LowPowerLab Forum

Hardware support => Low Power Techniques => Topic started by: Felix on April 20, 2016, 04:06:04 PM

Title: Making a lower power Moteino
Post by: Felix on April 20, 2016, 04:06:04 PM
I do have to agree it would be nice to make something that runs from an (2x)AA source, no regulator, 8mhz internal resonator, super optimized for low power. The gain would be something like going from the current 4uA (Moteino) to under 1uA. Quite doable and something I'd like to make one day.
Title: Re: Making a lower power Moteino
Post by: joelucid on April 20, 2016, 04:32:39 PM
I wholeheartedly agree. Battery operation is really the sweet spot of the 328p yet to get most out of it you first have to perform a multi-limb amputation on a Moteino. Let go of the regulator and the oscillator plus the flash and add an rtc so users can finally go 'legit' with freq hopping ;)
Title: Re: Making a lower power Moteino
Post by: Felix on April 20, 2016, 08:38:37 PM
I agree back.
I do need to mention that the regulator is really a sweet feature of the mainstream Moteinos and without which I probably wouldn't be in Moteino-selling business at this point, maybe I'd just be another forgotten 4 year old hackaday post for some cool-but-who-cares-anyway hand assembled SMD board. I have to also say that the regulator is not by far the #1 issue in achieving low power either.
Title: Re: Making a lower power Moteino
Post by: perky on April 21, 2016, 11:08:18 AM
The reguator is nice as it allows it to be powered from higher voltages, maybe the type could be changed for an LDO that has a lower voltage spec and possibly lower quiescent current? As it stands it would work down to about 2.7V, but alkalines go down to 1V per cell so you'd need one working down to 2V. The AS1360-33-T looks interesting here, it's pin compatible with the MCP1703 in SOT-23 package but works down to 2V with 20V max input voltage and 1.5uA Iq, also has a lower dropout voltage for given current.

http://www.mouser.com/ds/2/588/AS1360_Datasheet_EN_v2-472961.pdf

Mark.
Title: Re: Making a lower power Moteino
Post by: Felix on April 21, 2016, 11:48:44 AM
The reguator is nice as it allows it to be powered from higher voltages, maybe the type could be changed for an LDO that has a lower voltage spec and possibly lower quiescent current? As it stands it would work down to about 2.7V, but alkalines go down to 1V per cell so you'd need one working down to 2V. The AS1360-33-T looks interesting here, it's pin compatible with the MCP1703 in SOT-23 package but works down to 2V with 20V max input voltage and 1.5uA Iq, also has a lower dropout voltage for given current.

http://www.mouser.com/ds/2/588/AS1360_Datasheet_EN_v2-472961.pdf

Mark.

Nice find. Price hurts though :) it's 3X+ more than the current regulator in small quantities but getting pretty good at 1000X+.
The problem with the HopeRadios (RFM69, 95) is the VDDmin = 2.4V. So that is quite limiting, the lowest AS1360 usable variant is 2.5v.
Title: Re: Making a lower power Moteino
Post by: perky on April 21, 2016, 12:41:24 PM
I think the RFM69 series have a min VDD of 1.8V according to the spec sheets (RFM69W and RFM69HW).

I'm not sure what you mean that 'the lowest usable AS1360 variant is 2.5V'. LDOs can and generally are used for voltage clamping, if you put a 2.4V input for example into a 3.3V LDO the pass transistor is fully on so the voltage out follows the voltage in, i.e. in that example it would be 2.4V, until it regulates. Therefore you'd choose a 3.3V variant of LDO if you want to restrict your output voltage from ever going above 3.3V. It's important to get one with a low input voltage spec so you know the circuit starts to work (i.e. the pass transistor turns on) at that low voltage.

Mark.

Edit: I notice the voltage range you specifiy for Moteinos is recommended 3.5V to 9V. This will happily work down to 2.7V (assuming the other components also work down to 2.7V).
Title: Re: Making a lower power Moteino
Post by: Felix on April 21, 2016, 02:18:19 PM
The abs minimum for HW is 2.4V. Disregard the Hope datasheets, they are incorrect. There is an RF switch component on the board that limits it to 2.4v, according to its datasheet. It might work below that but its out of spec.
Regardless of that, the W uses the same silicon chip from Semtech, the SX1231h, which is a 2.4v minimum also: http://www.semtech.com/images/datasheet/sx1231h.pdf
See the datasheet for the regulator you referenced, it has fixed output voltages, the closest greater than 2.4v is 2.5v - so that's where I got the 2.5v.
Title: Re: Making a lower power Moteino
Post by: perky on April 21, 2016, 03:03:25 PM
Well thanks for pointing that out! I'm in the middle of a project that uses 2 cells, I thought I had some headroom so glad I put an ideal diode circuit on it now  :-\

I'm still not sure why you would choose a regulator with a regulated voltage closest to 2.4V, unless you want all the circuit all to run at 2.5V. That means any IO would also be at 2.5V and maybe would make programming and interfacing harder from a logic level point of view.  Putting in a 3.3V LDO with a low minimum working input voltage like the AS1360-33-T would still work down to 2.4V, it would also allow it to be run at 3.3V in 3.3V systems without logic level translation. I don't think there's any power benefit from doing that given the current consumption is a linear thing with voltage and you'd be dropping voltage across the regulator didssipating the power that would have been dissipated in the circuit.

Anyway, the main thing is you've prevented me from making a potentially costly mistake from a production point of you, so thank you!

Mark.
Title: Re: Making a lower power Moteino
Post by: Felix on April 21, 2016, 03:57:51 PM
I'm still not sure why you would choose a regulator with a regulated voltage closest to 2.4V, unless you want all the circuit all to run at 2.5V.
I thought you wanted to run from 2xAA ?
If you want 3.3v to be in the range of 3.3/5v devices, in that case rather than a LDO you would use a booster with 2xAA which shouldn't be too bad. But you'd want something with at least 200mA output.
Title: Re: Making a lower power Moteino
Post by: perky on April 21, 2016, 04:27:34 PM
Right, I've been digging into this a little deeper. The Semtech sx1231 can output +13dB and has a minimum voltage limit of 1.8V. The sx1231h has a max output of +20dB (100mW), and a minimum voltage of 2.4V.

Now the RFM69 uses a sx1231h (as confirmed by the version number register which is different to the sx1231). However, the chip is marked RF69, and there is a datasheet for it:
www.orcam.eu/res/Datablad/rf69v12.pdf

This shows that for up to +17dB the voltage can be as low as 1.8V (using PA1 and PA2, normal power settings) . From +17dB to +20dB it uses PA1, PA2 with high power settings and needs 2.4V. So it appears the 2.4V spec is for the full output power using PA1, PA2 and high power settings.

The RFM69CW is restricted to +13dB from the RFIO output, and doesn't use PA1, PA2 or the high power settings, and according to the RF69 datasheet only needs 1.8V. Even though this is missing in the sx1231h datasheet the thing that makes it credible is that the sx1231 is also +13dB without the PA amplifiers and also has a minimum voltage of 1.8V.

