Has been made redundant for GivEnergy....

Has been made redundant for GivEnergy....

Ok looks like they are going down...
EVM starts talking about givenergy at 3:20 into the video, interesting to listen to what he has to say about the company culture, the support staff, the kit, the leadership, some of the decisions being made and challenges the company has
Definitely worth a watch.
In 5 hours, reached 12,000 views and 2,000 likes. Surprised how many people are following this.
Watched it yesterday. He says he's not bitter but I think he is & rightly so IMO.
It certainly won't help GE in terms of new sales. I'm surprised GE didn't put a gagging clause in his redundancy terms . A similar thing happened to me some years ago & I was prevented from talking about my departure from a company for 2 years.
I hope 100% YT works out well for him.
I agree which his statement on management who don't know how to provide support. It is rife in the NHS and a reason why I retired early.
geoffreycoan Too many adverts - every 60-90 seconds. Life's too short for that crap.
I got one ad in the whole thing. Must admit I didn’t really understand what his job was. Still wouldn’t wish redundancy on anyone (with their exception of some politicians…) having done it 3 times.
SimonClarke I watched part of it yesterday and got a couple but today is just ridiculous.
Watched it last night, he tried to be very diplomatic about GE's future... but the message I took away wasn't a good one. Circling the drain seems the best analogy.
I didn't get any ads.
Use a VPN set it to Albania.
I think some serious consideration needs to be given by those who have good knowledge of the kit, dev of GivTCP and anyone else who can help out in planning for the fall of GE and providing some kind of self support as a community.
Pretty sure there will be a few helping out

I think most things are already covered for local control. The EV charger is going to be a pain as its a bit hit and miss on local with timeouts but atleast they can be replaced a bit cheaper than a PV system.
I've made a start back on the wiki again for storing firmwares & changelogs.
Vestas I think I only had one advert through the whole video
Maybe the increased adverts is because the video is proving popular and google is stuffing more videos in as a result? EVM definitely needs the adsense revenue
I didn’t fully understand his role at givenergy. He made videos for consumers (how to reset your wifi), and for installers, and he said it was 3 days a week and he’d been working for Giv for a couple of years.
Must be a limit to the number of relevant videos you can make so maybe he did more than that that we’re not aware of.
Getting fired by teams call when its not even a covid excuse is pretty appalling. Especially firing someone who has a nearly 100k youtube subscribers, “management” must have known he would be making comment about it.
Very sad.
I did get any adverts, but I do use 'Brave' as my browser for youtube
Vestas Adverts are inserted by Youtube, not the creator.
I have not seen any adverts in YT for the last 10 years - I have adblock for that crap.
CGoode Gagging clauses aren't as enforceable as most HR departments would like to believe. They are only really binding if they can be shown to be mutually acceptable - typically they aren't, irrespective of what was signed. But it's a small world and usually better to resist disparaging a former employer irrespective of how bad they were. Tempting as it might be! Similar goes for trying to stop you working for a competitor.
hoggy I'm not sure calibration is covered at this time. I think the BBC Basic Inverter has a "hidden" control for it but the app doesn't play well these days, not on later versions of Android anyway.
Also I don't think we have a "reset all registers to default" function on anything local. I know I've had to use the portal "reset to defaults" funtion a couple of times due to register corruption.
Both of those are going to become issues when GE goes tits up (which I'm now convinced of as I wasn't previously aware of the £8 million "loan" from GE Property Ltd to GE Ltd in December 2025).
sponev Gagging clauses aren't as enforceable as most HR departments would like to believe.
Especially since said "HR dept" will be getting the chop along with everyone else. Also of course a contract is no longer valid when one of the parties ceases to exist.
Ok I missed this thread worth a read...
https://www.speakev.com/threads/who-on-earth-are-powerdog.194673/page-2#post-3885182
PowerDog is the new GivEnergy...

