Hi,
Does anyone know whether it’s possible to use the murata CMWX1ZZABZ-091 and achieve ultra low power usage (microamps) at all? The datasheets only refer to a transmit and receive mode power usages and both are > 20ma. It seems strange that there’s no mention anywhere of a deep sleep mode and it’s power usage.
Kim
Yes, we use the module here:
and have measured the sleep current to be ~2 uA. We are also using this module in asset trackers and environmental data loggers.
There will soon be an Arduino core for it too.
That’s a beautiful module there! 
I have one. I haven’t used it yet, I will need to hunt down sketches for putting it into deep sleep and waking up on gpio changes for my purposes. Currently I’ve been using the arduino with a teensy plugin. I’ve noticed that there are a lot of differences and experiences in utilizing deep sleep on the different Arduino platforms. Any experiences, tips, links you can share to use with that grasshopper I’m sure will be appreciated from the people on this forum.
Cheers,
Kim
I agree 
OMG would love to build a tiny node with that one, but reflowing this one may be a new challenge
The Arduino core for the CMWX1ZZABZ is almost ready; I am as anxious to put it to use as anyone. As soon as the Arduino core is available I expect to post several sketches on Github demonstrating Grasshopper use including how to access the low power modes (which is straightforward). Maybe by the end of this month…
This is one of the projects I am planning to use it for:
Its a 23 mm x 23 mm sensor tile with BME280, BMA280, and VEML6040 with 16 MByte SPI flash for logging, LiPo battery charger, programmable via USB using the Arduino IDE.
The sleep current is < 10 uA and daily usage at 0.1 Hz data logging rate is likely to be < 10 mA per day, still TBD. But the idea is to use it as an ultra-low-power environmental data logger and long-range edge node, etc.
It’s just a prototype for testing but I like the form factor and there is still room for more (low-power) sensors.
Awesome,
Any tips on how you’re soldering all of this (on both side)? custom reflow oven or dedicated manufacturer?
Indeed a great node! Any final cost estimate already?
Is anyone here aware of a simple breakout for this module, that would help tinkerers a lot 
You mean like Grasshopper above?
An Arduino core on this will be very interesting.
Grasshoppers ordered 
@Charles, I use a hot air gun and hot plate with low temperature ChipQuik solder paste for this kind of prototyping. Easy once you get the hang of it, but if you’ve never done it probably have to work up to this level of intricacy. For production, like Grasshopper or the Cricket asset tracker, I use fabs in Beijing or Hong Kong, of course.
@kiwiclan, I have other dev boards that use the STM32L4 family of MCUs (no SX127X) that work like the Teensy but allow very low power operation, 2 uA stop current and low 80 MHz current consumption compared to the Teensy 3.X:
See here for an example:
I know this is a TTN forum but you asked about Arduino-based, low-power usage so there you go…
@Megabyte, hope it’s just a matter of weeks now for the complete Arduino core, but in the mean time you can always use an ST-Link and the ST Hal. Grasshoppers on their way!
Hi Kris,
The ‘great node’ and ‘cost estimate’ / price indication relate to the Pesky products LoRa Sensor tile 
With breakout I mean just a bare breakout with no extra’s that just breaks out as many GPIO’s as possible and that can be used by tinkerers to build new prototypes based on this very nice CMWX1ZZABZ Murata/STM module with minimal cost. But indeed the Grasshopper can be seen as a breakout-deluxe for this module 
Hey Kris, I was talking to someone that suggested that in order to install an arduino core on the ST MCU in the murata module for the grasshopper you would be throwing away and not using the murata developed lorawan stack. This doesn’t seem right to me, That’s not really the case is it?
Kim
No, this is not right. At least, we have a full LoRaWAN stack as part of the system layer accessible via the Arduino core. Whether this is exactly the same as the Murata stack probably not, since both the ST HAL for the L082 as well as the stock LoRaWAN stack are bare minima.
We are not just grabbing ST’s Arduino core and shoe-horning it into the Murata Module. We have built a system layer from scratch including both L082 and SX1276 system calls supporting all peripherals in a power efficient way and then created an Arduino layer on top of this into an integrated, well-designed whole.
Bottom line, the user will be able to use the 19 GPIOs and associated peripherals of the L082 as well as LoRa p2p and full LoRaWAN with the Arduino IDE as easily as any other device out there, but likely more power efficient and faster…
Excellent! You have me on the edge of my seat 
Look forward to it.
Kim
Hi Kris,
Indeed great news, really looking forward to it!
I’m a bit confused though by what you mean with:
Is the “stock LoRaWAN stack” the stack from ST that they developed for this Murata module and got the LoRaWAN certification on AFAIK, or the LMIC stack that most people use when developing for Arduino, as that would void the LoRaWAN certification I guess? (BTW I don’t think a “Murata stack” exists as they refer to ST for the software)
Anyhow both the grasshopper and the Lora sensor tile would be great modules for both tinkerers and other developers using the STM development environment.
OK, clarification from my colleague Thomas who is developing the system layer and Arduino core:
"We are using the official Semtech Stack (https://github.com/Lora-net/LoRaMac-node) as a starting point. Reworked the region code to be space efficient. Reworked the radio stack for the same. Totally different porting layer internally from the original stack for power reasons.
ST is using the same stack as a starting point, with their own porting layer, which then goes to the HAL. Lot’s of bugs there …"
Didn’t mean to disparage (too much) SEMTECH’s stack or ST’s HAL but the Grasshopper system layer (including the LoRa/LoRaWAN stack and L082 HAL) and the Arduino core are new, complete, space-efficient, designed for low-power usage, and reasonably bug-free. Of course there will be bugs and we expect the initial users (including ourselves) will help identify them so we can fix them as soon as possible…
Thanks @onehorse, that clears up things for me. Looks like it’s going to be the ultimate LoRaWAN node platform to me
Of course, that’s what we’re hoping to achieve!
Fast moving space new nodes at the moment, hope this doesn’t get overtaken before its ready…