Magnetostriction

Understanding the origin of the noises made by a machine is important for knowing the machine is in good health. My Powerspout has always made an unhealthy whine a lot of the time, and at the present time when power generation is at its maximum, the noise is particularly concerning.

Listen to the sound:

Although it has made this noise ever since it was commissioned 8 years ago, only recently have I come to understand what its cause is: it is caused by a phenomenon called magnetostriction. 

Magnetostriction is when minuscule changes in dimension occur to the steel cores of the alternator's stator windings as the magnets of the rotor pass over them. The constantly changing magnetic field causes the cores to hum a bit like tuning forks. 

All voltage transformers on the grid network can be heard to have a low pitched hum for the same reason but their hum is based on the mains frequency of 50 Hz; in the case of the Powerspout's alternator the frequency is of the order of 1000 Hz so the hum sounds like a whine. 

I calculate it to be about 1000 Hz from there being 56 magnets, each passing a stator core 20 times per second: so there are 56 x 20 = 1,120 changes in magnetic field every second for each of the 42 poles of the stator; (turbine rpm is 1200 at the present power level of 920 W; 1200 rpm = 20 rps).

The bearings in the Powerspout have run for over 4 years without change and without added grease, and I am particularly listening out for signs they may be coming to the end of their life. The whine of magnetostriction makes this rather difficult but there is a window of opportunity to hear how the bearings are doing when the turbine is shut down and as the rotor comes to rest. 

I only hope I don't miss the sounds of imminent bearing failure if they appear.

Year end results for 2020-21 water year

Here are the 3 graphs I post each year showing the output of my turbine for the past 12 months; in each, the bold black line is the data for the past year:

1. Daily power output:

There was an early pick-up of generation with rain falling in October, and generation was sustained more or less at maximal generation (917 W) through to March when the weather became much drier; generation then picked up again from mid May, during which month an unprecedented 192 mm of rain fell (usual being 90 mm); thereafter output dropped steadily as autumn dryness set in.

2. Cumulative energy (kWh)

This graph uses the same data as graph 1 but presents it in a different way, this time showing the cumulative number of units of electricity (kWh) generated as the year proceeded; it can be seen the end of year total (5167) is the most units ever generated, just exceeding last year's total of  5132; the FiT payments received for this energy will be £1,338 for generation plus £216 for export, totalling £1,554. This amount more than covers our electricity purchase costs which for the same period will be about £400.

3. Power duration curve 

Again this graph uses the same data but shows it in yet a different way. The graph indicates the number of days the turbine performed at various levels of power output; it can be seen it did not operate at full power for as many days as last year (the pale pink line) but it exceeded last year, and all previous years, in generating for more days at the intermediate output band between 700 W and 200 W. There was not a day in the year when there was an output less than 140 W. 

Conclusion: 

With this being the best of eight years in which my Powerspout has operated, it is a very pleasing result; perhaps the years are getting wetter; certainly the National River Flow Archive has reported that the calendar years 2015 to 2018 were probably abnormally dry years, and the wetter years following are likely to represent nothing more than a return to the long term 'normal'. 

Measuring magnetism

In the last diary entry, I mentioned I use two types of rotor: the standard magnetised Type 2 and the more highly magnetised Type 2+. 

Thoughts which have long intrigued me have been: how much more magnetised is the Type 2+ rotor, - and as a separate question, - are the magnets in the 14 'magnet tiles', 4 magnets per 'tile', - are they each of equal strength.

Underlying these thoughts has been the basic question: is it possible to measure the strength of the magnets.

In this blog I want to show how I devised a way to make some measurements using the force generated on a soft iron cylinder held a constant distance away from each magnet. 

The apparatus uses the technology from a kitchen scales of the digital type, the working principle of which is a strain gauge Wheatstone bridge.

The finished test rig looked like this; the force displayed is in grams:

The detail of the 'binocular cantilever strain gauge' looks like this:

The schematic representation of the arrangement looks like this:

The deformation on applying load looks like this:

The wiring diagram looks like this:

The voltage change across the Wheatstone bridge, created by tension and compression of the four strain gauges, is very small; it needs to be amplified and processed for display in the LCD screen as grams weight.

Conclusion:

What did I learn from this little experiment: in truth not much ! The individual magnets of each kind of rotor all seemed to be equally magnetised; the force exerted by each of the Type 2 magnets was, as seen in the picture, around 424 grams whilst for Type 2 + magnets it was around 200 grams more.

One limitation of the system was found to be that the rotors are not precisely circular; this had the effect of reducing the air gap in one place and widening it in another; since the 'pull' exerted on the soft iron cylinder is very dependent on the distance from the surface of the magnet, this limited the reliability and precision of the experiment.

But it was a fun thing to do, - and I had had the 'guts' of the kitchen scales sitting in a drawer for over 10 years awaiting some useful purpose. It was nice finally to discover why I had been keeping it all those years !

My chart for optimum Powerspout operation.

Powerspout owner / operators who rely on a water source that is seasonal will be familiar with the challenge of needing to optimise the output of the turbine when using different sized nozzles for the different seasons.

In this blog, I'm posting the chart I use to help me with this challenge.

For me, the variables are the type of rotor, the amount of rotor packing, and the type of stator. 

When flows are good and a big nozzle is in use, so much torque is produced by the jet hitting the pelton that the more highly magnetised Type 2 + rotor is needed to keep turbine speed down to a speed where maximum power is produced.

At the other extreme, when flows are least and there is a small nozzle, the torque lost in overcoming the attraction between the rotating magnets and the iron cores of the stator is such a large proportion of the total torque being produced by the pelton that the number of poles in the stator has to be reduced from 42 to 18 in order to keep shaft speed up.

Between these two extremes, the chart gives me a system, which I've built up from experience over 8 years, to guide as to when to change from one type of rotor or stator to the other, and as to how much packing to place beneath the rotor for each of the 13 nozzles I employ.

The chart also has a column for efficiency, and this is 'whole system' or 'water-to-wire' efficiency. Numerous factors contribute to this figure, and these include: how good the penstock is; how good is the alignment of the jet on the splitter ridges of the pelton cups; how well cut are the nozzles to give a good jet profile; how much drag effect there is from grease in the bearing housing; how much transmission loss there is between turbine and inverter, (this will vary with the current); how good the efficiency of the inverter is at different input voltages; and how inefficient the Smart Drive alternator becomes by packing off the rotor. 

Peak efficiency can be seen to be 56.8% and this is achieved with nozzle XII, delivering 2.43 l/s at 1102 rpm, generating 723 watts leaving the inverter.

As I write this, the nozzle in use is number VII and the Powerspout is producing 321 watts; for this time of year this is exceptionally good and is a reflection of the rather wet summer we are having so far; unlike most people, I'd quite like it to continue that way !

To change or not to change ...

In July, my Powerspout will have run for 4 years on the same set of bearings; in that time I've not greased them once; they've run only on their factory fill of grease and, having run continuously, the total hours will be just over 35,000.

EcoInnovation's recommendation is to grease regularly and to change the bearings every year (8,760 hours); I've not followed that advice, not because it's bad advice, but because I wanted to see how long I could get bearings to last.

Mind you, ...the bearings I'm using are a premium type from SKF, designated E2 Energy efficient and more expensive than their standard Explorer series; the claims for them are: "longer service life", "longer grease life", "reduced frictional loss" and "lower cost of ownership";

...and I've modified the sealing arrangement at the 'wet end' of the shaft to reduce the risk of water getting in and causing the grease there to be degraded; see here.

So the title of this post: "To change or not to change..." looks at whether after 4 years, the time is ripe for a new set.

According to the SKF leaflet, bearing service life is almost always limited by grease life; for their E2 bearings, SKF give the following chart to estimate how many hours the grease in E2 bearings should last; the estimate is based on an L10 grease life, and that is defined as the period of time at the end of which 90% of a sufficiently large group of seemingly identical bearings are still reliably lubricated.

The red arrow on the chart is placed to indicate the 'life curve' for an A value of 40,000 and that 'life curve' can be seen to intersect the logarithmic y axis at around 80,000 to 100,000 hours, the same as if the A value was 100,000. This means the L10 grease life should be at least 80,000 hours, - or 9 years of continuous running.

This being the case, I am not going to change the bearings this year; I'll hold out at least for another 12 months and think again when the next anniversary comes around in July 2022.

For the technically minded: 

The diagram shows 'grease life curves' for various values of A, under varying bearing operating temperatures, and for bearings working with a load value P = 0.05 * C; this value of P signifies the lowest loading and was assumed to be the value most likely to be applicable for the load experienced by the bearings operating in a Powerspout; the leaflet gives de-rating factors to lower the L10 hours for bearings subject to a higher load value.

the operating temperature of bearings in a Powerspout is almost certainly less than the lowest value given in the diagram (50 deg C); as the dashed lines of the life curves show, this will mean an arbitrary reduction in the value of L10 hours for any given A value.

the A value for the size of bearing used and the particular conditions under which the bearing is operating is calculated from:

A = n * dm

where: n   = rotational speed in revolutions / min (for my turbine 1000 r/m)

            dm = mean diameter, mm, of smallest bearing, 6005 size, ( [25 + 47] * 0.5 = 36 )

Thus: A = 1000 * 36 = 36,000 (which is near enough 40,000)

Post script added 11 September 2022.

One year on from when the above was written, the bearings are still sounding perfectly OK, so I have again decided not to change them and wait another 12 months.

On holiday but in touch.

I've just been away on holiday, - and owners of Powerspout turbines have decisions to make when they go away.

Those lucky enough to have a water supply which is constant can decide without thinking: - the turbine can be left running. But for those, like me, who have to change nozzles, planning is needed.

One option would be to shut down the turbine, but that would mean missing out on useful generation; an empty house still uses electricity and it's nice to know the hydro is meeting that demand.  Besides, I've set myself the target of generating all through each year without any breaks, and that means I don't come close to considering shutting down.

The alternative then is to keep going, and this is where it gets tricky; my crystal ball has to be dug out so I can gaze into the future to make a prediction about how much water is going to be available while I'm away.

The reason for this is that Powerspouts have nozzles with an orifice size which is fixed, and they are not automated in any way; many small hydros are automated; they have a nozzle orifice which is variable in size and can be adjusted dependent on how much flow is available, the flow being sensed by the level of water in a tank at the top of the penstock.

People can, and do, automate their Powerspouts along this line but keeping things simple is the line I prefer to take.

So the exercise with the crystal ball is to discern what size of fixed nozzle to leave the turbine with, which will still, I hope, be the right size for whatever is the flow at the end of my time away.

The things to look for in the crystal ball are recent rainfall, time of year, anticipated rainfall, temperature, and vegetation cover in the catchment area of the water source.

It doesn't much matter if you compute everything and get it wrong; if the flow dries up more than you thought it would, it's not a disaster.

What happens in my set-up is the head tank empties, water coming into it immediately goes out again down the penstock, setting up a new operating head somewhere down the pipe. This lower level of head ensures flow through the nozzle is less and the end result is a new equilibrium is reached where water arriving equals water leaving.

The operating efficiency of the system is rubbish when the turbine runs like this, - it labours at a much too slow rpm and power output is greatly reduced, - but the system does manage to keep generating and no damage is done.

Because of the technological wonders of the age we live in, there is no need to be in ignorance of how your turbine is doing while you're away. The photo shows that at the top of Carn Llidi in West Wales, I could check to see if the crystal ball had spoken truthfully.

In fact, when I got home I discovered it had not; far from drying up, 5 inches of rain had fallen in the ten days I was away; I could have been generating twice as much as the 220 W I set the turbine up for.

I need to get a new crystal ball !

How I use the electricity my Powerspout generates - in summer months

In the previous post I showed how electricity from my Powerspout is used in winter; in this I show how it's used in summer.

The seasons are quite different; winter sees generous hydro and little solar; summer is the opposite with lots of solar and little hydro. 

There are other differences: in summer the fire in the house is not alight and the space (storage) heater for the priority 2 diversion load is switched off; together these differences create a quite different pattern for what happens to the electricity which is generated.

As with the previous post, I have taken data from one particular day, an unusually sunny day from the look of the solar generation record, to illustrate what happens in summer; the two plots below display the data.

Plot 1

Plot 2

Explanation of the plots

as before, the vertical axis of plot 1 is power in watts, of plot 2: watt hours, and the horizontal axis of both is the time of day.

the green line represents power being generated, - in the hours of darkness this is only the contribution from the Powerspout (283 W), but when the sun is up, it rises to 3000 W as the PV array adds its contribution; the PV comprises 12 panels, each rated at 275 W (ie 3.3 kW peak).

there being no input from a fire to give the house hot water in summer, the first priority for any spare power is for this purpose; the diverted power appears as intermittent blocks of red bars, intermittent because a washing machine, dish washer and bread maker are programmed to come on during the night; when they draw power, none is available for diversion, especially since the Powerspout is only producing 283 W;

from Plot 2, a total of 2 kWh can be seen to have been diverted to house hot water by 08:20, - little of it coming from the Powerspout and most coming from the PV in the first 1.5 hours after sunrise.

after house water is up to temperature, power is then seen to be sent to the second priority load, signified by blue bars, until that too reaches its thermostat setting; 

the priority 2 load in summer is an immersion heater in the hot water tank of the annex;  when both annex and house hot water are up to temperature, no diversion loads are available and excess power is fed back into the grid; there are just short duration top-ups to one or other diversion load from time to time.

the total energy exported to the grid on this day was ~ 14.7 kWh 

imported energy from the grid was 2.7 kWh.

Conclusion

The big difference between summer and winter is the amount of energy which is spare; in summer, at least on a sunny day like the day studied, 14.7 kWh was put back into the grid, and to this can be added the 3 kWh which went to the priority 2 load because it is not usually necessary for us to have hot water in the annex;

in winter there is really none to spare; diversion to the annex which is optional in summer changes to being essential in winter; the 7 kWh which goes there is needed to take the chill off the rooms and stop water from freezing.

As I said in the previous blog post, the purpose behind analysing how our home generated electricity is used is to be able to make an informed decision about opting for home battery storage; I mentioned that one factor in that decision is what happens to the grid price of electricity.

This week I was notified of another increase ! 

On April 1st 2021 our off-peak energy moves up from 12.44p to 14.26p, standard day energy from 18.67 to 20.65 and the standing charge per day from 24.1p to 25p.

Now, I just need the price of batteries to come down !

How I use the electricity my Powerspout generates - in winter months

In January and February my Powerspout produces more electricity than at any other time of year; its power output is usually a constant 924 Watts giving an energy yield of just over 22 kWh per day.

The question I want to answer in this blog post is: how does this electricity get used ?

Below are two plots which show how the power and energy was used on one, unexceptional day last year: the 20th February 2020; the data was captured by the diversion device I use (a Solarcache) which has the ability to record information every 6 seconds to an SD card.

Plot 1

Plot 2

Explanation of plot 1.

the vertical axis is power in watts; the horizontal axis is the time of day.

the green line represents power being generated; during the hours of darkness, fairly obviously this will only be from the Powerspout, but between 08:30 and 16:00 a variable contribution from solar panels is added giving the spiky appearance mid-way through the day.

power in excess of what is being consumed in the house as 'base load' is diverted by Solarcache to two 'non-essential loads'.

these two non-essential loads are prioritised so that one, a 3 kW immersion heater giving the main house its domestic hot water (DHW), receives power first until the water is up to the temperature of its thermostat;  when power is being fed to this load the plot represents it by the red bars.

when the thermostat opens, Solarcache senses there is no load to supply and starts sending power to the second priority load, represented by blue bars; 

this second load is comprised of another 3 kW immersion heater supplying hot water to our 'granny' annexe plus, in parallel with it, a 3 kW storage heater for space heating in the annexe; the storage heater is set to allow air to convect through it at all times so it never gets hot enough for its thermostat to open; in this way the priority 2 load never switches off and is always available to receive excess power.

'base load' consumption in the house at any moment can be determined from the plot fairly exactly by the number of watts by which generation exceeds the power being sent to one or other diversion load; I say fairly exactly because Solarcache always allows a trickle of power to flow back to the grid, of about 40 to 80 W; this amount will contribute to the difference between generated and diverted power and makes it difficult to be precise about base load consumption.

base load is made up mostly of lighting (LED or compact fluorescent) but includes also a Grundfoss central heating pump (48 W), a fridge and a freezer (111 W each) and innumerable switch-mode battery charging devices; we have no TV.

when a power hungry household appliance is switched on, requiring more power than is being generated by the sum of hydro and solar generation, the diversion loads automatically switch off and the extra power needed to supply the appliance is drawn from the outside grid. 

the household appliances which make up the power hungry category are: kettle, hob, oven, microwave, vacuum cleaner, washing machine and dishwasher, - the latter two only being power hungry when they are heating water which they have to do as each is only fitted with a cold water supply.

the discerning reader will detect the times when power hungry appliances were in use on 20th Feb 2020 by the white gaps within the red and blue bars.

the provision of DHW in the main house does not only rely on its immersion heater; the tank also has a heating coil which receives hot water from the back boiler of a log burner in our lounge and in winter this is burning 24/7; thus the water quite quickly gets hot enough to open the immersion heater thermostat and for this reason the cumulative energy taken by the priority 1 load is less than that of the second priority load in the annexe.

Explanation of plot 2.

the vertical axis is in watt hours (Wh); dividing by 1000 gives the more common unit for energy: the kilowatt hour, - kWh, - aka a unit of electricity for metering purposes.

the blue and red plot lines represent respectively the cumulative energy diverted to the house load and the annexe load;

as explained above the priority 2 (blue line) receives the greater share because DHW in the main house receives additional energy from a log burner and can switch off, whilst the priority 2 load can never switch off and is always available to soak up power.

Conclusion.

this is how the system works in winter when winter conditions prevail: hydro power is plentiful, the lounge fire is alight and solar power is minimal.

the situation in summer is quite different and will be the subject of a later blog post;

the 'brains' of the system is the Solarcache and it is a huge pity this device is no longer available; mine has been bomb proof since installation 5 years ago and I am not aware of anything being marketed now which has quite the same functionality.

the SD card (8 GB size) which stores all the data has been quietly recording everything since August 2015 and last week covid lockdown stimulated yet another job to be tackled that had long been put off, - I ventured to download the accumulated data; there was 2.3 GB of it with each day accounting for 1.3 MB; considerable patience was needed because it took ages to get the data off the card to my desktop, and then MS Excel went into extra slow mode trying to crunch the figures.

such a huge amount of data is probably excessive but its usefulness is in being able to see how much surplus energy my combined hydro and solar installation yields at different times of the year; such information can help decide whether storing the energy in a battery, rather than use it for diversion load heating, might be a worthwhile road to take. 

For the moment, I don't think that is the way I want to go, but if battery prices come down and grid electricity prices go up, at least I'll have the hard data to help me make the decision.

A fuller description of Solarcache is in this blog post.