Photo Voltaic - Immersion heater power diverter - SAFETY WARNING!
- Thread starter Jeremy Harris
- Start date
PhilHornby
Senior Member
They don't really explain "why" you need titanium heaters (and all the info. I can find seems to be hard water/corrosion related). The warranty to which they refer, is presumably the 25 year warranty on the tank itself. 
Buzby
Senior Member
Is there an electro-chemical reaction between Copper and Steel that might lead to excessive corrosion, or lots of fizzy bubbles ?
Maybe Titanium is not so reactive.
See page 6 of this : http://www.free-instruction-manuals.com/pdf/pa_1936615.pdf
Hi,
I don't have an unvented system myself, but it appears that they operate at mains water pressure (usually many Bars, i.e. > 100 feet head of water) so perhaps the Immersion Heaters just need to be "stronger" or more resistant to leaks? But also it seems that most unvented cylinders are Stainless Steel (not copper) and I've definitely seen the recommendation that these should use Titanium or similar heaters, so perhaps there is a corrosion issue here?
Personally, I have (so far) fitted a "PV" Immersion heater (1 kW) which avoids "overloading" the Inverter/Battery (i.e. pulling power in from the Grid) particularly if/when the Microwave Oven / Washing Machine, etc., cut in at the same time. It's also "kinder" on the "Smart Plug" which (temporarily) controls it - a previous Plug died within 3 months of powering my original 3kW Immersion (even though it rarely if ever needed to break the 12A current).
The issue of (Jeremy's) Power Cycling the heater with (say) a period of some seconds is interesting, because repeated expansion/contraction might occur at that frequency. Of course the heater is normally cycled at 100 Hz or at a much lower frequency (minutes/hours) by the thermostat, so there could be an "issue". Perhaps the control program should estimate the approximate overall duty cycle (i.e. 1 / N) and switch individual cycles accordingly (where N is ideally an odd number)? However, I'm concerned that taking individual "half-pulses" might "confuse" the control loop(s) in the PV Inverter, which AFAIK now use high-frequency switching techniques and don't rely on a big/resonant inductor or transformer. It does appear that the myenergi "Diverter" uses a High Frequency Switching Mode (Class D) "Audio Style" 50 Hz Amplifier to "wind down" the amplitude of the 230 volt mains to the heater, in a similar manner to the old "Variac" transformers.
Cheers, Alan.
I don't have an unvented system myself, but it appears that they operate at mains water pressure (usually many Bars, i.e. > 100 feet head of water) so perhaps the Immersion Heaters just need to be "stronger" or more resistant to leaks? But also it seems that most unvented cylinders are Stainless Steel (not copper) and I've definitely seen the recommendation that these should use Titanium or similar heaters, so perhaps there is a corrosion issue here?
Personally, I have (so far) fitted a "PV" Immersion heater (1 kW) which avoids "overloading" the Inverter/Battery (i.e. pulling power in from the Grid) particularly if/when the Microwave Oven / Washing Machine, etc., cut in at the same time. It's also "kinder" on the "Smart Plug" which (temporarily) controls it - a previous Plug died within 3 months of powering my original 3kW Immersion (even though it rarely if ever needed to break the 12A current).
The issue of (Jeremy's) Power Cycling the heater with (say) a period of some seconds is interesting, because repeated expansion/contraction might occur at that frequency. Of course the heater is normally cycled at 100 Hz or at a much lower frequency (minutes/hours) by the thermostat, so there could be an "issue". Perhaps the control program should estimate the approximate overall duty cycle (i.e. 1 / N) and switch individual cycles accordingly (where N is ideally an odd number)? However, I'm concerned that taking individual "half-pulses" might "confuse" the control loop(s) in the PV Inverter, which AFAIK now use high-frequency switching techniques and don't rely on a big/resonant inductor or transformer. It does appear that the myenergi "Diverter" uses a High Frequency Switching Mode (Class D) "Audio Style" 50 Hz Amplifier to "wind down" the amplitude of the 230 volt mains to the heater, in a similar manner to the old "Variac" transformers.
Cheers, Alan.
