ESP12S

Peter Scargillesp12S, ESP826623 Comments on ESP12S

ESP12SI’m looking forward to trying out the new ESP12-S – this one apparently has a better antenna – and by the look of the schematic (see page 15 – thanks Alexander) the resistors on reset, GPI))0,2 and 15 are already there.

That should simplify layout somewhat!

ai-thinker.com – www.ai-thinker.com/uploadfile/2016/0909/20160909073755898.pdf

Hmm, board layout with 0.1 pins, 3v3 reg with decent heatsink area underneath, maybe 3 mosfets…

So this should work with existing layouts. I’ve kept back an IoTBEAR adaptor and ordered a couple of these from Ebay sub- £2 each.

23 thoughts on “ESP12S”

  1. Hi Pete, the provided link at ai-thinker.com looks dead. I’m looking for the ESP-32S module schematic. Could you manage to provide the pdf file or another link ?

  2. I’ll be watching to see your thoughts. I have been using the ESP-12F and am getting ready to release a commercial product…I noticed the ESP-12S on Aithinker’s website and a few other places. There seems to be some mention of wifi performance improvements as well, but no one has done any extensive testing. Its also interesting about the pull-ups, I wonder if there will be an issue if you already have pull-ups on your board.

    1. That’s a very good point Curtis about existing boards but if I remember rightly they used pretty high values so I would not expect this to be an issue. As you say, some concrete comparative testing would be good.

    2. Tested 2 ESP-12S…. Pure garbage . Reset and dump all the time . After 1 of the two died I opened the shield …. inside i found a Flash called “Buy a Micro” with code25032ASSIG…. I would sa ASS Flash…….It’s a joke ? How can We experiment IoT on devices that stop to work after 5th reflash? Come On…

      1. I would be willing to bet that’s a power supply issue – either that or you’ve been VERY unlucky – these things are used all over the place in quantity. I have an ESP12 here and I’ve been re-flashing it endlessly without any problem at all.

        1. Well I hope to be just unlucky, really. but the ESP-12S seems to be more affected . And in general the ESP8266 is somewhat instable…To me the flash inserted need to be of industry level quality…..and not “Buy a Micro” branded… I would be ready to pay more for having a more accurate ESP8266 board really well done and well equipped.
          I wil buy a new set and see If i can replicate the issues..
          C.

          1. Well I wish you luck and as far as the ESP12-S is concerned I’ve not used enough of them to make a definitive statement – however – I believe you are absolutely wrong on the “somewhat instable” comment – they are not somewhat unstable – they are generally rock-steady. They are in use in several products including the Sonoff devices which many of us love. I have them in use in two properties night and day controlling heating and lighting and on the odd occasion I’ve had trouble – it has been power supply related. I was involved in a project some time ago where we put hundreds of ESP-01 units on boards for a government project – they sat night and day sending all sorts of information back to what became a very large database – no problem. I would have no hesitation in recommending these devices – but of course it is always possible you bought a rip-off. I always buy mine from AliExpress – who tend to be one of the cheapest places to buy them as well.

  3. I read a while ago that ESP folks said that ESP8266s are actually 5V tolerant.
    Any confirmation?

    1. Updated… If Swee-An says they are 5v tolerant (apart from power obviously) I’m happy with that. I have been known to feed serial straight from a 5v FTDI but always thought that this was more down to lack of power coming out of the FTDI serial out pin. You learn something every day. No wonder I get away with it.

      1. The CEO confirmed that the data pins were 5V tolerant, though obviously not the power input. I feed them 5V somewhat regularly with no problems.

        1. Well, I have to tell you Jason that this goes WITH my experience – i.e. firing serial in at 5v I’ve never damaged anything – but as for general input of 5v to GPIOs this goes against everything I’ve read… so I would just say to others before you act on this – check for yourself in the specs…

          Look at this for example – https://hackaday.io/page/2024-esp8266-is-5v-tolerant-after-all

          You end up no better off at the end than at the start!!

          When you say the CEO – do you mean Swee-Ann Teo ? If so then I’ll be happy with that . a bright man (I was part of an Espressif tour to Boston a couple of years ago and I spent some time with him and Ivan Grokotkov of Esp-Arduino fame).

          2. the esp8266 are 5v tolerant, see page 17 and 18 from datasheet.

            http://download.arduino.org/products/UNOWIFI/0A-ESP8266-Datasheet-EN-v4.3.pdf

            “All digital IO pins are protected from over-voltage with a snap-back circuit connected between the
            pad and ground. The snap back voltage is typically about 6V, and the holding voltage is 5.8V. This provides protection from over-voltages and ESD. The output devices are also protected from
            reversed voltages with diodes. ”

            it works without any problem.

            1. compatible with TTL logic level and not 5V CMOS. VIH = 0,7*VDD
              = 3.5V
              ESP VOH(max) = 3,3 a logic 1 is not guarantee.

      2. Pete,

        Thanks for the update. Really good to know the IO v. Power difference.

        Have a great trip to the Sunshine…

    1. Good – but given that the ESP current draw has peaks around that – I think I’ll stick with the 1 amp jobs – also gives a little spare for any peripherals… the quiescent draw is irrelevant compared to the power-hungry ESP8266 of course…

      1. Very low quiescent current draw can be useful for some applications i.e. lengthy deep sleep intermittently woken battery powered, seldom charged applications. The LF33 is also pretty good and has a very wide input voltage range..which is sometimes nice..but if you don’t care about your regulator being a power sucking beast and you want some headroom then yeah it’s probably always going to be something like an AMS1117

        1. The point being Joe – the ESP takes so much current that saving a few microamps here or there will make no difference at all.

            1. I just read one of your articles on an ESP board with LIPO charging circuit so I can see why you’d be interested in that angle – personally the power aspect has no effect on any of my projects however, nice to know there’s a PCB out there.

              I like the use of the AliExpress charging boards

              https://de.aliexpress.com/item/Free-Shipping-5V-1A-Micro-USB-18650-Lithium-Battery-Charging-Board-Charger-Module-Protection-Dual/32453058256.html?aff_platform=aaf&cpt=1492069148097&sk=VnYZvQVf&aff_trace_key=6df1cd0613944557bfc5d745a8b14ee6-1492069148097-00990-VnYZvQVf

              So cheap it isn’t worth doing your own layout. A shame their charging limit is one amp – indeed I’d guess knocking 30% off that would be good – which makes them ideal for ESP8266. I wonder if there is a similar board out there which can go up to 3amps.

              I ponder that because I recently bought a cheap board to go with a Raspberry Pi which included a Lithium battery – it was so appallingly bad they refunded me after a round of emails – on removing the power, the Pi reset – which somewhat defeats the object of using a backup board. It is something I’ve thought of doing for a while, a board with usb connector, battery and a range of mounting holes to handle pi, Orange Pi and some of the Armbian units but with enough power to handle a Pi3. Such boards exist for the Pi but they are not cheap.

            2. that was my point… devices made for logging, unmanaged, for the longest time possible… i see many remove leds and ldo from their arduinos to reduce power consumption… Great Scott on youtube is one of them and i think he made a video on this…

Comments are closed.