Vestas the Rubicube app has the ability to stop calibrations I think. It is just literally the same register for that, you just send it 1 instead of 0 to start it (or 3 to do a top calibration) so I imagine he's already on with that.
Reset to defaults doesn't really exist (well it does, but it erases the whole inverter back to factory - which isn't good as it removes Grid codes, export settings, literally everything - forcing you to recommission it from scratch. So when you click reset to defaults in the portal I beleive it is actually some sort of list of things it resets one by one rather than a dedicated inverter function- which explains why it doesn't always sort everything. I imagine we can come up with a similar list.
M_J yes, this has been known/suspected for a while and you'll see one of the regulars here posting in some of those threads. The anger expressed there is understandable but threats of class action etc might just make things worse - anyone trying to re-establish under a different guise might just chuck the towel in. It's a clusterf@ck whichever way you look at it.
hoggy do you think this could be added from the RubikCube app (assuming willingness to do it of course) or should I be looking to set up HA this weekend if a deeper dive is needed? I was hoping to avoid that to be honest as I've got lots of other stuff on at the moment - but needs must.
I've posted on the RubikCube thread to seek an opinion but I don't want the guy to feel under pressure. We're very fortunate to have that app as similar doesn't exist for other systems as far as I can tell. I've been shopping around for alternatives to GE but local third-party party control doesn't seem to be a strong feature for any of it.
hoggy another approach would be to list those entries which have non-default values, allowing the user to inspect anything unexpected. That way, they can reset just the ones that are wrong, rather than resetting everything then having to set up timers again.
(Or you can just dump all regs to a text file, and the diff against an earlier dump to see what's changed.)
sponev Its understandable because if anyone other than GE uses the data/has control of the portal then they're breaking several laws.
Osler has ZERO right to the data/control being transferred without individual customer consent - and that includes receiving the data in the first place.
So if GE goes bust then anyone receiving data from GE equipment is immediately in breach of enough laws to end up with a 6 figure fine. I certainly intend to report them to the ICO immediately after they go bust & press involvement is almost inevitable so I wouldn't put much hope in Powerdog surviving long.
Vestas if his plan is simply to try to force it through and perhaps introduce some form of a subscription model, then that's out of order. Problem is a lot of folk would have no other option. In that case, he'd deserve whatever he got.
sponev I don't think he's the sort of person who thinks laws apply to him.

sponev I'd imagine he already has this for his own so i don't doubt he'd be able to quickly pivot if not already.
@DD yes I have similar thoughts on resetting - get a few people to drop registers and report back what "normal" looks like.
I've implemented a diagnostics thing for my own use that highlights anything that looks wonky for all the most common reasons (why Givenergy didn't do this I've no idea, most of the time it's either paused or the power level is set low or it's stuck in permanent charge window from the quick actions.)

Note this is not intended to replace all these apps people have made (this just sits on a Raspberry Pi Zero running in the background with a pretty interface) as theirs will be more polished but if there is merit it can be pushed out I guess. I need to think about what to do with engineering mode as you can cause some havok with that if you don't know what your doing.
Vestas from what we've seen so far and the bizzare commentary in the annual report, you may well be right. I'm not sure how this ends well.
hoggy Some form of engineering mode will be required as those registers which aren't exposed via GivTCP are the very ones which require a "reset to defaults".
I've had two occasions (pre-fast f/w which has CRC on comms) where "engineering" registers were the ones corrupted. On both occasions the inverter was completely non-functional - no power from/to batteries or PV strings. Without a "reset to defaults" the inverter was to all intents and purposes functionally dead.
hoggy Do you know if triggering the 'start calibration' command is equivalent to scheduling a calibration via the portal?
When I triggered a calibration manually via the BBC Basic app, it just kept running and drained both of my batteries to an unhealthily low voltage, then just kept trickle-discharging for 16+ hours until I manually aborted the calibration and triggered a 'top end calibration' afterwards.
This leads me to think that possibly the portal calibration method may be a 'managed' sequence, which starts/stops a discharge, then starts/stops a top-end, based on the parameters of your system (stated battery capacity in Ah etc)... but I could be talking BS...
hoggy This looks perfect for local control; I for one would gladly pay for a ready-made solution like this, with all available GivTCP commands exposed via a nice GUI.
Vestas Hmm, interesting. How many of these 'hidden, engineering-mode' registers are there? Have they been documented anywhere public? How can one invoke/modify them via local control?

hoggy I've made a start back on the wiki again for storing firmwares & changelogs.
Please can you remind/tell me the wiki address?

ectoplasmosis It's not managed by default from the portal (the portal does catch on that your doing one and you get the app notifications) however I've triggered these locally lots (well, in the grand scheme of things, the system is over 5 years old now) and it just walks through the sequence without hand holding step 1 to 7.
I only have 1 battery though so i imagine if you have multiples and one is saying something different to the other it sits and waits for something specific. That or maybe the BBC app just keeps sending start cal over and over (maybe the idea was to trigger it then close the app while it did it's thing)
Ultimately how often do you run calibrations in reality that if it did get stuck, you just cancel it and trigger it again/top calibration - would it be the end of the world? - probably not.
Although the offer is appreciated I think the cloud/portal is the bit with the most chance of being bought. I imagine Axle or Octopus or whatever would be very keen on getting control of such an asset even if the rest of it went. Captive audience and all that. Even some research companies would be very eager to have LIVE data of PV output from across the entire UK (I know I would for TerraVolt, think of the forecasts accuracy you could have with that rather than using models and best guesses)

