ESP8266 Developer Zone

Re: Power consumption problem when the AP is not avaliable

2015-09-18T23:53:30+08:00

Espressif: thanks for the new force-sleep functions in SDK 1.4! Looks pretty awesome!

Question: if I enable force-sleep (wifi_fpm_open) does that disable automatic sleep? Or is the force-sleep in addition to the automatic sleep managed by the SDK?

Statistics: Posted by tve — Fri Sep 18, 2015 11:53 pm

Re: Power consumption problem when the AP is not avaliable

2015-09-18T20:04:05+08:00

So there will be a function that we can disable WiFi communication (i.e. power off the RF section?).
When will the next API be released?

Kind Regards

Daniel

Statistics: Posted by pumzele — Fri Sep 18, 2015 8:04 pm

Re: Power consumption problem when the AP is not avaliable

2015-09-17T15:46:59+08:00

Anyone know when the next API will be released?
I would really like the RF power off functionality.
I am using the ESP8266 in a low power application as the processor and the WiFi is just to set up/ download info. So it is only being used once in a while. 90% of the time nothing is able to connect to it.

Statistics: Posted by pumzele — Thu Sep 17, 2015 3:46 pm

Re: Power consumption problem when the AP is not avaliable

2015-09-16T23:52:56+08:00

@blubb: Probably yes, but to be honest I don't mind calling another function as long as I get the desired result.

Statistics: Posted by Npt — Wed Sep 16, 2015 11:52 pm

Re: Power consumption problem when the AP is not avaliable

2015-09-16T19:30:02+08:00

Should this not be done automatically? Closing RF in between reconnection attempts (if there are pauses inbetween)?

Statistics: Posted by blubb — Wed Sep 16, 2015 7:30 pm

Re: Power consumption problem when the AP is not avaliable

2015-09-16T17:08:01+08:00

Thank you very much! It's nice to see a company respond to users' needs.

Statistics: Posted by Npt — Wed Sep 16, 2015 5:08 pm

Re: Power consumption problem when the AP is not avaliable

2015-09-16T16:44:34+08:00

While a replay Cheap Fifa Coins or pre-match area is going on, the gamer designs along with the experience seem fantastic.

However, when they begin to FIFA 16 Coins make the things become vibrant, they are to end up to become a little unorganized.

Statistics: Posted by Guest — Wed Sep 16, 2015 4:44 pm

Re: Power consumption problem when the AP is not avaliable

2015-09-16T15:38:23+08:00

The API that can close RF will be released at new document in next version.
Thinks for everybody.

Statistics: Posted by ESP_Alfred — Wed Sep 16, 2015 3:38 pm

Re: Power consumption problem when the AP is not avaliable

2015-09-16T13:49:45+08:00

I would agree with tve rather than blubb. The way I understand ESP_Alfred's post earlier, the problem isn't that the chip is trying to connect in too short intervals, but rather that it doesn't power down the RF despite not doing anything.
If there was a possibility to tell the RF to go to sleep (or, if there is, if we knew how to use it), it would be easy to implement any kind of reconnect algorithm, including intervals of any length or exponential back-off.

Statistics: Posted by Npt — Wed Sep 16, 2015 1:49 pm

Re: Power consumption problem when the AP is not avaliable

2015-09-16T12:20:31+08:00

Yes, or the ability to actually turn the RF off (and back on) so we can control when it tries to reconnect. The API calls are there, but they don't behave correctly.

Statistics: Posted by tve — Wed Sep 16, 2015 12:20 pm

Re: Power consumption problem when the AP is not avaliable

2015-09-15T21:07:39+08:00

I guess the problem is that ESP is constantly trying to reconnect and what we need would be an API to set a time interval between these attempts.

Statistics: Posted by blubb — Tue Sep 15, 2015 9:07 pm

Re: Power consumption problem when the AP is not avaliable

2015-09-15T14:34:53+08:00

@ESP_Alfred: can you please explain how I can use the esp8266 with a power budget that does not allow for continuous 70mA? I have a battery operated system that functions well as long as the wifi DTIM allows it to remain at <20mA most of the time. What should my software do when it detects that the system has lost connection with the access point in order to prevent continuous 70mA power draw?

Statistics: Posted by tve — Tue Sep 15, 2015 2:34 pm

Re: Power consumption problem when the AP is not avaliable

2015-09-14T16:26:08+08:00

@ESP_Alfred: I see. Without knowing the internals of the chip, this behaviour seemed a bit counter-intuitive. Thank you for clarifying!

