emonDC Development

Happy new year everyone,

I’ve been developing a board for DC sensing.

Below is an old post from 2 years ago. I’ll update this first post when I have the new board. Cheers.




The particulars to this board are numerous! DC sensing, as I’ve learnt, is a complex subject with many safety related design elements.
With help from the Megni team and the community here I’ve developed a board with the following features, originally based on the emonTx v3.4 although now very different:

  • DC sensing based on shunt monitor LMP8481, using external shunt up to 500A or internal shunt soldered across bridge on board. (LMP8603 was found as a better replacement.)
  • Bidirectional or unidirectional sensing with 2 shunt monitors for battery / generator application.
  • Make use of 12-bit ADCs, make an accurate implementation.
  • ESP-12S wifi module usage. STM32 option for lower power
  • MicroSD card slot for storing data locally. Web-interface for accessing SD card
  • Suitable for up to 60V DC systems. Onboard switching power supply based on TL2575-05, powered from <60V DC supply. (TL2575 never used)
  • Connection options for onboard and external shunts.
  • Accurate voltage sensing

Other thoughts…

  • Overvoltage protection. Done
  • RFM69 radio. Implemented in forthcoming stm32 revision.

This is a step in the direction of a charge controller for solar and wind.

I’ve left the RFM69 chip off the board due to the new wifi module being used. Please try change my mind on that if you think the RFM chip would be useful.

All here on github. GitHub - danbates2/emonDC: DC Monitoring, internet connected datalogger compatible with OpenEnergyMonitor.org project.

1 Like

Judging by questions asked here, I think the ability to offer either (or both, but both would likely be superfluous) would be an advantage. In a number of instances, correspondents have mentioned that the location of their monitor node doesn’t have Wi-Fi coverage. I don’t know the numbers, it is probably not that many anyway, but it is something to consider.

There is a way to get RF on this board so I’ll include it then.

I’m also going to put together a guide on using the spare analog input on the Tx for DC sensing, using either a small board with a shunt monitor on it, or a hall effect IC.

back from a shindig and felt inspired for some late night electronics design as you do, so a quick sketch of a tiny ESP DC monitor.

contains:
ESP wifi chip
microSD card slot
voltage regulator
shunt monitor
shunt

to do list:

replace shunt monitor with suitable one for 56V max
check voltage regulator for suitable power output and voltage range.
find 2 channel ADC suitable for measuring both voltage and current, at the moment it’s only current.

the idea is to get as compact as possible for BATTERY ONLY applications.

thought, suggestions welcome

Hi @danbates

Are you any further along with this project?

Thanks
Paul

Hi @Pukka Paul,

Close to production… I’ve got a functioning prototype which I’ve put together in the Megni lab, I’ve got more to do on the software side of things, added functionality for the ESP / SD card should be soon, what I want is to be able to download a txt file of the logged data, getting there…

See Small DC power datalogger for 48V ebikes
for a compact version of something similar I’m developing in parallel.

Bests,
Dan

Hi. Did you get any further with this project? I am off grid running a 48v nominal (max 65v) setup with solar, wind and hydro components. I’m looking for a remote/network/internet accessible DC monitoring solution for the wind and hydro components of my system as I am having to combine the outputs of both into one shunt/monitor in my Outback setup. Can you help me please? Thanks. Nathan

Hi Nathan,
The PCBs are arriving in 2 weeks. An order of 50.
My SMT assembly guys will then do their thing, so an added week, then packaging.
A safe estimate of when they’ll be ready for shipping will be early April.
The final boards should have WiFi, 433MHz RF optional, MicroSD card datalogging and battery powered RTC.

Great! :grinning:

Just a little note to say I’ve put out for prototyping what are hopefully the final revisions of the two emonDC boards… Will put another update here in about 2 weeks when they’ve arrived and been soldered.

2 Likes

I’ve been making decisions over the power supply. I need to put my thoughts out, part sharing, part sanity check.

I’m going modular.

DC Buck Converter eagle file attached. My thinking is that in certain contexts 5V will be available anyway.
Archive.zip (89.3 KB)

I’ve got emonESP running alongside the 12-bit ADC MCP3202.
When I post to emonCMS there’s a significant drop in the number of samples. In the serial output below it drops to 3640, then back up to a more typical 5440.
Are delays in the code on the emoncms and http side of things causing delay() milliseconds causing potential sample opportunities to be lost I wonder?
I think the question is… is emonESP asynchronous?

String to post:CA:50.17,VA:45.03,PA:2259.01

/input/post.json?json={CA:50.17,VA:45.03,PA:2259.01,psent:55,psuccess:55,freeram:16928}&node=espadctest1&apikey=eee8fc8218573e61453a6bf940377b7a
Plain old HTTP
ok
Samples - 3640
channel0avgA - 2055
done_currentAcalibrated - 50.17
channel1avgA - 2055
done_voltageAcalibrated - 45.03
powerAcalc - 2259.01
Whaccumulator - 370.86
numberofintervals - 591

Samples - 5440
channel0avgA - 2055
done_currentAcalibrated - 50.17
channel1avgA - 2055
done_voltageAcalibrated - 45.03
powerAcalc - 2259.01
Whaccumulator - 371.49
numberofintervals - 592

a modified version of emonESP attached.
src.zip (82.7 KB)

I’ve visited Dennis Ferranti where these PCBs get manufactured.
Some day now…
emonDCduo-ESP v1.1.zip (149.5 KB)

1 Like

Well well, look what arrived in the post this week, about time too.


1 Like

Very interesting! Is it “Designed in North Wales UK by Megni” ? and “Designed in North Wales. Made in the UK”? They kept that quiet.

[edit - I see from your other thread the PCB’s are made in Wales (See emonDC Development). Would this post not be better off placed there?

1 Like

True, can you shift it?

done!

1 Like

Cool thanks.
I did most of the designing, I think my initials are on there somewhere. I’ve had enough help along the way seemed a good idea to put Megni on there. Perhaps that’s the nature of most design projects though and I should’ve put my name on it.
A fair proportion will be made in the UK, not as much as I originally wanted. Over the last year I’ve had the PCBs for prototypes made in the UK. I then went to a manufacture ‘Dennis Ferranti’ or DFM in Bangor, North Wales, near my home, near the Megni offices, the same place where the TX gets assembled.
After getting all the PCBs made so far in the UK I assumed my UK based manufactures DFM were doing the same, and I’d assumed they were aware of the reasons I were doing it local, turns out they’ve had them made in China! which I was a bit disappointed about as I thought it was clear I was trying to do everything as locally as possible, in retrospect I could’ve been more clear on this. If another batch is made I’m getting them done in the UK. Although there’s nothing wrong with Chinese boards or the principal of international trade as we’ve all got to work together on one planet, I’d just feel better about UK or Europe made boards for political and personal reasons.
My guys at DFM in North Wales will be doing the SMT assembly, and possibly the hand soldering pending quotes.

OK. I’m going to get my name on them, instead of ‘designed by Megni’.
I want this to be my responsibility and after all, I’ve design it :slight_smile:

2 Likes

emonDC has been designed by @danbates with support from @TrystanLea and myself. The design is entirely Dan’s. We assisted with some advice and providing equipment. Dan has done all the hard work! See Dan’s post above regarding the silkscreen update for future versions.

DFM assembles most of our other boards for OpenEnergyMonitor. They are located 10min from our office here in Bangor, North Wales! I specify UK PCB manufacture for all our boards. There isn’t a huge difference in price for volume production. I much prefer to do as much manufacturing in the UK as possible. It’s really great being able to pop into the factory to help smooth production issues. I usually pop in for a morning every couple of weeks.