Comparing solar immersion heater diverters (water heating).

[nowty]

I’m on a general, non-renewable specific Electrician group on fb and the reports of problems from myenergi gear is worrying. Granted the others are much less well known so it maybe a case of you only hear of popular products failing but there is also reports of poor tech support. Make of that what you will.

In what way is the Myenergi gear worrying ?

[Jinx]

[quote=Jinx post_id=22987 time=<a href="tel:1669151222">1669151222</a> user_id=356]
I’m on a general, non-renewable specific Electrician group on fb and the reports of problems from myenergi gear is worrying. Granted the others are much less well known so it maybe a case of you only hear of popular products failing but there is also reports of poor tech support. Make of that what you will.

In what way is the Myenergi gear worrying ?
[/quote]

Seemingly constant reports of failures/bugs and useless tech support. May not be a fair representation as I said but it is noticeable. Seems such a shame as a British, innovative company making a premium grade product, I’d really like to see them succeed.

[nowty]

[quote=Jinx post_id=22987 time=<a href="tel:1669151222">1669151222</a> user_id=356]
I’m on a general, non-renewable specific Electrician group on fb and the reports of problems from myenergi gear is worrying. Granted the others are much less well known so it maybe a case of you only hear of popular products failing but there is also reports of poor tech support. Make of that what you will.

In what way is the Myenergi gear worrying ?

Seemingly constant reports of failures/bugs and useless tech support. May not be a fair representation as I said but it is noticeable. Seems such a shame as a British, innovative company making a premium grade product, I’d really like to see them succeed.
[/quote]

I asked because I have an Immersun which is the same technology that myenergi bought out when it went bust. Mine's been faultless in about 9 years of operation apart from a connector block problem which is well known and easily fixable.

I have used an older Immersun Mk1 and a Solarmiser but both gave quite severe flicker problems to my lights.

[Oldgreybeard]

Everything I've read and experienced suggests that the Myenergi stuff is well-made, well-supported and generally reliable. My first PV diverter was home made, essentially much the same as Robin's Arduino based one on the OEM forum, but using a PIC instead of an Arduino, plus a dedicated fast 14 bit A-D to enable very fast sampling of the current and voltage waveforms, to get an accurate true power and direction. That worked very well, but I wanted to get rid of home made devices, just for future proofing (sooner or later I won't be able to maintain home made stuff).

I then opted to buy an Apollo Gem, as it was the cheapest unit that allowed a remote, radio linked, CT (essential for us, as the meter and outdoor DB is in a kiosk well away from the house). The Apollo Gem was fine, until I installed the battery system. I quickly found that there was no way to adjust the threshold on the Apollo Gem, so that caused the Sofar to take priority all the time, when I wanted the hot water to take priority.

The Eddi and Harvi combination fitted the bill neatly, and was the only other diverter I could find at the time that had the option to have the CT connected via a wireless link. I did call Myenergi regarding setting the Eddi up to take priority over the Sofar and the tech support was brilliant, explained exactly what setup changes I needed to do, and how to fine tune the start up setting. Couldn't fault them, TBH.

Absolutely zero flicker with the Eddi, too, unlike both the Apollo Gem and my home made unit. Brilliant bit of kit in my view, and worth every penny.

[Jinx]

[quote=Jinx post_id=22987 time=<a href="tel:1669151222">1669151222</a> user_id=356]
I’m on a general, non-renewable specific Electrician group on fb and the reports of problems from myenergi gear is worrying. Granted the others are much less well known so it maybe a case of you only hear of popular products failing but there is also reports of poor tech support. Make of that what you will.

In what way is the Myenergi gear worrying ?

Seemingly constant reports of failures/bugs and useless tech support. May not be a fair representation as I said but it is noticeable. Seems such a shame as a British, innovative company making a premium grade product, I’d really like to see them succeed.

I asked because I have an Immersun which is the same technology that myenergi bought out when it went bust. Mine's been faultless in about 9 years of operation apart from a connector block problem which is well known and easily fixable.

I have used an older Immersun Mk1 and a Solarmiser but both gave quite severe flicker problems to my lights.
[/quote]

I just did a quick search on the group and tbf it seems most issues are the chargers rather than solar diverters but again I’d guess that’s the more popular product.
An RCD fault is just met with a ‘return complete unit’, some sparks reporting several units failing in a row so whilst it might not be such an issue for a single customer they are falling out of favour with installers as you can imagine.

-

[sharpener]

In 2011 they weren't widely available so I built my own using an analogue power calculation circuit driving a triac-based heating control "brick". Even with some suppression components this chops the waveform something horrible so I was very suprised the Victron inverter worked with it OK. But it does.

Its CT is on the grid side of everything so battery-charging has priority. The actual immersion heater is amongst the house loads.

[Oldgreybeard]

Its CT is on the grid side of everything so battery-charging has priority. The actual immersion heater is amongst the house loads.