Can I ask one more favour? Like Harold, I don't know how to "close the RF" without a complete reset of the device. Could you perhaps post a small code snippet?

Statistics: Posted by Npt — Mon Sep 14, 2015 4:26 pm

Re: Power consumption problem when the AP is not avaliable

2015-09-14T15:42:42+08:00

ESP_Alfred wrote:
Npt wrote:(I tried to post in this thread a couple of days ago, apparently my comment was not approved by the moderators. I'll try one more time...)

I have encountered the same unexpected behaviour of the chip. I think we need to distinguish between two scenarios here:
(A) The ESP involuntarily loses connection to an AP (for example the AP is shut down or goes out of range).
(B) The ESP voluntarily ends the connection (wifi_station_disconnect) or never establishes a connection in the first place (wifi_station_set_auto_connect(false)).

In case of (A) I'd say it's ok if the ESP use more power in order to try to re-establish the connection. If this is not desired, it is your responsibility as the API user to stop it, maybe after a certain timeout. So in my opinion everything's fine here.

However, in situation (B) I think there's a bug in the API. You would expect that the power consumption eventually goes down after disconnecting. The opposite is the case: I measured a stable ~70mA after disconnect as opposed to ~20mA (with short spikes every 1s or so) when connected and idle.
In my application I need to do some computations on the CPU before sending the data. One would assume that it's better to do the computations offline and only connect after they're done. Turns out, the opposite is true: you have to connect to an AP to save power. That is weird and definitely a bug that Espressif should look into.

for situation (B), It's normal.
As you say : a stable ~70mA after disconnect as opposed to ~20mA (with short spikes every 1s or so) when connected and idle.
please reference to the attachment of current map.

Thanks for your reply. I agree with Npt, and consider the behavior (B) as a bug, because after the wifi_station_disconnect() call, I also called wifi_get_sleep_type() and ensure it returns "2" which refers to 'MODEM_SLEEP_T' as you suggested. My current chart still is exactly the same as your's.

Statistics: Posted by Harold L. — Mon Sep 14, 2015 3:42 pm

Re: Power consumption problem when the AP is not avaliable

2015-09-14T13:52:39+08:00

Npt wrote:
(I tried to post in this thread a couple of days ago, apparently my comment was not approved by the moderators. I'll try one more time...)

I have encountered the same unexpected behaviour of the chip. I think we need to distinguish between two scenarios here:
(A) The ESP involuntarily loses connection to an AP (for example the AP is shut down or goes out of range).
(B) The ESP voluntarily ends the connection (wifi_station_disconnect) or never establishes a connection in the first place (wifi_station_set_auto_connect(false)).

In case of (A) I'd say it's ok if the ESP use more power in order to try to re-establish the connection. If this is not desired, it is your responsibility as the API user to stop it, maybe after a certain timeout. So in my opinion everything's fine here.

However, in situation (B) I think there's a bug in the API. You would expect that the power consumption eventually goes down after disconnecting. The opposite is the case: I measured a stable ~70mA after disconnect as opposed to ~20mA (with short spikes every 1s or so) when connected and idle.
In my application I need to do some computations on the CPU before sending the data. One would assume that it's better to do the computations offline and only connect after they're done. Turns out, the opposite is true: you have to connect to an AP to save power. That is weird and definitely a bug that Espressif should look into.

for situation (B), It's normal.
As you say : a stable ~70mA after disconnect as opposed to ~20mA (with short spikes every 1s or so) when connected and idle.
please reference to the attachment of current map.

Statistics: Posted by ESP_Alfred — Mon Sep 14, 2015 1:52 pm

Re: Power consumption problem when the AP is not avaliable

2015-09-11T14:00:58+08:00

(I tried to post in this thread a couple of days ago, apparently my comment was not approved by the moderators. I'll try one more time...)

I have encountered the same unexpected behaviour of the chip. I think we need to distinguish between two scenarios here:
(A) The ESP involuntarily loses connection to an AP (for example the AP is shut down or goes out of range).
(B) The ESP voluntarily ends the connection (wifi_station_disconnect) or never establishes a connection in the first place (wifi_station_set_auto_connect(false)).

In case of (A) I'd say it's ok if the ESP use more power in order to try to re-establish the connection. If this is not desired, it is your responsibility as the API user to stop it, maybe after a certain timeout. So in my opinion everything's fine here.

However, in situation (B) I think there's a bug in the API. You would expect that the power consumption eventually goes down after disconnecting. The opposite is the case: I measured a stable ~70mA after disconnect as opposed to ~20mA (with short spikes every 1s or so) when connected and idle.
In my application I need to do some computations on the CPU before sending the data. One would assume that it's better to do the computations offline and only connect after they're done. Turns out, the opposite is true: you have to connect to an AP to save power. That is weird and definitely a bug that Espressif should look into.