Now I'm curious as to what the other 2.4V component is, the RFM69CW and RFM69W have only passives and crystal, unless you're saying it's an oscillator? All the reference designs use a crystal and both pins 4 and 5 of the sx1231h are connected to the device and the other two are ground, so I'm sure it's a crystal.

My conclusion is that the RFM69CW and RFM69W are OK down to 1.8V.
Mark.
Title: Re: Making a lower power Moteino
Post by: perky on April 21, 2016, 04:30:46 PM
Well I'm actually doing an embedded design, it uses 2 alkaline cells to run a micro with a RFM69CW. All logic works down to 2V, no interface to the outside world.
Edit: Except a programming interface and FTDI serial interface for debugging. For that I power via USB, hence there's a STLQ015M33R 3.3V regulator with VBATT as input and micro/radio power as output. The battery input uses an ideal diode circuit so it can be OR'd together with the USB 5V (which uses a pair of diodes in series first to knock it down a bit) to form VBATT. So this stops the internal power supply going over 3.3V, and still allows the battery to drop to 2V or so, and allows 3.3V signalling for programming and debugging.
Title: Re: Making a lower power Moteino
Post by: joelucid on April 22, 2016, 01:33:36 AM
Quote
The abs minimum for HW is 2.4V. Disregard the Hope datasheets, they are incorrect. There is a component on the board that limits it to 2.4v.

Felix, what's the part that prevents use below 2.4v? I have a project using the rfm69hcw where I've gone a bit overboard with battery life optimizations. It runs from a buck converter off of a 9v battery. During sleep I run the board at the lowest possible voltage by only switching the buck on for less than a ms at a time and then coasting down from a 100uF cap.

This results in a vcc profile that starts at 2.3v and coasts down to around 2V. I've found that I can operate the rfm69hcw down to slightly less than 2v that way in rx - at least well enough that I can detect RSSI.

Of course once I do detect a signal I switch on the buck and bring vcc up to 3.3v so I don't have data on tx from that project. I do have some from my coin cell motes (using the rfm69hw) where I have a feature that monitors vcc during the tx cycle and reports the lowest seen voltage to the gw. When I tested that I sent packets at tx levels normally unsuitable for coin cell motes and could also see success with minimum voltages down to around 2v.

Joe
Title: Re: Making a lower power Moteino
Post by: emjay on April 22, 2016, 05:20:17 AM
@Joe,

The daughter boards are significantly different in component count between the RFM69(C)W and the RFM69HW.
The crystal is just a two active pin crystal and does not force the lower voltage limit.

Title: Re: Making a lower power Moteino
Post by: joelucid on April 22, 2016, 05:26:53 AM
Quote
The daughter boards are significantly different in component count between the RFM69(C)W and the RFM69HW.
The crystal is just a two active pin crystal and does not force the lower voltage limit.

I'm aware of that - I use both. But my experience is even the hw/hcw ones work below 2.4v just as spec'd in the hoperf data sheet.
Title: Re: Making a lower power Moteino
Post by: perky on April 22, 2016, 05:33:28 AM
I think it might be the rf switch, it's the only other component on the reference schematic that could be an active part. Some of those go down to 1.8V. What we could do with if a full BOM of the parts used on both variants, that would sort out the crystal ppm issue too.
Mark.
Title: Re: Making a lower power Moteino
Post by: joelucid on April 22, 2016, 06:00:44 AM
Quote
I think it might be the rf switch

I bet you're right. It seems to be this one: http://datasheet.octopart.com/UPG2179TB-E4-A-NEC-datasheet-26379.pdf

The spec says control voltage needs to be >2.5V for a logic high. That would explain what I saw: I switch to TX at a voltage >2.5V in both scenarios.
Title: Re: Making a lower power Moteino
Post by: joelucid on April 22, 2016, 06:10:19 AM
Btw kind of entertaining: if this is the only problematic component you could still use the module <2.5v using pa0 on rfio. Potentially a useful side effect during the golden years of a coin cell mote  :)
Title: Re: Making a lower power Moteino
Post by: perky on April 22, 2016, 06:24:02 AM
How did you figure out which part they used? Do you have a schematic?
Title: Re: Making a lower power Moteino
Post by: joelucid on April 22, 2016, 06:24:55 AM
With a magnifying glass.
Title: Re: Making a lower power Moteino
Post by: perky on April 22, 2016, 06:29:37 AM
Ah, you can see the G4C marking, ok.
Title: Re: Making a lower power Moteino
Post by: WhiteHare on April 22, 2016, 06:43:02 AM
This would seem to have huge implications if running off two alkaline AA cells, which is what's shown in the photo on the OP of this thread: looking at constant current discharge curves of alkaline AA cells, a significant fraction of the useful life of the battery happens below 1.2v.

So, just to be clear, in your guys judgment based on the RF switch part, we shouldn't be running below 2.4v as far as the RFM69 radio is concerned?

[Edit: Ultimate Lithium AA L91's are a different story.  They would still seem to be a perfect fit, as most of their useful life happens at greater than 1.2v.]
Title: Re: Making a lower power Moteino
Post by: joelucid on April 22, 2016, 06:57:57 AM
The switch only exists in the rfm69hw, not on the w. So I think the w likely runs down to 1.8v. The hw can't switch the high power pa onto the antenna below 2.5v. It still can receive below. It could also send using the low power pa. And it could tx at high power as long as the voltage at the beginning of tx is >2.5v.

The last one of these is important for coin cell motes. There will likely work even though voltage will go under 2.5v during tx since in the beginning voltage is pretty high.
Title: Re: Making a lower power Moteino
Post by: TomWS on April 22, 2016, 07:51:48 AM
With a magnifying glass.
And young eyes!

 ::)
Title: Re: Making a lower power Moteino
Post by: perky on April 22, 2016, 07:54:27 AM
There are some really interesting questions to ask here.

First, why is the rf69 chip specified down to 1.8V for powers less than +17dBm and 2.4V above when the sx1231h specification doesn't? Could it be that Hoperf only wanted to license one chip (the sx1231h) and not two for use in both low and high power modules, and sought assurances from Semtech about these specs before adding them to the rf69? Did that volume then drive the price of the sx1231h down so it is actually cheaper than the sx1231 in volume (seems to be), and Semtech want to keep them separate so the sx1231h doesn't get used in those low voltage, low power applications even though it is fully capable?

In my opinion it looks like the rfm69CW and rfm69W devices will work down to 1.8V, the H variants appear to have a component on board (rf switch) that precludes it and is instead 2.4V.
Mark.
Title: Re: Making a lower power Moteino
Post by: TomWS on April 22, 2016, 07:55:42 AM
The switch only exists in the rfm69hw, not on the w. So I think the w likely runs down to 1.8v. The hw can't switch the high power pa onto the antenna below 2.5v. It still can receive below. It could also send using the low power pa. And it could tx at high power as long as the voltage at the beginning of tx is >2.5v.

The last one of these is important for coin cell motes. There will likely work even though voltage will go under 2.5v during tx since in the beginning voltage is pretty high.
Unfortunately, the RFM69 design does not permit using PA0.  It would be great if it did (you would have full power output range), but it doesn't.  If you want to TX, you need, at a minimum, PA1 turned on.

Tom
Title: Re: Making a lower power Moteino
Post by: joelucid on April 22, 2016, 08:00:15 AM
Quote
Unfortunately, the RFM69 design does not permit using PA0.  It would be great if it did (you  ;) would have full power output range), but it doesn't.  If you want to TX, you need, at a minimum, PA1 turned on.

Sure - but the reason is that the Rf switch disconnects rfio during tx. If the voltage is too low for the switch to work you can suddenly use pa0 even though the design prevents its use when voltage is within spec limits.

I wonder how one would do it though since you don't know whether the switch will work. Maybe switching all pa's would solve that problem. But then you use even more power than usually.

Anyway I didn't really mean that seriously ;)
Title: Re: Making a lower power Moteino
Post by: WhiteHare on April 22, 2016, 08:15:01 AM
So, is it fair to say that if running an HW with AA alkaline cells, the new version of the "optimal solution" is to keep the LDO and run the Moteino using 3xAA batteries?  Because if we really do need 2.5v, and not 1.8v as previously believed, then that 3rd battery really would seem to offer more run-time by allowing you to drain all the batteries all the way down to dead.   
Title: Re: Making a lower power Moteino
Post by: perky on April 22, 2016, 08:24:34 AM
My understanding is the switch switches between RFIO and RF_BOOST (which in turn uses PA1 and PA2), and switches to RF_BOOST during TX. So just using PA0 on RFIO won't work. If there's not enough voltage for the switch to work even the RFIO connection could be dodgy, it might be high impedance but low enough for the receiver to work in 200R mode but too high for transmission with a lower impedance antenna.
Mark.
Title: Re: Making a lower power Moteino
Post by: joelucid on April 22, 2016, 08:33:24 AM
Oh it depends on so many factors that one would have to test it. Personally I think 2xaa batteries which are just able to keep the voltage >1.8V over a full Tx cycle of say 10ms at 120mA might well have an open circuit voltage >2.4v. Whether you could still turn the switch at the very beginning - who knows. Try it out and let us know!
Title: Re: Making a lower power Moteino
Post by: Felix on April 22, 2016, 09:03:27 AM
Good catch @perky, indeed the RF switch is the UPG2179TB, and yes the limitation is actually 2.5v, with a "typical"  of 3.0v. So even the 2.5v is a pretty edgy use case.
When I build some RFM69 clones I tried to find that switch but it's obsolete, not stocked anywhere so I had to use a different one. The switch Hope uses is found on virtually all boards they make (semtech based at least) including the LoRa radios and has been used for years. It seems they either bought millions some time back and the supply is limited, or it's a clone from asia, or maybe it's still source-able through more obscure means.
Anyway it was fun making some RFM69s:

(https://farm1.staticflickr.com/584/21088513486_b5e2ea546f_n.jpg) (https://flic.kr/p/y8w9m7) (https://flic.kr/p/y8w9m7)

Does anyone know if the RF topology of the 1231h allows for an integrated component to move all the matching and perhaps the RF switch into a single component (and what that component type might be)?

[This should maybe be a new topic]
Title: Re: Making a lower power Moteino
Post by: WhiteHare on April 22, 2016, 09:24:50 AM
@Felix  Looking at the datasheet, it only gives the recommended operating voltage range for 25C.  I don't see any graphs showing the effects of different temperatures.  Do you have any info as to how sensitive those numbers are to temperature? 
Title: Re: Making a lower power Moteino
Post by: Felix on April 22, 2016, 09:36:22 AM
@Felix  Looking at the datasheet, it only gives the recommended operating voltage range for 25C.  I don't see any graphs showing the effects of different temperatures.  Do you have any info as to how sensitive those numbers are to temperature?
Unfortunately no, I don't have more than what the datasheet reveals.  :-\
Title: Re: Making a lower power Moteino
Post by: perky on April 22, 2016, 09:40:24 AM
@Felix: Wow, good stuff Felix! I have to say the build quality of yours compared to theirs is so much better, just look at the soldering :-O

@WhiteHare: I think these things work by forward diasing diodes to enable rf through and reverse biasing to block, if that's the case and it isn't just a logic level thing then there could well be a temperature dependence.

Mark.
Title: Re: Making a lower power Moteino
Post by: joelucid on April 22, 2016, 11:03:49 AM
Felix, have you ever tried to use a RFM69HW at less than 2.5V or is the assumption that it doesn't work based on the data sheet alone?
Title: Re: Making a lower power Moteino
Post by: Felix on April 22, 2016, 11:25:17 AM
Felix, have you ever tried to use a RFM69HW at less than 2.5V or is the assumption that it doesn't work based on the data sheet alone?
Not that I can remember.
I know you like to push things to the limit, I do too sometimes (ie Moteino 16mhz @ 3.3v). But with this RFM69 at less than 2.5v it can be very difficult to debug if it starts to edge into some unstable condition, and it's easy for people to point fingers in other directions. That is why I would not run it at less than that.
Title: Re: Making a lower power Moteino
Post by: WhiteHare on April 22, 2016, 12:52:21 PM
Funny how perspectives change.  In the face of uncertainty, keeping the LDO and running from a 9-volt battery "just to be sure" suddenly looks like an attractive, compact, and rational choice if running an HW....  I don't really feel like running a ton more experiments just to test whether the datasheet might be wrong.
Title: Re: Making a lower power Moteino
Post by: joelucid on April 22, 2016, 01:33:56 PM
Quote
Funny how perspectives change.  In the face of uncertainty, keeping the LDO and running from a 9-volt battery "just to be sure" suddenly looks like an attractive, compact, and rational choice if running an HW....  I don't really feel like running a ton more experiments just to test whether the datasheet might be wrong.

Well lets not jump to conclusions. I'm running a test now: RFM69HW on two aa Eneloops with 3 Ohm in parallel - shouldn't take too long. As of now it's happily transmitting at 17 dBm at 2.2V, creating an RSSI of -34 at the receiver.

One interesting tidbit: The RFM69HW Spec V1.0 specifies minimum voltage at 2.4V. V1.3 has the new minimum voltage of 1.8V for 17dBm. So this was an active modification of the data sheet rather than just an error in the spec.
Title: Re: Making a lower power Moteino
Post by: perky on April 22, 2016, 01:38:52 PM
Mmmm, if I remember the old datasheet said 2.4V specified over the the full power range, whereas the new one breaks it down into finer detail so yes, this is a clarification rather than a correction. Similar to the difference now between the RF69 and sx1231h datasheets.
Mark.
Title: Re: Making a lower power Moteino
Post by: joelucid on April 22, 2016, 04:09:18 PM
Quote
Well lets not jump to conclusions. I'm running a test now: RFM69HW on two aa Eneloops with 3 Ohm in parallel.

Interesting - I almost needed to pinch myself. Luckily I had a Moteino with regulator removed and fuses set to no BOD lying around. As I said with this setup at 17dBm the receiver saw the test packages sent 2.2V at -34 RSSI in the beginning. From there is just continued down, very reliably, less than 0.3% packet loss.

It continued all the way down to 1.69V when the 328p just couldn't keep up appearances and suffered from ram corruption. Until then the radio hadn't missed a beat:

Code: [Select]
Apr 22 22:03:18 espgw.lx  [11] [RX_RSSI:-36,1,8,45]
Apr 22 22:03:18 espgw.lx  test4 -> nd:11,nr:285,pw:12,vc:170,rt:3
Apr 22 22:03:21 espgw.lx  [11] [RX_RSSI:-37,1,8,45]
Apr 22 22:03:21 espgw.lx  test4 -> nd:11,nr:286,pw:12,vc:170,rt:3
Apr 22 22:03:24 espgw.lx  [11] [RX_RSSI:-35,1,8,45]
Apr 22 22:03:24 espgw.lx  test4 -> nd:11,nr:287,pw:12,vc:170,rt:3
Apr 22 22:03:27 espgw.lx  [11] [RX_RSSI:-36,1,8,45]
Apr 22 22:03:27 espgw.lx  test4 -> nd:11,nr:288,pw:12,vc:170,rt:3
Apr 22 22:03:30 espgw.lx  [11] [RX_RSSI:-35,1,8,45]
Apr 22 22:03:30 espgw.lx  test4 -> nd:11,nr:289,pw:12,vc:170,rt:3
Apr 22 22:03:33 espgw.lx  [11] [RX_RSSI:-38,1,8,45]
Apr 22 22:03:33 espgw.lx  test4 -> nd:11,nr:290,pw:12,vc:170,rt:3
Apr 22 22:03:36 espgw.lx  [11] [RX_RSSI:-35,1,8,45]
Apr 22 22:03:36 espgw.lx  test4 -> nd:11,nr:291,pw:12,vc:170,rt:3
Apr 22 22:03:39 espgw.lx  [11] [RX_RSSI:-35,1,8,45]
Apr 22 22:03:39 espgw.lx  test4 -> nd:11,nr:292,pw:12,vc:170,rt:3
Apr 22 22:03:45 espgw.lx  [11] [RX_RSSI:-38,1,8,45]
Apr 22 22:03:45 espgw.lx  test4 -> nd:11,nr:294,pw:12,vc:169,rt:17
Apr 22 22:03:51 espgw.lx  [11] [RX_RSSI:-38,6,8,45]
Apr 22 22:03:51 espgw.lx  test4 -> nd:11,nr:296,pw:12,vc:169,rt:30
Apr 22 22:03:54 espgw.lx  [11] [RX_RSSI:-37,1,8,45]
Apr 22 22:03:54 espgw.lx  test4 -> nd:11,nr:297,pw:12,vc:169,rt??.
Apr 22 22:04:00 espgw.lx  [11] [RX_RSSI:-38,1,8,45]
Apr 22 22:04:00 espgw.lx  test4 -> nd:11,nr:299,pw:12,vc:169,rt:33
Apr 22 22:04:06 espgw.lx  FreeHeap: 17440
Apr 22 22:04:06 espgw.lx  [11] [RX_RSSI:-38,1,8,45]
Apr 22 22:04:06 espgw.lx  test4 -> nd:11,nr:301?test4 -> ??#014?B?00?WZ6Q^??
Apr 22 22:04:12 espgw.lx  [11] [RX_RSSI:-36,1,8,45]
Apr 22 22:04:12 espgw.lx  test4 -> nd:11,nr:303,pw:12,vc:169,rtNYV
Apr 22 22:04:15 espgw.lx  [11] [RX_RSSI:-36,1,8,45]
Apr 22 22:04:15 espgw.lx  test4 -> nd:11,nr:304,pw:12,vc:169,rtNYV
Apr 22 22:04:43 espgw.lx  [11] [RX_RSSI:-38,6,8,45]
Apr 22 22:04:43 espgw.lx  test4 -> nd:11,nr:313I?ج?"?A?k???t]#021
Apr 22 22:04:46 espgw.lx  [11] [RX_RSSI:-35,6,8,45]
Apr 22 22:04:46 espgw.lx  test4 -> nd:11,ns:314,pw:12,vc:168,rt???F
Apr 22 22:04:55 espgw.lx  [11] [RX_RSSI:-35,6,8,45]
Apr 22 22:04:55 espgw.lx  test4 -> nd:11,ns:317?-r:2??#006?#6???.
Apr 22 22:04:55 espgw.lx  [11] [RX_RSSI:-36,1,8,45]
Apr 22 22:04:55 espgw.lx  test4 -> nd:11,ns:317?o*'`???#013\??!?|a#021??K


RSSI pretty much unchanged at -36 vs -34 at 2.2V.

So I guess this should put the hypothesis that the rfm69hw only works down to 2.5V to a conclusive rest.

Joe
Title: Re: Making a lower power Moteino
Post by: WhiteHare on April 22, 2016, 05:27:29 PM
Are you stopping there, or will you be testing it all the way down to 1.8v?
Title: Re: Making a lower power Moteino
Post by: joelucid on April 22, 2016, 05:34:04 PM
Reread the post. It ran down to 1.68v!
Title: Re: Making a lower power Moteino
Post by: WhiteHare on April 22, 2016, 07:19:07 PM
Great work!  And fast too!

I'm perplexed though.  The fact that it goes so low is great and definitely more reassuring than if it were a higher number, but, trying to be objective about this:  why is there such a seemingly large disconnect between that result and the minimum voltage shown on the datasheet that Felix linked to?  Can anyone here perhaps shed some light on that?  For example, if, hypothetically, all the MMIC chips ever used on the RFM69 modules were tested using Joe's method, might there be a wide distribution of minimum voltages in the MMIC chips that got manufactured, and perhaps Joe simply had good fortune with the one he happened to test?  Or does it not work like that?  I presume the manufacturer has a large dataset from ongoing QA testing, and so I would think they'd be in the best position to put solid numbers in their datasheet.  I should think it would undermine their sales to state a higher minimum voltage if a lower minimum would work just as well, and so I don't know why they would do that if there were no reason to.
Title: Re: Making a lower power Moteino
Post by: TomWS on April 22, 2016, 10:15:15 PM
RSSI pretty much unchanged at -36 vs -34 at 2.2V.

So I guess this should put the hypothesis that the rfm69hw only works down to 2.5V to a conclusive rest.
But, but... RSSI of -36?  Were these two devices like right next to each other?  What's the behavior when the starting RSSI is -54?  Or a more realistic -64?   AGC can mask a wide range of sins.

Tom
Title: Re: Making a lower power Moteino
Post by: joelucid on April 22, 2016, 10:30:52 PM
WhiteHare, I think we just don't know. Is this a Chinese clone with different specs? Who knows. But for the module: if it's spec'd like a duck and performs like a duck I think one has to assume it's a duck. 3 tiny letters on a component notwithstanding.

Tom, actually the gw still has the agc testing code from recently. The 1 after each measurement shows that agc selected the highest gain again for each packet. Proving again that agc just doesn't work with the current rfm69 library.

I have a fixed version where it does work. In that case I measure RSSI of less than -20 for nodes as close together as these two. But using a rssithresh below noise floor actually has some advantages. I think I wrote all about it in that agc thread.
Title: Re: Making a lower power Moteino
Post by: Felix on April 25, 2016, 08:39:57 AM
joelucid, great testing, it's definitely a good thing to see this work down to 1.8v and even below. I see this as a good margin overclock case, same as the atmega328p running 16mhz at 3.3v without a hitch.
So I guess a very low voltage mote can be achieved, running from a combined 2xAA to depletion without worry that the components can't handle it (at least not the radio).
We should split this topic, what's a good starting post to do that? I'd like to continue the discussion in the direction of narrowing down some specs for such a mote. My interest is growing. :)
Title: Re: Making a lower power Moteino
Post by: joelucid on April 25, 2016, 09:17:16 AM
Quote
We should split this topic, what's a good starting post to do that?

Seems to me all but the very first message are off the original topic. Speaks to our forum discipline :-)

I think you could separate right there.

In terms of specs I would think:

a) Eliminate the regulator
b) Eliminate the resonator
c) Potentially switch to 328pb

Then comes the timer question. Part of this is being able to run lower power than with the WDT or listen mode. And part of it is being able to track time more precisely. I think some options are:

1) tpl5010 - super simple, super small, very low power timer. Not precise.
2) am1815 - super low power RTC, not temp corrected.
3) 32.768khz crystal on 328p - this gives you a precision timer at <1uA. But a lot needs to be done in software.

1) could probably just take the place of the regulator given how small it is. Same for 3) and the resonator. If you want to add 2) you probably need to let go the flash - it wouldn't hurt me, I've never had much use for it.

With 1) or 2) you'd have the lowest power arduino clone on the market. At least last time I checked nobody had that.

I think 2) is more attractive overall. But it has the difficult package issue.

I guess another thing to consider is throwing in a si7021 at least as option. Everybody should have a handful TH motes at home and it would be great to have something less expensive than the Moteino + Weathershield option available.

Joe
Title: Re: Making a lower power Moteino
Post by: SadE54 on April 25, 2016, 09:59:34 AM
What are the benefits of the atmega328pb over 328p ?
Title: Re: Making a lower power Moteino
Post by: joelucid on April 25, 2016, 10:11:24 AM
Quote
What are the benefits of the atmega328pb over 328p ?

Cheaper and more peripherals.
Title: Re: Making a lower power Moteino
Post by: SadE54 on April 25, 2016, 10:37:00 AM
I looked at the datasheet . Seems nice : 3 16 bits counters !  ;)
I guess this one is not officially supported by the Arduino platform ?
Title: Re: Making a lower power Moteino
Post by: WhiteHare on April 25, 2016, 12:03:43 PM
FWIW, I'd like to see an am1815 on a Moteino.  It's one of the flagship lowpower RTC's, and it has a small footprint.  It's also difficult to hand solder.  I think those qualities taken together might help some people justify buying a pre-made board (i.e. buy rather than make).
Title: Re: Making a lower power Moteino
Post by: Felix on April 25, 2016, 02:02:38 PM
@joelucid,
RE the MCU:
a) and b) are no problem. This would imply different fuses also. Maybe fuses should also be discussed.
c) This will likely not be a reality due to lack of support. When I switch to 328pb it will be everything I make.

RE the timer:
Unfortunately am1815 cannot be sourced from any major reputable distributors. The RTC is available here (http://www.digikey.com/product-detail/en/abracon-llc/AB1815-T3/535-11936-1-ND/3661487).
32k crystal + lots of work also probably not a good option. Simple is still a feature i'd like to keep.

Si7021 could be included. The 4MBIT FLASH may also be optional, regardless of anything, having memory is attractive to lots of people.

- We should also start talking about battery.
- We should also start talking about board format, layout, pinout, perhaps battery connections depending on the battery.
- We should also start talking about programmability (ISCP, FTDI) and powering it when attached to a FTDI. I am thinking it should only be able to work when its own power supply is attached (ie battery), since the 5V from FTDI will fry it (no regulator) - the FTDI also has a 3.3v power output (via solder jumper) but it's limited to 50mA so that won't do it either.
Title: Re: Making a lower power Moteino
Post by: joelucid on April 25, 2016, 02:40:08 PM
Felix, I'll have more on the other items later. Just quickly: the am1815 is broadly sold by Abacon as ab1815. Digikey, mouser etc all carry it. It's fairly inexpensive and VERY low power.
Title: Re: Making a lower power Moteino
Post by: Felix on April 25, 2016, 03:16:26 PM
Felix, I'll have more on the other items later. Just quickly: the am1815 is broadly sold by Abacon as ab1815. Digikey, mouser etc all carry it. It's fairly inexpensive and VERY low power.
Doh, I just searched to AM1815 so of course I got nothing. AB1815 reveals it (http://www.digikey.com/product-detail/en/abracon-llc/AB1815-T3/535-11936-1-ND/3661487). Thanks for pointing that out. At $1.95 (1qty) it's quite a pricey part compared to other things. The required crystal (http://www.digikey.com/product-detail/en/abracon-llc/ABS07-120-32.768KHZ-T/535-11937-1-ND/3724046) adds another $0.70.
In 1000qty this combo would come to $1.31, not bad.
Which of the AB18X5 variants would be preferrable? I2C or SPI?
How about the more package friendly (still pretty low power at 1uA) MCP7951x/MCP7952x (http://ww1.microchip.com/downloads/en/DeviceDoc/22300A.pdf)?
Title: Re: Making a lower power Moteino
Post by: joelucid on April 25, 2016, 04:11:38 PM
I've used i2c but realized that spi would have been much better: when you have an rtc the temptation is to use it listen mode style. That means setting a couple of registers every other second or so. And with i2c that takes too much time/power.

I think 1uA is too much these days for simply keeping the time.
Title: Re: Making a lower power Moteino
Post by: Felix on April 25, 2016, 05:07:33 PM
Right, so it sounds like the 1815 variant.
1uA is too much yes, when 15nA is an option :)
That and the fact that the MCP is scarce would make the AM1815 a good option (ignoring the QFN package).
Title: Re: Making a lower power Moteino
Post by: joelucid on April 26, 2016, 02:12:09 AM
Quote
- We should also start talking about battery.

