OEM Shop products and direction

Organisation
1

Appreciate I’ve not been paying attention for a while, but I’d like to put it out there that the shop looks like it’s silently dropped products…

Here’s some examples:

EmonEVSE - seems to have totally disappeared
M-BUS Hat - likewise (the usb alternative doesn’t meet the need quite as well… remote HP monitor with a Pi Zero)
Relay Boards (3ch + 1ch) - also disappeared

Various other old products also seem to have silently disappeared (emonDC for example).

Why would you not continue to stock some of these things ? Even if they don’t sell super well, being able to show the breadth of your ecosystem is impressive.

BTW it is very very annoying that product pages disappear from the shop entirely leaving nothing but a 404 error. This makes finding information on older products very very difficult indeed as the shop had links to all the relevant information. I tried finding information on the MBUS hat the other day and it took 15 minutes of searching before I found the GitHub repo.
Please can you consider leaving old product pages in the shop and just mark them as out-of-stock or discontinued ?

I think it would be also useful to do more BLOG posts, particularly on OEM direction … often things are found deep in unrelated foum topics… e.g. emonpi3 developments

Speaking of direction…

A few things jump out at me from reent catching up in the forums

  1. The firmware nightmare. I’m lucky I understand it but this must baffle the casual user. I get why you changed to LPL radio format, but the way it was done was bad. You went through ashort period of RFM69 native and huge fanfare of CM monitoring using it, including a detailed analysis of which radio format to use only to change a few short months later to LPL. THat’s bad enough but the crazy selection of firmware choices for the pi2/tx5 is bewildering, each with different choices to be made, some having some features, others not (e.g. temp monitoring). There has to be a better way.

  2. Changes again to another microprocessor… atmega 328, AVRDB, STM32 (development effort at least) , and now another one in emonpi3… The AVRDB was touted by Trystan to be everything they’d ever need. This is going to create huge support nightmare for OEM with all these different versions in the wild. Maybe a rethink of the architecture is required as opposed to trying to get the microporcessor to do so much (e.g discrete AD convertors verses all in one chip … iotawatt style design?)

  3. I really don’t get the use of RJ45 socket for power (emonVS)… especially on a device that can have a RJ45 network socket. This isn’t user friendly and prone to things being connected wrongly. Also, moving away from RJ45 for the 1wire/pulse counters to terminal blocks makes the emonpi2 look , well, DIYish… The price of emonVS is also eye watering.

  4. On emonpi2 design, another bugbear is the small screen. This is much harder to read than the 2 line LCD on the orginal emonpi

  5. Staying with emonpi2 design, the knockout panel, while practical cheapens the design. When a PI is fitted, the end panel doesn’t surround it neatly, leaving gaps between top of network socket. It looks much worse than the very smart emonpi endcaps which are tightly wrapped around all connectors. While I appreciate this may be practical, it takes away from device aesthetics

  6. Abandoning emontx4 expansion… I saw the thread on this, but to totally abandon this option that’s already been designed and sold isn’t fair to existing tx4 owners. Forcing them down pi2/tx5 route when all they needed was the expansion board doesn’t sit well (I did read the threads on this).

Sorry if this seems like a rant, but I really am struggling to understand OEM direction, although the focus on heat pump monitoring is groundbreaking and excellent I must say, but it does feel like it’s at the expense of other products.

This is just my 2c can I don’t want it to sound negative. It’s one mans perspective. And I wouldn’t be without OEM and the ecosystem, but I can’t help thinking if I was new to it today and not entrenched I would have went with OEM… it was the complete ecosystem of products that I slowly bought into over time.

Its worth saying the decision to keep it all open source should be applauded and has created a fantastic community (& great software and products)… the ability to hack on it, change things and contribute certainly cant be under estimated… as well as this forum to give feedback :slight_smile:

Love to hear your thoughts…

1 Like
2

Hello @alandpearson

We are a small team, at the moment I have my time split between heat pump monitor development and the emonTx4/5 and emonPi2 hardware development. A keen eye will see how development on one drops away as I focus on another, or my ability to keep on top of forum threads :slight_smile: I would love to engage more on heat pump monitoring threads. Glyn is pretty much flat out building heat pump monitor orders and running the admin side of the shop alongside Gwil. We’ve also all got young kids which certainly adds another dimension :sweat_smile:

Zooming out, looking at our product history, between 2012 and 2022 our electricity monitoring hardware mostly ran on the same platform ATmega328, YHDC CT sensors and AC-AC adapter voltage sensors, the circuitry was mostly the same through the emonTx2, emonPi1 and emonTx3. While in practice the accuracy of this basic monitor was quite good, the maximum possible error due to the tolerance of the sensors was quite high! and so we really wanted to pursue a significant upgrade in the hardware capability.

The emonTx4 together with the lower tolerance CT sensors and emonVs voltage sensor provided a significant improvement in that maximum possible error due to circuit component tolerances, dropping it down to ±1.2% Overview — OpenEnergyMonitor 0.0.1 documentation. The higher ADC resolution also helps. The increase to 6 and 12 CT sensor inputs and full 3 phase voltage support is also a huge step forward alongside making that same capability available across both the emonTx4/5 variants and the emonPi2.

We’ve certainly made a lot of changes in a relatively short space of time and some decisions e.g not realising that the RJ11 would not be able to carry the power requirements of a full RaspberryPi and therefore needing to switch to the RJ45 was unfortunate. This said a few mistakes are to be expected and we would honestly never ship anything if we tried to get every single detail perfect. This is partly the result of trying to do a lot with a small team and budget. We could choose to do a lot less but then we enjoy having a varied range of interests, it’s what makes working on this project worthwhile for us and keeps the motivation going 15 years down the road now from where we started :slight_smile:

I will leave this there for now. Hope that gives a bit of a clearer picture of where we are coming from. If you or anyone else are interested in specific expansion boards for the emonTx4 please let me know. I’m currently working on reducing the number of firmware variants, more to come on this in the next few days.

1 Like
3

Thank you so much for taking the time to reply.

That’s quite some info on the accuracy improvements.

Appreciate all your points, and how hard it is to run a business that is in manufacturing, software development, installation & advice in the UK. It’s quite an achievement and especially 17 years still going strong.

Please don’t interpret my post the wrong way, I’m a huge fan of OEM, and I don’t think there is anyone doing anything like it, certainly not with the passion, skills or knowledge to glue it all together.

Keep up the good work, just wanted to provide some feedback and as with the other members of the community, I’ll do all I can to support you.

1 Like
4

Thanks @alandpearson

I will chat to Glyn and Gwil about what we can do to keep pages alive for products that are no longer stocked on the shop. I agree that links going to 404 are a pain. The product listings are all there on the backend and so making the listings available again even if not stocked could be possible.

Realised I got the numbers of years we’ve been going wrong in my post above, it’s 15 years :slight_smile:

2 Likes
5

Hi Alan,

Thanks for your honest feedback.

The past few years have been tough for hardware manufacturing, we’ve had to deal with severe component shortages, which have heavily influenced what we can and cannot design and manufacture.

Keeping older products stocked in the shop and continuing manufacturing batches for older hardware is a significant a drain on resources. As a small team, we need to focus our resources on our current products.

The EmonEVSE has been discontinued, we still have some OpenEVSE kits in stock but going forward its best if customers order directly from the OpenEVSE store. Most customers were choosing the OpenEVSE kit rather than the pre-assembled EmonEVSE. Continuing regulatory certification for the EmonEVSE was becoming a burden, and we recently came to the end of a manufacturing run and due to the high setup costs and high MOQ for injection moulding the enclosure, it doesn’t make sense to do another run with low sales volume.

We have a number of these in stock, I’ve updated the stock count:

This has been discontinued a number of years ago, it never sold very well. We can’t compete with other low cost WiFi relays which are on the market.

This is a good point. We’ve now created an ‘Archived’ shop category , we’ll try and ensure older discontinued products are moved to this category rather than fully removed.

2 Likes
6

Heh, I must have been one of the last people to buy a EmonEVSE.

1 Like
7

Thank you both for the responses.
From an outsider point of view, it seems you are victims of your own success.
Thanks for creating the archived product category it will be very useful… hope you add all the old products there :slight_smile:

I’d like to make one final point on the older products. You are discontinuing things with no replacement… emonEVSE is a prime example. You point to openEVSE, but a lot of people want a prebuilt, working device. When emonEVSE was launched it was the only 22kW 3phase charger on the market that could do solar divert. It was also (and still is) the only one that would fit with my existent emon* ecosystem.

And this is the point. You have created a great ecosystem of products that all work with each other, are customisable and create stickiness to your business. To choose only to focus on a few new products I think will hurt you in the long run and having a smaller selection of things you can sell ultimately less profitable. What’s the point of an emonpi if there is nothing else it can talk to ? Sure it can monitor electricity consumption, but so can a smart meter. Yes its got 12 channels… I can imagine the market being small for a stand alone device…

But if you have an ecosystem that can charge my car (emonpi - using solar divert), monitor my heat pump (emontx + a self built emonHP system & m-bus hat (before you started offering the kit) ), monitor my room temperatures (emonTHs), control my underfloor heating (3 ch relay + thermostat) and display useful data (emonGLCD + emoncms dashboards) then you are onto something.
When I look at the shop and see only a small handful of products it doesn’t seem you are taking advantage of what you have built over the years.

Of course it costs money to manufacture and stock these things, and the more products you have the more people you need to do it, and cost increases. I can’t imagine how long it takes to assemble an emonEVSE or put together an emonHP bundle. Much as I’ll get flack for saying this, when you’ve created a product, done some initial production runs in Wales, would it not be cost effective to get manufacturing done abroad ? Then you guys could concentrate on the fun stuff, new products, growing the HP monitoring side of your business etc.

I don’t envy your position and don’t have the answers… I just find it surprising good, successful, useful products that make up the emon* ecosystem no longer available and that’s a shame. Not that I personally would purchase them (I have already !), but I would hope others would. My neighbours look with envy at my emonEVSE as I slurp the free godjuice from PV and the same with hp monitoring (a group of ten houses built together all have heatpumps and around 6 EV owners). To be clear I’m not talking about products that evolve like emonpi/tx etc.

Again it’s just my 2p and thank you listening - that alone makes OEM a very special organisation, hugely appreciated by customers.

2 Likes
8

I wish we had more time and could execute on all fronts :grin: We are actually thinking of hiring to help with the heatpumpmonitor side of things.

I would say in terms of the emonTx4, emonTx5 and emonPi2, that emonTx + emonPi combo that others have asked for will be back available in full very soon. I really like the new emonTx5 design, the expansion board to the full 12 CT inputs is also properly mechanically secured this time and the new set of common firmwares is more consistent across the 3 variants.

image
image1503×849 121 KB

2 Likes
9

Sorry I missed this. That is surprising (low sales) and it’s not surprising as well. Back in the day most EV sales were giving a free(!) charger & install, which I guess left the market small for 3rd party chargers. Maybe emonEVSE was niche… which is such a shame… The high cost of manufacture and low sales are certainly not good combinations :frowning:

[edit] I also didn’t realise openEVSE store offer fully assembled units. It’s a shame you couldn’t resell them and make a small profit… goes to the ecosystem point above (new users would be able to see the ecosystem vs existing folks who know).

10

Perhaps we could highlight the compatibility of our system with OpenEVSE and direct people to the OpenEVSE shop? In a way showing that we are part of a network of compatible open source projects is probably as good as actually stocking these ourselves?

3 Likes