Statistics: Posted by Npt — Fri Sep 11, 2015 2:00 pm

Re: Power consumption problem when the AP is not avaliable

2015-09-11T01:45:37+08:00

ESP_Alfred wrote:
Harold L. wrote:Hi, we've tried all related APIs, but no lucky, ESP8266 consumed too much battery when AP is not exist, are you still working with this problem?
Glad to hear from you, thanks!

hi, Harold L
We have a lower power working mode ,base on deep sleep mode. please give your e-mail address to me, we will connect to you.

My address is: haroldmars[A.T.]icloud.com
Thank you a lot!

Statistics: Posted by Harold L. — Fri Sep 11, 2015 1:45 am

Re: Power consumption problem when the AP is not avaliable

2015-09-10T19:42:32+08:00

Harold L. wrote:
Hi, we've tried all related APIs, but no lucky, ESP8266 consumed too much battery when AP is not exist, are you still working with this problem?
Glad to hear from you, thanks!

hi, Harold L
We have a lower power working mode ,base on deep sleep mode. please give your e-mail address to me, we will connect to you.

Statistics: Posted by ESP_Alfred — Thu Sep 10, 2015 7:42 pm

Re: Power consumption problem when the AP is not avaliable

2015-09-10T05:06:44+08:00

Hi, we've tried all related APIs, but no lucky, ESP8266 consumed too much battery when AP is not exist, are you still working with this problem?
Glad to hear from you, thanks!

Statistics: Posted by Harold L. — Thu Sep 10, 2015 5:06 am

Re: Power consumption problem when the AP is not avaliable

2015-09-03T14:57:13+08:00

I am struggling with the same unexpected behaviour of the chip. I set the auto connect to false so that I could do some computations before firing up the WiFi connection and transmitting the results. However, instead of saving power, this method seems to consume more.

I only have a very basic DMM, so take my measurements with a grain of salt. There's also a power LED on my board, but that will only affect the absolute values, not the relative ones.
After power on and calling wifi_station_set_auto_connect(false) and wifi_set_opmode(STATION_MODE) in user_init, I get ~75mA. When I call wifi_station_connect() the current drops(!) to ~20mA with short spikes every second or so. Calling wifi_station_disconnect() makes the power rise again to ~75mA. These measurements are all taken after a short wait, when they've stabilised. Each operation (boot, connect, disconnect) causes a short rise in power, which seems logical to me.

What doesn't seem logical is that idling without being connected to an AP takes almost four times as much power as idling with a connection. Just to be clear: I'm not talking about a connection that was lost involuntarily. I understand that in that case the module would try to regain the connection by upping the power. But it shouldn't happen if the connection was willingly shut down (wifi_station_disconnect) or never established in the first place (wifi_station_set_auto_connect(false)).

And yeah, the fact that we can't enable or disable RF without a reboot, really sucks.

Statistics: Posted by Npt — Thu Sep 03, 2015 2:57 pm

Re: Power consumption problem when the AP is not avaliable

2015-08-24T14:59:53+08:00

I'm trying to locate the code I mentioned, but am failing right now. Have you tried using wifi_station_set_reconnect_policy? Or setting wifi_station_set_auto_connect to false and managing reconnection attempts yourself?

Statistics: Posted by tve — Mon Aug 24, 2015 2:59 pm

Re: Power consumption problem when the AP is not avaliable

2015-08-24T11:18:21+08:00

tve wrote:
I do believe this is a bug in the SDK. The bug being that the average power consumption goes up by a significant factor in STA mode if the AP is not reachable.
The work-around I've found is to write a callback for wifi status changes and when you get a disconnect event implement a back-off retry mechanism that disables wifi altogether and then turns it on again periodically (and then off again if no AP is reachable). In your use-case, you can probably get away with turning it on briefly every minute or even every 5 minutes.

I think you are right, but I didn't find a way to disable/enable wifi without rebooting. I found the only way to shut it down is entering deep-sleep mode and rebooting with "system_phy_set_rfoption(4);". If we want to enable it again, another reboot is inevitable.

A more weird thing is, calling 'wifi_station_disconnect()' may also cause the abnormal high current.

Do you have a workable way to enable/disable wifi?

Statistics: Posted by Harold L. — Mon Aug 24, 2015 11:18 am

Re: Power consumption problem when the AP is not avaliable

2015-08-24T11:07:39+08:00

ESP_Alfred wrote:
1、As you see form table 4, the current rising is inevitable when 8266 send packet.If you want to reduce this part of the current, You could modify the parameters in the "esp_init_data_default.bin" via "flash download tool" to reduce the power of Tx.

2、8266 have some sleep mode, may be you adopt this mode in your occasion.

1. I'm sorry you may not catch my point... I've made some translation to help us understand each other

As I've said, "Could we come up with a compromise like introducing some 'gaps' between each Tx retrying? It seems that now the strategy is little bit like a 'flood'. Can we take a strategy like TCP's retransmission algorithm(Exponential Back off) ?"

It means: 我们能找到一种方式，当AP不存在的时候，不是像现在这样洪水一般的发包寻找AP，而是引入一些间隙，隔几秒钟试探一次？这并不会妨碍重新连接的过程，却能大大降低这种情况下的平均功耗。TCP的重传机制（重试时间按指数退避）是一个可以参考的例子。

2.I've tried all sleep modes, and found only after entering deep-sleep mode and rebooting with "system_phy_set_rfoption(4);" can reduce current. But unless another deep-sleep rebooting, it'll never connect even if the AP appears again! This is a bad experience for our customers.

Statistics: Posted by Harold L. — Mon Aug 24, 2015 11:07 am

Re: Power consumption problem when the AP is not avaliable

2015-08-20T01:09:53+08:00

I do believe this is a bug in the SDK. The bug being that the average power consumption goes up by a significant factor in STA mode if the AP is not reachable.
The work-around I've found is to write a callback for wifi status changes and when you get a disconnect event implement a back-off retry mechanism that disables wifi altogether and then turns it on again periodically (and then off again if no AP is reachable). In your use-case, you can probably get away with turning it on briefly every minute or even every 5 minutes.

Statistics: Posted by tve — Thu Aug 20, 2015 1:09 am

Re: Power consumption problem when the AP is not avaliable

2015-08-19T17:43:31+08:00

Harold L. wrote:
Thanks for your reply! Yes, we have very strong demand, because we're designing a hand-held device powered by battery, and customers need to have a longer battery life outside(e.g. on the streets, parks, restaurants...). Our device will upload datas through WIFI only after they coming back to home.

Could we come up with a compromise like introducing some 'gaps' between each Tx retrying? It seems that now the strategy is little bit like a 'flood'. Can we take a strategy like TCP's retransmission algorithm(Exponential Back off) ?

Thanks again, saving power is really matters to us, hope we can find a way together!

1、As you see form table 4, the current rising is inevitable when 8266 send packet.If you want to reduce this part of the current, You could modify the parameters in the "esp_init_data_default.bin" via "flash download tool" to reduce the power of Tx.

2、8266 have some sleep mode, may be you adopt this mode in your occasion.

Statistics: Posted by ESP_Alfred — Wed Aug 19, 2015 5:43 pm

Re: Power consumption problem when the AP is not avaliable

2015-08-19T11:25:40+08:00

ESP_Alfred wrote:
there are two reasons for this current:
1. when the AP power off, 8266 will send package for disconnect.
2. 8266 will change data rate to try to connect AP, because 8266 don't know how about AP. May be because the router signal is weak.
High data rate means high current, although a short duration.

Do you have a strong demand on this ? Please tell us your usage occasion.

Thanks for your reply! Yes, we have very strong demand, because we're designing a hand-held device powered by battery, and customers need to have a longer battery life outside(e.g. on the streets, parks, restaurants...). Our device will upload datas through WIFI only after they coming back to home.

Could we come up with a compromise like introducing some 'gaps' between each Tx retrying? It seems that now the strategy is little bit like a 'flood'. Can we take a strategy like TCP's retransmission algorithm(Exponential Back off) ?

Thanks again, saving power is really matters to us, hope we can find a way together!

Statistics: Posted by Harold L. — Wed Aug 19, 2015 11:25 am

Re: Power consumption problem when the AP is not avaliable

2015-08-18T11:48:05+08:00

Harold L. wrote:
Btw, you should focus on the relative value, the rising-edge of the current when power off your router(130mA - 70mA = 60mA) instead of the absolute values, because there exists some bias-current caused by things like LCD screen, external ADCs in my measurement.

Surely, those things would not be the reason of this problem since they just have not way to be informed when the router is powering on/off.

there are two reasons for this current:
1. when the AP power off, 8266 will send package for disconnect.
2. 8266 will change data rate to try to connect AP, because 8266 don't know how about AP. May be because the router signal is weak.
High data rate means high current, although a short duration.