PhilHornby
Senior Member
The OEM supplied Immersion heaters are Thermowatt 1N1-14, which appear to be 'Incolloy' (never heard of it before today!). Thermowatt's website is particularly useless, but some of their suppliers FAQs are a bit more enlightening. They suggest Titanium for hard water areas...
Which still leaves the minor puzzle, as to why the Tank manufacturer suggests titanium for heavy-duty/commercial/constant usage. I've emailed them; let's see if they're interested in providing after sales support for their £1200 hot water tank... (My oil tank supplier isn't
)I'd contemplated something similar - but by means of adding a large diode! (I know what you're going say about Power Factors etc etc
)Yes, this aspect is what I'm curious about. Maybe the presence of 300litres of luke warm water serves as a massive heat sink
The issue of (Jeremy's) Power Cycling the heater with (say) a period of some seconds is interesting, because repeated expansion/contraction might occur at that frequency.
Yes, this aspect is what I'm curious about. Maybe the presence of 300litres of luke warm water serves as a massive heat sink
I would expect a period of several minutes before you would see a measurable change in the temperature of 300 liters of water with weight of 658lbs (299Kg). That should provide some degree of stability for the temperature near the heating element - obviously the element is much hotter than the water at full power but running at some reduced power level for a longer period of time could have an on-to-off "cycle" of only a few degrees instead of the typical "at water temperature" to "at full heat" temperature differences.
Yes, this aspect is what I'm curious about. Maybe the presence of 300litres of luke warm water serves as a massive heat sink
I would expect a period of several minutes before you would see a measurable change in the temperature of 300 liters of water with weight of 658lbs (299Kg). That should provide some degree of stability for the temperature near the heating element - obviously the element is much hotter than the water at full power but running at some reduced power level for a longer period of time could have an on-to-off "cycle" of only a few degrees instead of the typical "at water temperature" to "at full heat" temperature differences.
PhilHornby
Senior Member
I thought I'd just add another snippet of information.
I couldn't really understand the 'energy bucket' concept; as in, why not just use zero and some S.O.T. upper-value, like 3600?
I'd tried a variety of values and apart from changing the ON/OFF timings, they all appeared to work equally well. In the end I stuck with Jeremy's original values for
As we transitioned into Winter and the amount of Solar power decreased, I noticed something: The ADE7555 was sometimes detecting "Export", when the "Anti-creep" function of the company meter wasn't (yet) active - it lagged some way behind.
The manual for the company meter, says that at some preset minimum current, the Impulse LED comes on and stays on (rather than giving a 20mS flash to indicate energy usage). I'd naively assumed that you got 'free' electricity, when consumption dropped this low
I think the company meter might do the same borrowing trick to even things up. The ADE7555 was adamant current was flowing to the grid, yet the company meter insisted that it wasn't.
The Import/export detection doesn't seem to be something I could influence through mis-calibration (only the amount of import or export).
Either way, not using up all the exported energy in the bucket, seems to give the pair of them the chance to co-exist, when operating at the margins.
I couldn't really understand the 'energy bucket' concept; as in, why not just use zero and some S.O.T. upper-value, like 3600?
I'd tried a variety of values and apart from changing the ON/OFF timings, they all appeared to work equally well. In the end I stuck with Jeremy's original values for
upper and lower (2700J & 900J).As we transitioned into Winter and the amount of Solar power decreased, I noticed something: The ADE7555 was sometimes detecting "Export", when the "Anti-creep" function of the company meter wasn't (yet) active - it lagged some way behind.
The manual for the company meter, says that at some preset minimum current, the Impulse LED comes on and stays on (rather than giving a 20mS flash to indicate energy usage). I'd naively assumed that you got 'free' electricity, when consumption dropped this low
I think the company meter might do the same borrowing trick to even things up. The ADE7555 was adamant current was flowing to the grid, yet the company meter insisted that it wasn't.
The Import/export detection doesn't seem to be something I could influence through mis-calibration (only the amount of import or export).
Either way, not using up all the exported energy in the bucket, seems to give the pair of them the chance to co-exist, when operating at the margins.
Last edited: