2016-08-04 at 8:30 PM #1694
I’m thinking of trying for a mayfly modbus poller, along the lines of my previous post envirodiy.org/imp-calling-sensor/
with a particle.io wifi/Photon or cellular/Electron3G
Anybody tried interfacing the modbus/RS485 yet.?
For prototyping seems like an external 12V, and a down switcher for Mayfly/5V but also cellular/5V@1A connection, and RS485 board.
Of course a pluggable “shield” would need to do the reverse. Start with a Solar panel 5W/openCircuit 18V, charging a Li battery, and then switchable step up to 5V@1A and switchable 12V.
A field deploy-able innocuous enclosure also needs space for cables and drying desiccant.
2016-08-15 at 4:08 PM #1701
Well I’ve done some design thinking on this. I’ve focused on just generating the +12V. The issue with the Mayfly circuit is that the switched pwr comes from the 3.3V which also powers the processor – and a power hog on the 3.3V_Sw is going to cause a processor power glitch – yes I’ve had them in the past and they are nasty to debug 🙂
So I want to generate +12V, and possibly the easy way to do it is with
SUNKEE LM2577 DC-DC Adjustable Step-up Power Converter – available on Amazon.com (amazing)
The design needed for the boost circuit is to be able to switch it on when needed.
The +5V boost NCP1402 does have an enable on it but it is permanently enabled. It is tied to the 3_3VSw – which is powered by the LDO SPX3819 which is rated for 0.5A.
So not very power efficient, drop input voltage (LiPo 4.2V or +5V) to 3.3V and then boost to +5V – but workable
The NCP1402 is rated for 130mA, so that is not likely to cause a glitch on the 3.3V
Ideally I would
cut trace to C13/L1 and tie power line to C7/D3
Then the NCP1402 U3-En SJ5-3 cut & strapped to Port-D22/(3.3_SW) OR another Port
So I wonder it the layout for the Mayfly is easily available to see how the traces are done.
I’ve ordered the $6 SUNKEE LM2577 DC-DC Adjustable Step-up Power Converter and a Mayfly to try it out.
The other option I considered was to use +12V -start with a SLC battery, and then drop the voltage to +5V for Mayfly.
All the switching is done at 12V/SLA low impedance, with an interesting
$15 CWE Arduino Dual Channel Smart FET Driver Board with Current Sensing 8-28VDC
and a switch converter 12Vto +5V for mayfly.
Theoretically it would have been then possible to add +5V/1A cellular phone later – but this would get expensive for power modules and take more design now.
One of the reasons I considered +12V is I am using an Olimex-STM32-H407 – which takes +12V, for prototyping modbusm.
Personally I think the STM32F4 family is going to provide a lot of versatility for the future –
http://www.stm32duino.com/ for a lot of small boards now and
The key is a rugged mechanical environmental enclosure, solar power, power switching managed by the SMT32F4xx, and low power sleep capability, and using its unique USB Host capability for pluggable USB modems. This can then communicate with the Mayfly over the XbeePro modules. So that’s the longer term plan.
2016-11-29 at 3:13 PM #1813Shannon HicksKeymaster
Neil, these were all good observations about the 5v boost circuitry on the Mayfly. The NCP1402 has been obsoleted by the manufacturer, so I’m looking for alternative ways to generate 5v and possibly other voltages onboard the Mayfly.
2016-11-29 at 9:24 PM #1817
Great. If you want cct comments I’d be happy to participate. I’ve used LM2623 +5V/1A, and also can do +12V
- You must be logged in to reply to this topic.