How does that help? A quick, back-of-fag-packet, arithmetic, allowing for a cable velocity factor of 0.66, suggests that if the diverter grid CT was, say, 10 metres closer to the meter than the inverter grid CT, then it would detect an import about 40ns before the inverter. For an export, the miniscule time difference would be reversed, so the inverter would detect the export about 40s before the PV diverter. This is an approximation from the Heaviside equation for a 50Hz signal in a cable (which as as with many others was derived from Maxwell's equations) and was initially concerned with telegraphy delay times.

I do not believe that any bit of kit using a CT could possible respond within 40ns, it's far too short a time for anything but a pretty sophisticated measurement system to detect, let alone measure with any degree of confidence, I think. I would suggest that there is some other factor that makes your system always prioritise hot water, rather than the miniscule time difference caused by the CT position. It's most probably just down to your PV diverter circuitry always responding more quickly than the inverter circuitry, so it always wins the race.

In my case, it seems that most of the time the Sofar would win the race, until I added the second CT to the Eddi, which then allowed the Eddi to win pretty much all the time. My best guess is that the Eddi can compare the grid and battery inverter CT signals far more quickly than the Sofar can respond to the change in direction from just a grid CT.

[sharpener]

Nothing to do with the speed of propagation!

Default state of diverter is OFF.

Inverter does the following in order of precedence:

(i) meets other house loads (immersion is off)
(ii) charges battery
(iii) exports

Because of its location, only at this point does the diverter CT see the export and turn on the immersion heater proportionally, to the extent needed to reduce the export to zero. Due to the finite loop gain it actually needs about 100W export to turn it on fully but I can live with that.

From reading their data sheets over the intervening years I think many other diverters work on more or less the same measurement/control loop principle, even though their waveform control may be more sophisticated than mine.

If it ain't broke don't fix it! Currently waiting for a new battery module, more solar panels and an expert to look at the rainwater system (corroding pipes) all imminent so there is plenty to be getting on with.

[Oldgreybeard]

Nothing to do with the speed of propagation!

Default state of diverter is OFF.

Inverter does the following in order of precedence:

(i) meets other house loads (immersion is off)
(ii) charges battery
(iii) exports

Because of its location, only at this point does the diverter CT see the export and turn on the immersion heater proportionally, to the extent needed to reduce the export to zero. Due to the finite loop gain it actually needs about 100W export to turn it on fully but I can live with that.

From reading their data sheets over the intervening years I think many other diverters work on more or less the same measurement/control loop principle, even though their waveform control may be more sophisticated than mine.

If it ain't broke don't fix it! Currently waiting for a new battery module, more solar panels and an expert to look at the rainwater system (corroding pipes) all imminent so there is plenty to be getting on with.

Sorry, I still don't get it all. I have two CTs at the meter tails, maybe 50mm apart, one is the grid measurement for the PV diverter, the other the grid measurement for the inverter. To all intents and purposes both CTs are going to respond at exactly the same time (certainly within a few picoseconds) and their relative position makes zero difference as to which device takes priority. I could probably put them 50m apart and it still wouldn't make any practical difference.

There are massively larger, and slightly variable time delays in the PV diverter itself and in the inverter, and it's entirely the relationship between those very much longer time delays that determines which device "wins" the race when the import to export transition takes place. Once one device has "won" it will look like a load as far as the other device is concerned, so the potential export will not be registered.

In my case, the most significant issue is that all three of the PV diverters I've used (home made one, an Apollo Gem and now an Eddi, via a Harvi) have had to use a radio linked CT (because the meter and outside DB that feed the garage and charge point is outside in a kiosk) whereas the inverter CT at the meter is directly wired, via around 40m of outdoor Cat 5e cable, to the inverter in the battery outbuilding.

The delay from the radio link I'm sure makes a difference, and may well be the dominant factor when it comes to the Inverter almost always winning the race when the import/export direction changes. My fix was to just add another CT, directly wired to the Eddi (so no radio link) to measure the AC feed to the inverter (possible because the inverter feed comes from the house CU, as does the PV diverter feed).

The only thing that matters in terms of priority is time, how long it takes for each device to detect the import/export transition. Nothing else matters at all, really, although to some extent very small differences between devices can be masked by setting a threshold on one, which gives a tiny time advantage to the other.

[sharpener]

Before I installed the ESS it was very simple, the CT was on the meter tails in the CU, between the point of PV infeed and the grid. So any PV that was surplus to the loads turned the diverter on.

If I had left it like that and put the battery/inverter upstream of the CU i.e. even nearer the grid, then the surplus PV would have heated the water before charging the battery (since the immersion heater would have been driven to consume the surplus before* it got to the inverter).

I see from the above (and this thread https://camelot-forum.co.uk/phpBB3/view ... =11&t=1322) that is what you want to achieve, maybe there is a clue in what I have not done.

But in my case to avoid the immersion diverter taking precedence over the battery I moved its CT to the grid side of the Victron, now the Victron has got to achieve some (small) export to turn the immersion on.

Thassorl

*Before in the spatial sense not the temporal one, maybe that is the source of confusion.

(Somewhere there was a fuller description of my diverter's circuitry but I am blowed if I can find it ATM.)