480303 us, last update @ 77203754 us, debounce time is 110276549 us
10:56:45 COUNTER: Update of counter 0 @ 189101788 us, last update @ 187480303 us, debounce time is 1621485 us
10:57:41 CMD: CounterDebounce 200
10:57:41 MQT: stat/tasmota-test-01/RESULT = {"CounterDebounce":200}
10:57:50 COUNTER: Update of counter 0 @ 253368739 us, last update @ 189101788 us, debounce time is 64266951 us
10:57:50 COUNTER: Update of counter 0 @ 254286604 us, last update @ 253368739 us, debounce time is 917865 us
10:58:09 CMD: Counter1
10:58:09 MQT: stat/tasmota-test-01/RESULT = {"Counter1":8}

As you can see, for each passing of the stripe there are two counts. I tried with CounterDebounce set to 100 and 200 with no difference.

robin13 · 2020-02-12T11:07:19Z

I've done some more investigation - made these changes to xsns_01_counter.ino:

void CounterUpdate(uint8_t index)
{
    uint32_t time = micros();
    uint32_t debounce_time = time - Counter.timer[index];
    if (debounce_time > Settings.pulse_counter_debounce * 1000) {
        AddLog_P2(LOG_LEVEL_INFO, PSTR("COUNTER: Update of counter %d @ %d us, last update @ %d us, debounce time is %d us"), index + 1, time, Counter.timer[index], debounce_time);
        Counter.timer[index] = time;
        if (bitRead(Settings.pulse_counter_type, index)) {
            RtcSettings.pulse_counter[index] = debounce_time;
        } else {
            RtcSettings.pulse_counter[index]++;
        }
    }
    AddLog_P2(LOG_LEVEL_INFO, PSTR("COUNTER %d @ %d us, : %d"), index + 1, time, RtcSettings.pulse_counter[index] );
}

And from this I'm seeing that there are a lot of triggers to CounterUpdate:

11:48:32 COUNTER 1 @ 994830147 us, : 1
11:48:32 COUNTER 1 @ 994834169 us, : 1
11:48:32 COUNTER: Update of counter 1 @ 995122268 us, last update @ 994815314 us, debounce time is 306954 us
11:48:32 COUNTER 1 @ 995122268 us, : 2
11:48:32 COUNTER 1 @ 995126952 us, : 2
11:48:32 COUNTER 1 @ 995130062 us, : 2
11:48:32 COUNTER 1 @ 995140078 us, : 2
11:48:36 COUNTER: Update of counter 1 @ 998260029 us, last update @ 995122268 us, debounce time is 3137761 us
11:48:36 COUNTER 1 @ 998260029 us, : 3
11:48:36 COUNTER 1 @ 998270039 us, : 3
11:48:36 COUNTER 1 @ 998280400 us, : 3
11:48:36 COUNTER: Update of counter 1 @ 998580402 us, last update @ 998260029 us, debounce time is 320373 us
11:48:36 COUNTER 1 @ 998580402 us, : 4
11:48:36 COUNTER 1 @ 998584910 us, : 4
11:48:36 COUNTER 1 @ 998590050 us, : 4
11:48:36 COUNTER 1 @ 998600040 us, : 4
11:48:39 COUNTER: Update of counter 1 @ 1002210044 us, last update @ 998580402 us, debounce time is 3629642 us
11:48:39 COUNTER 1 @ 1002210044 us, : 5
11:48:39 COUNTER 1 @ 1002220031 us, : 5
11:48:39 COUNTER 1 @ 1002230062 us, : 5
11:48:39 COUNTER 1 @ 1002238069 us, : 5
11:48:40 COUNTER: Update of counter 1 @ 1002596283 us, last update @ 1002210044 us, debounce time is 386239 us
11:48:40 COUNTER 1 @ 1002596283 us, : 6
11:48:40 COUNTER 1 @ 1002600758 us, : 6
11:48:40 COUNTER 1 @ 1002610046 us, : 6
11:48:40 COUNTER 1 @ 1002620046 us, : 6
11:48:40 COUNTER 1 @ 1002629895 us, : 6
11:48:40 COUNTER 1 @ 1002631403 us, : 6
11:48:44 COUNTER: Update of counter 1 @ 1006260047 us, last update @ 1002596283 us, debounce time is 3663764 us
11:48:44 COUNTER 1 @ 1006260047 us, : 7
11:48:44 COUNTER 1 @ 1006270055 us, : 7
11:48:44 COUNTER 1 @ 1006277432 us, : 7
11:48:44 COUNTER: Update of counter 1 @ 1006576877 us, last update @ 1006260047 us, debounce time is 316830 us
11:48:44 COUNTER 1 @ 1006576877 us, : 8
11:48:44 COUNTER 1 @ 1006581614 us, : 8
11:48:44 COUNTER 1 @ 1006590048 us, : 8
11:48:44 COUNTER 1 @ 1006599321 us, : 8
11:48:44 COUNTER 1 @ 1006601158 us, : 8

