Chris Yelland critical of the regulations

NRS/SSEG Regulations
41

Wouldn’t the frequency of the grid relate to the massive generators’ frequency? These things are like giant flywheels with a huge momentum. I don’t think the frequency at which AC alternates at would depend on anything but those generators?

Which raises something I’ve been thinking about: Flywheels are like massive mechanical stores of energy, while batteries are chemical stores of energy. Would a solar system not benefit from having an actual flywheel providing the frequency (and some inertia to the supply) since batteries are DC and we all know that our solar panels have absolutely no inertia. As such, our batteries provide our PV systems’ inertia, but would it not be better to have an actual AC source between the inverter and our homes?

Then the inverter would make sure the flywheel keeps on spinning, and the flywheel doesn’t need to be massive, just enough to smooth power delivery and actually create a “true” sine wave?

Would a critically high figure not also be reported on? We probably didn’t have that situation, but if we loadshed a large area without turning the lights back on for a similarly large area, would this not cause the frequency to start increasing which could be as dangerous and potentially even more so - A runaway flywheel could cause some serious destruction… (I could be entirely wrong, because I haven’t done any reading on the subject, only thought a little about it the other day after I read @Phil.g00 post about the web farms disconnecting themselves from a grid with a dip in frequency).

42

The grid frequency is like a wall. A generator runs up against it and cannot increase the grid frequency (or only fractionally)
The DC equivalent generator will produce a higher voltage but this will be limited to the local grid voltage since the load will draw from this higher voltage first…

43

I’m not proposing a DC generator, but an AC one - So an actual flywheel that your inverter gets to spin at the grid’s frequency - Then you have a pure sine wave “inverter” as well as a little bit of inertia (in a more robust form than a chemical battery)?

44

I liked this idea …

Or THIS!?

45

You guys may enjoy looking on YouTube how to synchronise generators. They essentially throw the switch when the phases is ALMOST aligned (at 11PM on the dial so to speak) so the generator is essentially pulled into phase as it joins. It is indeed, as Richard says, a wall that is hard to push alone. All the generators in the country essentially synchronise on the same frequency.

That brings up Cahora Bassa. That comes down from Mozambique on a high voltage DC line, and then it is switched onto our grid using a couple really large Thyristors (which are switched by the AC side too!). So that is one practical way to move energy between networks that may not be synchronised.

46

That’s the easier of the two problems. Steam turbines have a governor just like most other mechanical engines. Open a valve and relieve some steam and bob is your uncle. No overspeed. The reverse issue (can’t keep speed with the present steam pressure!) is a bigger problem :slight_smile:

1 Like
47

Lots to unpack here.

Think of frequency as a system thing.
As @plonkster says think of voltage as a local thing.
(This isn’t strictly true all the time, but it is the best way to picture things).

System frequency is tied into power generation and the control centres try to match load and generation very closely to 50hz. (This is a power matching thing).

There are many things ( eg. motors, transformers) that get designed to work specifically at that frequency.
System faults can manifest as sudden load upsetting that fine balance with a momentary frequency dip, but then the protection will disconnect the faulty part of the network and hopefully a minimum amount of load and the net effect will be minimal. This is because the system tends towards stability because it has a lot of mechanical inertia to ride through the blips.
Within a very tight bandwidth, the national control centre actually has the tiniest bit of play.
It’s been 30 years since I toured South Africa’s national control centre at Simmerpan. (At that time national security was very high and tours weren’t just a thing).
I was quite surprised to see why frequency was tweaked within that very tight bandwidth.
Extremely technical devices were used to adjust the system frequency throughout the entire country.
The controllers had two clocks on the wall, an AC one and a DC one. With the tiniest of adjustments, the aim was to have both clocks reading the same time at the end of the year.

They reset the clocks every year. I remember being told that they had lost two minutes the previous year.

What also has been touched on in this thread, is flywheels and mechanical momentum.
Yes, it is a very important consideration, especially when considering generation by things like inverters.
Electronic generation doesn’t have mechanical momentum like a heavy spinning armature.
And by that, I mean electronic generation has none, nada, nothing, not some.
Even wind generation, although they use heavy turbines, once their inverters break the chain any mechanical momentum is lost.
DC power lines also break the chain.
Motors on the other hand will also contribute a bit of mechanical momentum.
Part and parcel of having that frequency stability are that heavy things take time to speed up and slow down.
As @Richard_Mackay says that brick wall effect is needed for system stability.
Frequency swing is not a binary event, it swings up beyond the 50hz target and back below it, over and under until it gradual settles. Like a pendulum gradually settling at its nadir.
Too far over and under and things start to trip at both high and low frequency, exaggerating the swing, making it more and more unstable until the entire system is lost.