PianSom
https://wiki.terravolt.co.uk - note that it's still pretty much as it got left (i.e. mostly empty) - I have put all the firmware stuff on but left as pending. Didn't want to tempt fate.
hoggy Do you happen to know what domain address the dongle is looking for when it connects to the "portal"?
Probably only does a lookup on boot as all I'm seeing with a packet sniffer is a single intermittent TCP connection to an AWS EC2 instance and some occasional ARP/DHCP chatter with the router/switches.
If you don't then I'll reboot it at some point and check.
Vestas are we advocating disconnecting and go remote only control if Giv goes belly up but the portal remains active temporarily? Presumably there's no risk of a disgruntled bad actor bricking the job as a final hurrah? It would be difficult to prove? Is it even the same connection?
In a former life I had to dismiss a fairly senior ops guy, who, after the event, remotely accessed an industrial control system with malicious intent. His access should have been revoked but wasn't. Messy. Hence my (probably unnecessary) paranoia.

Yes I know the domain it's speaking to but equally so does this person so I never bothered to spend much more time on it:
https://butnoidea.co.uk/2024/03/05/givenergy-inverter-networking/
The bit about "How does this work for the Inverter?"
I did start building a rudimentary proxy a while back to intercept it's comms and be able to deliver some commands I have a hunch on, directly to the dongle - I never got around to testing it fully - but I had my reasons at the time. It is on a never ending list of interesting but probably not that useful tricks to pull off.
hoggy I'm just making a note of it as that'll be the domain which gets redirected to another host. Obvious really and none of the ones I tried on spec 😃
sponev The "portal" is simply the IP address of the domain the dongle looks up at boot.
That IP address can be changed to point at whatever server/portal you like if you have control of the zonefile - so for example it could be portalDOTpowerdogDOTcloud - and that could be the same server instance its on now, in fact it could already be there 😉
That in turn means everyone's data (and control - potentially) for that domain gets sent to wherever it points.
That was OK if it was GE - but if they go bust then its not. The DRS/DFS/VPP market is the wild west now. Pretty much as predicted a couple of years ago we're heading for another financial scandal/taxpayer pays the bills scenario....
I intend pulling the plug (in a networking way by returning NXDOMAIN on lookups) once they go bust.
Vestas thanks for that and time I got a bit closer to how the comms is actually working. I thought I'd seen that last of wireshark and all that malarkey years ago. Bum 🙁
sponev Basically you only have to do two things -
1) Set the inverter/dongle to a static IPv4 address on your LAN;
2) Set a firewall rule prohibiting outgoing traffic for that address on your router.
That'll do much the same.
I have a Mikrotik router so I can force a NXDOMAIN reponse when the dongle boots & looks for the mothership.
That at least is lighter on resources and somewhat more elegant. It may of course cause unexpected side-effects when the dongle not just can't contact the mothership, but is told the mothership is dead 😃
Vestas nice, many thanks. Kinda feels inevitable now although I am a renowned pessimist 🙂
Vestas The "portal" is simply the IP address of the domain the dongle looks up at boot.
You would think so. That is how any sane person would do it. It doesn't make sense to do it any other way. But this is GivEnergy, and doing things in a non-standard way is the GivEnergy standard. For example look at how they modified the modbus standard to their own version. Or made a battery BMS communicate in a non-standard way.
Do your own packet capture and see for yourself how it is done. Don't rely on other peoples' guesswork and assumptions. Reverse engineer it yourself and say OMG.
Rubikcube I shall have a look at what hits the router this weekend.
Folks, you are worrying unnecessarily. I asked Google about givenergy’s administration and its AI told me that it wasn’t happening and they had strong company results:

So there you go, must be true 🤣
geoffreycoan That's confirmed it then!
sponev I agree, HR dept exit clauses aren't worth much or really enforceable.
However, my gagging order was delivered by my employers lawyers via a Court in London which I had to attend in person & swear I would obey!! All documented & officially noted. No contact or discussion with the press, social media (such as it was at the time), other ex-employees & anything else they could think of that could cast a shadow over them as a company.
There were a bunch of us all lined up in the dock all getting the same treatment from a barrister in a wig.
Pretty heavy stuff. I guess our previous employer didn't take kindly to the fact that a bunch of ex-employees decided to take legal action against them for their ethics, business practice & treatment of employees.
It was a battle of the lawyers & the only winners out of the whole episode were the lawyers (as always). Ours was a group action with very limited funds. Their lawyers were London based high court wigs who must have cost them a pretty penny which gave us some degree of satisfaction that they had to pay through the nose to shut us up. It's a long time ago now but it still grinds my gears (can you tell?) & that justice comes down to who has the deepest pockets.
Deep breath, relax & move on........
CGoode Deep breath, relax & move on........
At least you had a good, so kudos and most would have done nothing.
CGoode I haven't quite been in that situation but credit to you for standing your ground. However I have been scapegoated in business having turned round a failing company nearly twenty years ago. It was subsequently taken over and the new owners destroyed all the carefully crafted systems and procedures we'd put in place. Chaos ensued including a failed ERP implementation - I was the scapegoat for that. Fortunately I'd documented in detail why it wouldn't work and so subsequently "relieved" them of oodles of cash after the event.
I was upset by it at the time, particularly after a successful career, but with hindsight it was the best thing that could have happened. Subsequently I've run two of my own business alongside my wife which has set us and the family up very nicely as early retirement approaches. I also had the pleasure of watching that company demise and fold around 3 years after I left. So there's always hope after a tricky period in your career
Having been through a company failure as a result of incompetent owners, I wonder if that's why the current GE saga resonates so much. Seems like a lot of good folk have been f@cked over unfortunately.
Just an update here as its more appropriate than the community thread.
Had a play around and its easy enough to stop the inverter chatting to people you don't want.
However the dongle will reconnect to wifi every ten minutes if it can't contact the comms servers*. This is obviously going to be an issue.
I haven't tried yet but based on that I wouldn't be surprised to see the entire system reboot every 24 hours.
Also the system looks for a NTP server in IP space allocated to a specific customer & isn't in AWS IP space. That suggests its pretty old?
*this also explains some dongle reconnections I was at a loss to understand over the years
Vestas How does the dongle behave if it’s set to wired Ethernet only, via the DIP switches? Does it still try to ‘reconnect’ somehow if it can’t phone home? What exactly happens to the dongle when it ‘reconnects’?
ectoplasmosis The wireless dongle disconnects from the WLAN (SSID) and reconnects. As part of that it does a DHCP discovery and has an IP address assigned. As the hardware is ancient it takes 3-5 seconds to complete this "disconnect/reconnect" process.
No idea how a wired connection behaves but I think the dongle/system is simply trying to force a DHCP renewal in the hope of regaining connectivity with the comms servers. If it behaves the same on a wired connection then there should be very little loss of connectivity (well under a second) so it'll be less of an issue. NB - that's a guess, you'd have to try it yourself.
The £8.6M inter-company loan last year raised my eyebrows; alongside comments from EVM this is pretty worrying. I've got a 2nd AIO on order and have asked my installer to contact GE. Not good.
Rubikcube You would think so. That is how any sane person would do it. It doesn't make sense to do it any other way.
Have you considered there's a SIM version of the dongle?
Can't find a route via the default gateway (wifi) then proxy ARP via the SIM (or whatever). Was common enough 15 years+ ago.
Vestas The wireless dongle disconnects from the WLAN (SSID) and reconnects. As part of that it does a DHCP discovery and has an IP address assigned. As the hardware is ancient it takes 3-5 seconds to complete this "disconnect/reconnect" process.
Could try a static ip address ? On the web interface for the dongle, on the 'STA Interface Settings', there's a dropdown to let you choose DHCP or static IP address.
DD Afraid not. I'm not setting a static IP address on-device, that's just an accident waiting to happen. The inverter has a static IP address on the router.
Vestas Afraid not. I'm not setting a static IP address on-device, that's just an accident waiting to happen. The inverter has a static IP address on the router.
Not if you've designed your network properly. Keep statics and reservations at one end of the address space and use a DHCP range that does not include that space. There's no harm in adding a DHCP reservation as a way of documenting a static address, but there are still devices out there that don't play well with DHCP.
mrand31 Still not doing it. Been bitten on the backside before with that.
I cannot see any harm in reserving an IP address through the router and setting the same address on the dongle as static.
If it is the only static device on the network and is connected to the router at the time other devices request an address there should be zero chance of collision even without the reservation.
Reserving the address is belt and braces as it will then be removed from the pool available for other devices.
pwdst ....and when you change router years down the line you ARE going to remember this eh? No you're not - especially if you're troubleshooting something else.
Been there, got the T-shirt. Its just bad network management so I don't do it now.
Vestas Been there, got the T-shirt. Its just bad network management so I don't do it now.
I agree. Its best to only have one device determining what IP address should be used, and simplest if that device is the router. Spreading out IP addresses to different devices is a recipe for problems
I always export the list of reservations so I don't need to reconfigure every device that expects to be able to find something else by IP address - maybe that's just me.
I have changed routers at least three times in the last seven or eight years so well practiced by now.
Not bad network design, tidy:-
In a class C LAN or VLAN, create DHCP scope for entire network except network, router and broadcast adresses. Use the first 20/30/50 etc. addresses for static use (servers, printers, network switches, WiFi access points, yada yada yada) and exclude in DHCP scope. Rest of range permitted for general DHCP allocation with no static addresses. Create all reservations in excluded part of scope and never in general part of scope. Don't generally use a LAN or VLAN larger than class C size to minimise broadcast traffic.
Worked alright for 65 sites in the UK and another dozen abroad I administered.
I don't use the router for DHCP, dns or WiFi, I have dedicated equipment for that.