So it seems the issue is not in the code, but rather in getting a clean signal...

I suspect it is doing something like this:

where T1 and T2 are the first and second triggers respectively. The challenge here is that the time between the signal going low/high can be either quite fast (disk spinning fast) or slow (very little energy used), so I cannot see any good value for CounterDebounce which would work for all situations... :-/

I'll close this issue as I think it is not related to the tasmota code, but if you have any other ideas how to work around this I'd be very grateful!

arendst · 2020-02-12T14:42:13Z

Thx for your investigation.

I saw these spikes also when I had these "old" energy meters. Extending debounce time did solve it partly with a chance of missing a quick rotation.

Time to get a "Smart energy meter" ;-)

includes Arduino backport arendst#7488 backport arendst#7487 backport arendst#7486 backport arendst#7464 backport arendst#7434 backport arendst#7433 Security fix and malloc fix

fluppie · 2021-02-09T00:26:47Z

Could a simple capacitor "fix" this debounce issue? I see the same "problem" on a old 1994 Schlumberger gas meter and the TCR5000. I'm running Tasmota 9.2.0 release.

fluppie · 2021-02-09T00:40:14Z

With CounterDebounce 500 / CounterDebounceLow 50 / CounterDebounceHigh 50 it's working as expected :-)!

DST73 · 2025-01-23T08:11:40Z

I have to use these values. With the values above the counter counts about 4 or 5 times to fast

=>savedata 1
=>Teleperiod 30
=>CounterDebounce 12000
=>CounterDebounceLow 500
=>CounterDebounceHigh 500

but over the time it counts to low.

robin13 changed the title ~~Counter increments two~~ Counter increments by 2 for every pulse Jan 11, 2020

ascillato2 added awaiting feedback Action - Waiting for response or more information troubleshooting Type - Troubleshooting labels Jan 16, 2020

robin13 closed this as completed Feb 12, 2020

to-scho mentioned this issue Mar 29, 2020

Enhanced counter debouncing #8021

Merged

6 tasks

Jason2866 mentioned this issue Jul 27, 2020

Tasmota Stage Core 2.7.3.2 #8983

Merged

6 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Counter increments by 2 for every pulse #7488

Counter increments by 2 for every pulse #7488

robin13 commented Jan 11, 2020 •

edited

Loading

to-scho commented Jan 15, 2020

ascillato2 commented Jan 22, 2020

robin13 commented Jan 22, 2020

to-scho commented Jan 27, 2020

robin13 commented Feb 12, 2020

robin13 commented Feb 12, 2020

arendst commented Feb 12, 2020

fluppie commented Feb 9, 2021

fluppie commented Feb 9, 2021

DST73 commented Jan 23, 2025

Counter increments by 2 for every pulse #7488

Counter increments by 2 for every pulse #7488

Comments

robin13 commented Jan 11, 2020 • edited Loading

PROBLEM DESCRIPTION

REQUESTED INFORMATION

TO REPRODUCE

EXPECTED BEHAVIOUR

to-scho commented Jan 15, 2020

ascillato2 commented Jan 22, 2020

robin13 commented Jan 22, 2020

to-scho commented Jan 27, 2020

robin13 commented Feb 12, 2020

robin13 commented Feb 12, 2020

arendst commented Feb 12, 2020

fluppie commented Feb 9, 2021

fluppie commented Feb 9, 2021

DST73 commented Jan 23, 2025

robin13 commented Jan 11, 2020 •

edited

Loading