Of course you can run these from all kinds of batteries. The li-Mno2 chemistry is appealing because of all the small button cell types (cr2032, cr2450) as well as the small cylindrical shapes (cr123a or craa/2). But 2x LiFeS02 or even alkali - it all works. The Li-SOCl2 cells like the Saft are appealing if a little high-voltage because of their energy density and near constant discharge curve. Li-(CF)x (the BR coin cells) is great for long lived projects because of the low self discharge (and this actually matters once you have a 150nA sleep load!).

I think it would be nice if a coin cell or coin cell holder could directly be mounted for smaller form factor - similar to Tino or my small TH mote.

Quote
- We should also start talking about board format, layout, pinout, perhaps battery connections depending on the battery.

I personally find that in most my motes I don't use close to all pins. So maybe doing something with a less complete pinout would be attractive in particular for the coin cell powered moteino. That said it would also be useful to have the same pinout across the board - I'm not sure what's best.

Generally I find that Motes dedicated to one purpose tend to have the better form factor in the end. So I could also see you making a TH-Mote that only exposes a hand-full of extra pins for expansion.

Quote
- We should also start talking about programmability (ISCP, FTDI) and powering it when attached to a FTDI. I am thinking it should only be able to work when its own power supply is attached (ie battery), since the 5V from FTDI will fry it (no regulator).
- the FTDI also has a 3.3v power output (via solder jumper) but it's limited to 50mA so that won't do it either.

Good points. I actually used the 3.3V FTDI output for powering modded Moteinos without regulator. I wasn't even aware of the limitation. I think it would be better to have a new FTDI connector with regulator on board than to have batteries connected while programming. Batteries won't be able to sustain the programming currents - in particular the coin cell types.

Joe
Title: Re: Making a lower power Moteino
Post by: joelucid on April 27, 2016, 10:57:03 AM
Maybe another reason for preferring a non-regulator Moteino:

As I measured bit error rates today I converted a TH Mote into something I could use by replacing the cr2032 with 2 AAA eneloop. That worked well as described in the other thread.

Now as I had this ready I also wanted to test the noise floor on it using my methodology. The results are by far the lowest numbers I have ever measured:

Apr 27 16:50:57 espgw.lx  250:1 251:3 252:5
Apr 27 16:51:14 espgw.lx  251:2 252:4 253:5
Apr 27 16:51:30 espgw.lx  250:1 251:3 252:4 253:5
Apr 27 16:51:47 espgw.lx  251:3 252:4 253:5
Apr 27 16:52:03 espgw.lx  251:5
Apr 27 16:52:19 espgw.lx  250:1 251:3 252:3 253:5
Apr 27 16:52:36 espgw.lx  250:2 251:5 252:3 253:5
Apr 27 16:52:52 espgw.lx  249:1 251:3 252:4 253:5
Apr 27 16:53:08 espgw.lx  250:1 251:3 252:5
Apr 27 16:53:25 espgw.lx  250:2 251:3 252:4 253:5
Apr 27 16:53:41 espgw.lx  251:5

This is 5khz rxbw. On this mote you could run this profile with an RSSITHRESH of 249 (-125 dBm) and still use AGC. Pretty cool. I suspect it's a benefit of not getting regulator noise. I do have a 100nF cap directly at the radio pin.
Title: Re: Making a lower power Moteino
Post by: kobuki on September 07, 2016, 11:48:49 AM
Felix, is it reasonable to expect this product this year or even in the near future? Are planned features fixed yet in any case?
Title: Re: Making a lower power Moteino
Post by: Felix on September 08, 2016, 08:37:18 AM
I will first have to be convinced by the economics if I was to "mass produce" anything like what's mentioned here :)
The [ low power + cost to make/test/own + ease of use compared to a regular Moteino ] would have to be very convincing.
Title: Re: Making a lower power Moteino
Post by: kobuki on September 08, 2016, 08:45:18 AM
OK, in my interpretation it's "not going to happen" - though I think there would be significant demand, seeing the prevalence of the myriad of battery operated nodes everywhere and the ULP operation it obviates. We'll continue to desolder the LDO then :)
Title: Re: Making a lower power Moteino
Post by: Felix on September 08, 2016, 09:39:02 AM
I didn't say it won't happen. I plan to make a new Moteino type some time in the (near?) future. I just have to be very careful in what I put on it. IOW it's my job to make sure this will be a desirable node with features (including low power) that will distinguish it from the others, such that people will buy it and find it easy to use at a reasonable price. Otherwise I am wasting my time and money developing something that will be wanted by 10 people who are extreme low power fans. Will it be as described here? I don't know yet.
Sure, 20uA is great, but is 20nA vs 4uA worth the cost/complexity increase and usability decrease? I have to carefully evaluate that. There's a point where going so much lower power makes no sense because batteries shelf life don't last as long as that node could, or because technology changes and may obsolete that hardware, or because the owner won't live 50 years more to see that node deplete its battery.
Title: Re: Making a lower power Moteino
Post by: kobuki on September 08, 2016, 09:51:46 AM
Yes, I completely understand the reasoning. Though I think a new sister in the existing Moteino family could be sold by its established brand name. Imagine "New moteino using even less power!" :) You're well-known in the field and this relatively small, but ever-growing hobby and semi-pro market. For me the most useful change would be periodic wakeups with a precision timer (+RTC) and the possibility to add batteries below 3.3V like 2xAAA or CR123 without a mod. I agree, going below a certain sleeping power is senseless.
Title: Re: Making a lower power Moteino
Post by: Felix on September 08, 2016, 10:31:11 AM
Sure, all things to consider.
The one challenge and consistent theme I see is that no matter how wide range of features a board has, someone wants something else.
I guess focusing on low voltage and low power may be the delimiter between the general purpose Moteino which works with inputs up to 16v, and this new node type that can go to 1.8v perhaps.
Title: Re: Making a lower power Moteino
Post by: ChemE on September 08, 2016, 10:49:07 PM
I echo what others have suggested especially the RTC and Si7021 although the HTU21D has a lower sleep current if the datasheet is to be believed.  Getting back to fuses, would it make sense to fuse them to have BOD disabled so that they resume from sleep faster?  Also what about selecting components that are stable down to 1.8V as a method of further reducing power consumption?  There was some discussion as to whether or not the RFM69 could work at 1.8V but I didn't see anything conclusive.  The picopower uC's are good to 1.8V and the TH sensors are as well.
Title: Re: Making a lower power Moteino
Post by: Felix on September 09, 2016, 08:13:12 AM
The RFM69W will work at 1.8v "officially". The RFM69HW has an RF switch that requires 2.4v, but others have reported it working down to 1.8v without problems.
Title: Re: Making a lower power Moteino
Post by: kobuki on September 09, 2016, 10:26:20 AM
I echo what others have suggested especially the RTC and Si7021 although the HTU21D has a lower sleep current if the datasheet is to be believed.

I think putting on all kinds of sensors is not a very good approach. This is the point where it's very hard to satisfy everyone. There exist various easy to use modules and even LPL shields for T/H measurement and other things. Having a general purpose LDO-less board with a(n optional) timer or RTC would be simpler for both the manufacturer and the clients. Maybe solder-clad footprints for adding several kinds of sensors would be nice. For instance, Si7021 and the HTU21D are interchangeable IIRC. Setting clock source or BOD fuses is a somewhat more advanced task but can be done any time after purchase when flashing.
Title: Re: Making a lower power Moteino
Post by: ChemE on September 09, 2016, 04:20:48 PM
Felix,

You asked on the previous page about battery discussions.  For projects I see myself taking on running for 2 decades outweighs being easily swallowed so I prefer to run off 2xAAA rather than a coin cell.  2xAAA are also almost exactly the same width as a Mote just a bit longer; gives enough room to add a WDT and a footprint for a Si7021 (or Si7034 if we want to go nuts on lowest power).  An arduino+radio that can run on 2xAAA for 10+ years and fits in your mouth; surely that would sell.

PS thanks for the lightning fast processing and shipping of my two TPL5110s!
Title: Re: Making a lower power Moteino
Post by: Felix on September 09, 2016, 05:46:31 PM
For projects I see myself taking on running for 2 decades outweighs being easily swallowed so I prefer to run off 2xAAA rather than a coin cell.
An arduino+radio that can run on 2xAAA for 10+ years and fits in your mouth; surely that would sell.

PS thanks for the lightning fast processing and shipping of my two TPL5110s!

I agree and would rather do AAA than coin cell which will be a black hole of support for the outraged pack that will discover coin cells won't yield 10km at full TX power and isn't as plug and play as the "industry standards". 2XAA(A) can do all that with the defaults, well even now with a regulator-less Moteino. I like the easy to swallow concept :) Maybe my next company will be Swallow Labs, where everything made is can be easily swallowed, lol.

PS: My pleasure ;)
Title: Re: Making a lower power Moteino
Post by: MoebiusL on September 29, 2016, 10:01:35 PM
I built a very simple temperature sensor node some time ago that basically reads the temp from a TMP36 and sends it to a base station (a Moteino) every 4 seconds and sleeps (the real application will be sleep for much longer, but I'm speeding things up just for testing).

I've been trying to run Moteino from Li-Po for while, but for long running applications didn't work very well because of the self-discharge I guess. Also tried 2xAA without regulator and was only able to run down to 1.2V per cell as mentioned before on this topic (maybe because the TMP36 needs at least 2.4V). Also tried using 3xAA with the regulator but didn't get much of it either, it's a shame that RFM69 does not support 5V, so I could connect the 3xAA directly!

The idea of running at very low voltages is great, but I think it's challenging to have all sensors and components running at 1.8V... maybe in the future new sensors will lower the working voltage to help us  :D

I'm now testing some boards called Whisper Node, that I got from Kickstarter (https://bitbucket.org/talk2/whisper-node-avr/overview (https://bitbucket.org/talk2/whisper-node-avr/overview)). They are quite similar with the Moteino but there are loooots of tiny components. At the moment I'm running a similar basic temperature node with a single AA cell. The board has a step-up and when things are at sleep consumes a bit over 5uA from the battery, the whole board runs at 3.3V and MCU+temp.sensor seems to be happy.

The one thing I didn't like much is that many pins from the MCU are already used by the Whisper Node itself, like 2 voltage dividers, 2 leds and 2 buttons... it can be handy if you're using those, but won't be my case. Apparently you can cut the traces to free the pins for your own use, but just feels like I'm destroying the board - I guess everybody has different needs as Felix mentioned.

Well I'm just running the board for 3 weeks now and still can't notice a real voltage drop of the battery, I'll try to post something back when I get a bit more results... but I guess a step-up regulator might be the answer for lower voltages. I'll try to get a stand-alone step-up regulator, maybe this one: https://www.pololu.com/product/2561 to replace the LDO on one of my Moteino to do some testings as well...
Title: Re: Making a lower power Moteino
Post by: kobuki on September 30, 2016, 05:22:47 AM
Interestingly what you write doesn't really coincide with others' experience. Did you measure the power consumption with the original regulator-less boards during sleep? 2.4V is sure to be a limiting factor, though.
Title: Re: Making a lower power Moteino
Post by: MoebiusL on October 01, 2016, 07:44:10 AM
@kobuki, yes and the consumption was very low, never more than 2-4uAs. The problem with regulator-less is that the whole thing starts to become unstable as the AA voltage drops below 2.4V, even reducing the speed, as seems that the sensor doesn't like it. Even with some good results, at the end it was leaving lots of un-usable energy in the batteries.

That's why I'm doing some experiments using step-up regulator options, to try to extract as much as possible from the Alkaline AAs even if it cost extra 1uA or 2uA during the sleep. But at the end I'm getting a stable 3.3V all the way down to 0.8V on a single cell (I hope, as the tests still running).
Title: Re: Making a lower power Moteino
Post by: perky on October 01, 2016, 07:54:32 AM
One big problem with boost regulators is their start-up current, it can cause a current spike of several hundred mA if you're not careful. This is a problem when batteries are depleted because their internal resistance shoots up, a large current spike will drop the voltage to below the working threshold which will necessarily require the end point of the batteries to be higher . So you need some kind of slew rate control on the output of the boost regulator, or a programmable soft start.
Mark.
Title: Re: Making a lower power Moteino
Post by: kobuki on October 01, 2016, 07:59:09 AM
@MoebiusL, have you thought of using some other kind of temp. sensor that can go lower?
Title: Re: Making a lower power Moteino
Post by: joelucid on October 01, 2016, 08:13:26 AM
I would recommend to ditch the tmp36 for a si7021 and run off of 2x aaa without regulator. Sleep using the radio as wake-up timer gives you around 1.3uA sleep current. This is much simpler, quieter and more power efficient than using a booster. It would operate down to around 1.9v. Prepare not to change batteries in your life if using lithium cells.
Title: Re: Making a lower power Moteino
Post by: TomWS on October 01, 2016, 08:32:14 AM
That's why I'm doing some experiments using step-up regulator options, to try to extract as much as possible from the Alkaline AAs even if it cost extra 1uA or 2uA during the sleep. But at the end I'm getting a stable 3.3V all the way down to 0.8V on a single cell (I hope, as the tests still running).
I would opt for Lithium AAs rather than Alkaline and a voltage converter as you'll get useful voltage to the end of life without conversion inefficiency and much better low temperature performance.