Do you have a strong demand on this ? Please tell us your usage occasion.

Statistics: Posted by ESP_Alfred — Tue Aug 18, 2015 11:48 am

Re: Power consumption problem when the AP is not avaliable

2015-08-18T11:18:37+08:00

Harold L. wrote:
Btw, you should focus on the relative value, the rising-edge of the current when power off your router(130mA - 70mA = 60mA) instead of the absolute values, because there exists some bias-current caused by things like LCD screen, external ADCs in my measurement.

Surely, those things would not be the reason of this problem since they just have not way to be informed when the router is powering on/off.

Dear Harold L.
The table 4 is right.
When we disconnect from AP, we will send packets. So "Tx" been opened.

Statistics: Posted by ESP_Alfred — Tue Aug 18, 2015 11:18 am

Re: Power consumption problem when the AP is not avaliable

2015-08-17T11:36:57+08:00

Btw, you should focus on the relative value, the rising-edge of the current when power off your router(130mA - 70mA = 60mA) instead of the absolute values, because there exists some bias-current caused by things like LCD screen, external ADCs in my measurement.

Surely, those things would not be the reason of this problem since they just have not way to be informed when the router is powering on/off.

Statistics: Posted by Harold L. — Mon Aug 17, 2015 11:36 am

Re: Power consumption problem when the AP is not avaliable

2015-08-17T11:26:38+08:00

ESP_Alfred wrote:
Dear Harold L,
Could you tell me how get the value "130mA" , and how long this current exist.
Ths

Hi! Any of the cases below would lead you to the abnormal high current:
1. Power off your router after ESP8266 connected as a station, you will see the abnormal high current. In my situation, it continues as long as the disappearance of your router's signal.
2. Power off your router, then power on a ESP8266(configed as a station), you will also see it happens.
3. call 'wifi_station_disconnect' after ESP8266 connected as a station. the abnormal high current continues until you call 'wifi_station_connect'

Thank you!

Statistics: Posted by Harold L. — Mon Aug 17, 2015 11:26 am

Re: Power consumption problem when the AP is not avaliable

2015-08-14T21:53:06+08:00

Harold L. wrote:
Hi! We are designing a battery powered device on the top of ESP8266, and found there is abnormal high current during some situation.

Firstly set ESP8266 to station mode. The total current was 70mA. Once in a while went up to 100mA, it's OK!

But in both situations below, the total current went up to an abnormal high value: 130mA

1. When AP is not available. e.g. power off your router, or take ESP8266 out to the streets;
2. call 'wifi_station_disconnect'

We'd like to save power when WIFI is not available since our device is powered by battery. Wish you can give me some help, it really matters to us, thanks!

Dear Harold L,
Could you tell me how get the value "130mA" , and how long this current exist.
Ths

Statistics: Posted by ESP_Alfred — Fri Aug 14, 2015 9:53 pm

Re: Power consumption problem when the AP is not avaliable

2015-08-11T12:02:59+08:00

kolban wrote:
Howdy,
If I read the datasheet available here:

download/file.php?id=520

It seems to say (see Table 4) that the current draws you mention are normal....

I could be reading the table incorrectly ....

Neil

Thanks for your reply, but I think the table 4 only mentions the situation of Txing and Rxing packets. In my first situation, the AP is NOT exist at all, and no packets are sending or receiving.
In that scene, we hope only CPU are working -- recording data silently -- as long as possible.

Statistics: Posted by Harold L. — Tue Aug 11, 2015 12:02 pm

Re: Power consumption problem when the AP is not avaliable

2015-08-10T21:31:23+08:00

Howdy,
If I read the datasheet available here:

download/file.php?id=520

It seems to say (see Table 4) that the current draws you mention are normal....

I could be reading the table incorrectly ....

Neil

Statistics: Posted by kolban — Mon Aug 10, 2015 9:31 pm

Power consumption problem when the AP is not avaliable

2015-08-10T16:27:43+08:00

Hi! We are designing a battery powered device on the top of ESP8266, and found there is abnormal high current during some situation.

Firstly set ESP8266 to station mode. The total current was 70mA. Once in a while went up to 100mA, it's OK!

But in both situations below, the total current went up to an abnormal high value: 130mA

1. When AP is not available. e.g. power off your router, or take ESP8266 out to the streets;
2. call 'wifi_station_disconnect'

We'd like to save power when WIFI is not available since our device is powered by battery. Wish you can give me some help, it really matters to us, thanks!

Statistics: Posted by Harold L. — Mon Aug 10, 2015 4:27 pm