I hope I am getting across that too far away from 50Hz will be uncontrollable and catastrophic.

(When generation cannot balance the load, frequency instability is a threat without load-shedding.
This is also why if ZA ever lost the system, it would probably take months of a precarious load/generation balancing act to put it back together after a black start. I know that load-shedding is a PITA, but the alternative would be far worse. It’s akin to a doctor performing an amputation to save a life).

So the ratio of generation with momentum and generation without momentum becomes important.
Too much electronic generation and there is insufficient inertia to keep the frequency stable.
(By the way, this is another little technicality Mr Yelland doesn’t allude to).

@TheTerribleTriplet There are already devices that add mechanical momentum like flywheels. These are called synchronous compensators. Essentially, these are synchronous machines that are neither run as generators or as motors.
They are only supplied with sufficient power to spin themselves, they in turn don’t work as motors. By altering the DC excitation they are used for power factor correction because the wattless power can be varied. Although nowadays, pf correction is also being done by static ( read electronic) devices.

So ESKOM’s big generators ( also synchronous machines) are probably actually needed as a spinning reserve ( ala electrical flywheels) even if the generation is not required if ZA is to go renewable at a high percentage

So when all is said and done, there is a bit more to things than Mr Yelland would have you believe.

3 Likes
48

With regards to synchronizing two 3ph systems that have the same phase rotation.
There are a few criteria that have to be met within an acceptable window.
The systems:
have to be close in voltage.
have to be close in phase
have to be close in frequency.

Synchronizers are quite interesting things, because if you match the frequency perfectly, you can never alter the phase difference. So one system has to be at a slightly different frequency, so that one system is very slowly turning relative to the other.
Then it has to be set to trigger in advance ( so 11 o’clock, not 12 o’clock) so that the operating time of the switch and the relative slip is compensated for and it nails it.

From time to time, it happened at Vaal power station and at Arnot, a generator gets mistakenly switched in in complete anti-phase. This is actually the worst electrical fault in terms of fault current that is possible.
It cost the gen transformer at Arnot.
The Vaal event was before my time, but I understand the generator left home and decided it prefered to live on the other side of the river.

1 Like
49

So do you think it would be practical to add such a flywheel in a PV system at home? As I’ve said, putting it between the inverter and the loads. Or do you think it’ll have to be too big/expensive to be practical at a small scale, or cause too much losses?

Then your inverter’s job will be to keep it spinning at 50Hz, and it could even do it with DC (no conversion losses needed).

50

The only flywheels I’ve heard of are used by data-centers (esp banks) to maintain power in the event of a power outage.
I have never seen one but I believe there is enough stored kinetic energy to keep the power on long enough for the genset to be started and come on line…

51

I don’t need it to carry me for an interruption, I need it to give my home a frequency (instead of the inverter doing it) after the power is out (or I’m completely off grid) as well as provide a little bit of inertia. It doesn’t need to store tons of energy.

I guess my question is simpler: Why do we use electronics to create an “almost” AC signal instead of using DC to spin a flywheel at 50Hz and then use the flywheel to produce a pure AC signal. Is there almost no benefit to it and the cost would be much greater?

52

My opinion is it is far too expensive an undertaking for an individual. These are things that island communities may look into.
The way I see it if you have a massive solar farm with a massive inverter you are OK with a massive off-grid system.
Then there is a system frequency set centrally and everyone in the community has loads.
The same thing with a smaller off-grid system, it shouldn’t be an issue.

The problems arise if you try and make a big system out of a lot of little domestic size inverters. These inverters are geared up to export power because they continually test there is a stable grid frequency out there. But in that instance, there won’t be.