Tom
Title: Re: Making a lower power Moteino
Post by: WhiteHare on October 01, 2016, 09:16:06 AM
One big problem with boost regulators is their start-up current, it can cause a current spike of several hundred mA if you're not careful. This is a problem when batteries are depleted because their internal resistance shoots up, a large current spike will drop the voltage to below the working threshold which will necessarily require the end point of the batteries to be higher . So you need some kind of slew rate control on the output of the boost regulator, or a programmable soft start.
Mark.
Are there any boost chips that come with that kind of spike avoidance already built into them?  It would be handy to have a "go to" chip for those sensors that might need to run for short duration but at higher minimum voltages like 3.3v or 5v.  Even just knowing the proper keyword to look for would help in maybe finding something applicable in, say, Linear Technology's catalog of boost converters (of which there are many different types so as to cover a whole gamut of different use cases).
Title: Re: Making a lower power Moteino
Post by: perky on October 01, 2016, 12:17:56 PM
You can search for soft start external capacitor boost regulators, one such part is the ADP1612 that has an external cap:
http://www.farnell.com/datasheets/1793793.pdf

Note that a lot of boost regulators have a DC path from Vin through the inductor and the diode to Vout so they need a load switch on the front of it too if you want to isolate completely.

Some useful stuff:
http://www.analog.com/library/analogDialogue/archives/45-09/boost.html

Mark.
Title: Re: Making a lower power Moteino
Post by: MoebiusL on October 02, 2016, 11:32:30 AM
Good inputs, I just checked and the board I'm doing some trials uses a MCP16251 boost, just checked the datasheet and it has built-in soft-start (typical 1.5ms) and regulates the output via PWM and PFM depending on the required current, which seems to optimize the performance for small and larger currents.

Regarding the TMP36, that's just a test sketch. On other projects I need to have some voltage over 3.0V because of an ultrasonic sensor (US-100), etc. So going much lower without regulator limits too much the possibilities. I'll indeed try the lithium cells, shame they still a bit pricey. The conclusion I got is that's difficult to find a fit-all size when talking about low-power, you basically need to adapt the board for each situation.
Title: Re: Making a lower power Moteino
Post by: perky on October 02, 2016, 12:38:06 PM
The choice of boost mode of operation is quite important. The MCP16251 for example uses a scheme where the current ramps up to a certain max value then turns off which pumps that energy into the ouput, this max current limit value is designed for maximum output current and is usually about twice the maximum output current so even though the max current is limited to 100mA during the startup it can still take large current spikes when it is regulating (see figure 2-21 of the MCP16251 datasheet that shows even with 15mA output current the peaks are still 250mA or so).

So if you really want to reduce spikes to a minimum the best option is to choose a regulator that you can also control the maximum peak current. For example, the NCP1406 uses a max ON time scheme of 0.9us, it will turn off ramping up the current through the inductor if it either reaches its max current, or after 0.9us. This allows you to put larger value inductors in (which have a slower ramp) and effectively control the peak current through the inductor, and hence also through the input. You may for example only need 15mA of output current so you don't really want 250mA pulses at a low pulse rate if you can avoid it.

There's still a problem with the current ramping up at startup even with the NCP1406 with its soft start (3ms typical) and it's not isolated, so I put a slew rate controlled load switch on the front of mine as well. BTW I probably wouldn't use the ADP1612 for those reasons too, that shows an external cap for slew rate control but is quite a beefy chip for low current solutions.

Mark.
Title: Re: Making a lower power Moteino
Post by: steve v on October 04, 2016, 09:24:10 PM
Does anyone have a simple / lowest labor solution to bypassing the Linear Regulator on the Moteino ?
I plan on running an RFM95 Moteino with a pair of Lithium L91 AA batteries and I think it makes sense to take the regulator out of the equation.

Do I need to remove the device, and solder a bridge ? or Just add a jumper wire somewhere ?
Title: Re: Making a lower power Moteino
Post by: TomWS on October 04, 2016, 11:14:06 PM
Does anyone have a simple / lowest labor solution to bypassing the Linear Regulator on the Moteino ?
I plan on running an RFM95 Moteino with a pair of Lithium L91 AA batteries and I think it makes sense to take the regulator out of the equation.

Do I need to remove the device, and solder a bridge ? or Just add a jumper wire somewhere ?
IMO, if you're not using the VR, take it out.  It's easy to do (hint, lift single leg side first, then each of the other two).  I have a 'drawer' full of MCP1703s taken off Moteino boards that I use elsewhere - thanks, Felix!

Tom
Title: Re: Making a lower power Moteino
Post by: steve v on October 15, 2016, 02:31:32 PM
Thanks Tom,

I found I could remove the regulator with a wire snipper  and then power the Moteino from the 3.3V pin.   My board stops running at 2.81V

Does it make sense to reconfigure the Moteino 328P to  run below the 2.81 V,  any suggestions on how to do this ? I don't see any sketch examples for this ?  Or is this a fuse configuration ?
Title: Re: Making a lower power Moteino
Post by: TomWS on October 15, 2016, 06:45:31 PM
Thanks Tom,

I found I could remove the regulator with a wire snipper  and then power the Moteino from the 3.3V pin.   My board stops running at 2.81V

Does it make sense to reconfigure the Moteino 328P to  run below the 2.81 V,  any suggestions on how to do this ? I don't see any sketch examples for this ?  Or is this a fuse configuration ?
If you want to run reliably below 3.3v you'll need to change the fuses. The first change is the oscillator source - that will need to change to 8mhz or less.  Also check the BOD settings.  They should be set off or below 2.7V if you're going to run close to that.
Tom
Title: Re: Making a lower power Moteino
Post by: curious on September 13, 2017, 04:20:04 PM
Actually I run a project which builds Moteino compatible clones without voltage regulator @ https://www.canique.com (https://www.canique.com)

The problem with an Arduino Mini Pro e.g. or a standard Moteino is that if you run it with low voltages, you have high current consumption. E.g. a standard moteino will draw ~55uA @ 2.4V because of the voltage regulator.
A voltage regulator only makes sense when running above 3.3V.
Title: Re: Making a lower power Moteino
Post by: Felix on September 15, 2017, 05:08:59 PM
Actually I run a project which builds Moteino compatible clones without voltage regulator @ https://www.canique.com (https://www.canique.com)

The problem with an Arduino Mini Pro e.g. or a standard Moteino is that if you run it with low voltages, you have high current consumption. E.g. a standard moteino will draw ~55uA @ 2.4V because of the voltage regulator.
A voltage regulator only makes sense when running above 3.3V.

Awesome rebadge :)

Hey folks BTW I just released the vanilla Moteino with no-LDO, this batch has been sitting on my desk for about 6 weeks, just one of those things that get back burner-ed unintentionally. Sorry ChemE and other enthusiasts who have been asking for this variant.
Title: Re: Making a lower power Moteino
Post by: mielleriealphonse on November 11, 2017, 05:00:26 AM
Great Hacking of your own material ! Like it !

I am facing the challenge of having solar battery for one project. To solve that in a pratical and simple way I have decided to design a solar power supply feedding a small 600mAH LiFePo4. (3.2V) with a very small solar panel of 5V. This type of battery is a good choice to have the full power of RFM module and to have a very stable Voltage and most probably a long life (expecting 10 years).

Of course Moteino LDO need to be remove like in the vanilla version !