Not unless something like a synchronous compensator is part of that community system.
(And I am not saying, this is the primary function of a synchronous compensator, it is a by-product. So this is just theoretical, I don’t know if this is actually done or exists out there in the wild).
But even if it is viable, even though these machines don’t draw load as motors, they are still big heavy machines and they don’t spin at full rpm without losses.
So practically, only a community could collectively afford such a machine and there would still be a considerable constant energy usage overhead involved to run one.
It would probably take an amount of pioneering spirit, cost analysis and design for a community to undertake this.

Well, it does actually need to store tons of energy, it just doesn’t need as much to run.

And a synchronous machine does actually use DC. Its armature is a DC winding. The stator is AC winding that results in a spinning magnetic field. The magnetic attraction is so strong between the AC stator and the DC armature are in lockstep at synchronous speed.
Synchronous speed depends on system frequency and the number of poles in the stator winding.
For example, synchronous speed is 3000rpm for a 50Hz 2 pole synchronous machine.
There is no “slip” between the two magnetic fields.
A squirrel-cage AC motor actually doesn’t spin at the same speed as the AC magnetic field spins around in stator, there is a slip, it is always going slower. It will never actually reach 3000rpm.
This is why you can start an AC motor easily. It will speed up from zero rpm to its rated rpm over time.
You can’t do that with a synchronous machine, the mechanical stresses involved to go from 0-3000rpm instantaneously will destroy it. You have to speed it up with another technique to take it to nearly 3000rpm and then make the magnetic coupling (Turn on the DC) and then the rotor will lock into the stator magnetically. That “taking up to that nearly 3000 rpm” takes tons of energy because you need another big AC motor to do that work.
Now at the synchronous speed, you still have to supply the free-wheeling losses, which will be a comparatively smaller but still a constant amount of energy.

I hope this makes a bit of sense.

1 Like
53

I just reread this and had a little chuckle.
I’ll simplify your question even further because it is actually what you are asking although I don’t think you realize it:
You are actually asking: Why do we use an inverter instead of just building a power station?

54

Thanks! I’ll read it a few times and hopefully I understand it all then!

On the power station bit - is that effectively what is required to have a stable 50Hz flywheel spinning to give your system some inertia? I (obviously very ignorantly) thought it might be possible to simply spin up something (a flywheel in a type of housing) using your DC power source up to 50Hz and then keep it there - a mechanical inverter of sorts?

1 Like
55

That’s why I had a chuckle a mechanical inverter is better known as a synchronous generator.
The energy source that drives a generator can be steam, water or whatever. It’s irrelevant, it’s still a power station. You just want to use a DC motor.

Unfortunately, the only thing I know of that can imitate a monster chunk of metal spinning at exactly the system frequency, is another monster chunk of metal spinning at exactly the system frequency.

I am not saying the quest is not a noble one or that it’s impossible, but right now it hasn’t been invented as far as I know.

56

Haha ah okay, I think I’m starting to get it (at a very high level, not technically), but still need a few reads and thinks. In the meantime, does it absolutely have to me a monster chunk of metal? What if it is just a smallish chunk of metal (cheaper and more practical) since my home don’t need that much? As long as the standing losses to keep it running (the friction part) is less than 50W + the 10% of DC to AC conversion losses (presumably you won’t need to electrically convert any DC to AC anymore, you would just need to compensate for the friction losses) it would be just as efficient as my MPII. :grin:

57

I don’t know, what size it needs to be, I just know it needs to represent an immovable reference.
I’ll use an analogy, I could move with a 20 kg ankle weight, but I couldn’t move with a 10 ten ton ankle weight. Then again I couldn’t move with a 200kg ankle weight either.
But I am just 1 individual, Popeye ( with the strength of 10 men) could probably move after some spinach.
So Yes, for a single smaller installation it could definitely be smaller, but I don’t have a handle on how much. It will have to be proportional to the generation capability.

1 Like
58

So there’ll probably be some physics equations to tell us exactly what size is needed, but do I understand correctly that its inertia needs to be such that your demands cannot slow it down in the time it’ll take your source of power (battery or PV) to jump in and keep it spinning?

Man, I know absolutely nothing about these things but I find it incredibly fascinating.

59

Yes, it would be kinetic energy >> than electrical energy equation.
You have got the concept, but it is not just about slowing down it should also ensure a reluctance to speed up while you are generating.

60

Dare I suggest that we simply do away with AC altogether?? :smiling_face_with_three_hearts: