4 ms·

when renewables fail, they don't give you much warning (we're talking minutes at most) If your country has a lot of hydro (which has ramp times measured in sec

by m4x 13y ago

when renewables fail, they don't give you much warning (we're talking minutes at most) If your country has a lot of hydro (which has ramp times measured in seconds) this doesn't matter. Additionally, clouds don't magically appear and obscure an entire cities sunlight over the course of a few minutes. They pass over cities, so as one area loses solar generation another is regaining it. Or maybe it's an overcast day where the the solar generation is at a predictable low level all day. Neither of those cases pose any problem to hydro or any other plant that can ramp up and down over the course of a few minutes (and every nation has these - you can't run a grid without low-response time generators) In the case of plant failure, the system is much more resilient. And in the case that something disables most small plants (say, an earthquake), the system is much more affected than today (because rooftop solar just isn't built to power plant standards) I'm not really sure what you're saying here. That distributed generation gives you resiliency? In Australia and NZ, if the grid goes down then your solar inverters turn off too so it doesn't affect the grid's reliability at all. I believe most countries have this requirement. How does not being "built to power plant standards" have any effect? Your inverters are still required to generate clean AC, and if they can't match the grid closely for any reason they stop generating.

Swannie 13y ago
"In Australia and NZ, if the grid goes down then your solar inverters turn off" Yes... and that's precisely the issues here. How do you know when the grid is down? You'd see a drop across your grid connection? Makes sense. But what if the grid went down, but because all the local houses were supplying greater than 100% of the local load requirements, it never looked like the grid was down? That's the challenge. As I understand it, power utilities are considering "smart grid" as a distributed control mechanism to kill PV interconnects in such instances. However traditional grid control is normally built very stringently to allow fast detection of issues and fast isolation/resolution with redudant lines. If you have one rogue smart control point, you could end up electrocuting the engineer.
- m4x 13y ago
  If the grid supply goes down for some reason (natural disaster, car hits transformer, significant sustained overload blows pole fuse etc) then everything connected to the grid in that area will see some or all of the following effects: * sudden change in supply voltage * sudden change in frequency (and rate-of-change of frequency) * sudden change in phase angle You can't mask these signals and they are particularly pronounced on the section of grid that was affected (i.e. the part that is now blacked out). These (and more) are the specific effects that inverters monitor for, and they will disconnect if they notice anything unusual. Inverters aren't checking for the presence of a grid supply, they're monitoring it's health. Your local lines company dictates the trip limits based on your address and size/type of installation. As for safety, compare some scenarios that a lines worker could actually find himself in: 1) he has turned up to a fault and found that a HV line is on the ground and the auto-recloser is still trying to re-liven it at periodic intervals. The voltages are so high that walking anywhere near the fault could result in him being electrocuted due to a step voltage. The power levels involved are high enough to cause a major explosion. 2) he has to open or close a circuit breaker in a switch yard. He has no way to know whether the breaker is going to explode when he racks it in. At least he probably won't know about it if it does blow up. 3) he turns up to a fault on a 230V line, tests it and finds that there's still voltage present despite the earlier fault. He visits the houses connected to that part of the circuit who have a grid tied inverter and manually isolates them before investigating further. That last one is hypothetical, since I've never heard of an inverter failing to turn off and even if one did stay on initially it would very quickly trip on overload after trying to supply your house and your neighbors all by itself. Testing the safety of a circuit before you work on it is an everyday part of the job for electricians. Finding that a circuit is still live is very common, and having one re-liven unexpectedly happens too from time to time. You just take precautions, wear protective gear, and do your best to minimize the danger.

by m4x 13y ago

Swannie 13y ago

"In Australia and NZ, if the grid goes down then your solar inverters turn off" Yes... and that's precisely the issues here. How do you know when the grid is down? You'd see a drop across your grid connection? Makes sense. But what if the grid went down, but because all the local houses were supplying greater than 100% of the local load requirements, it never looked like the grid was down? That's the challenge. As I understand it, power utilities are considering "smart grid" as a distributed control mechanism to kill PV interconnects in such instances. However traditional grid control is normally built very stringently to allow fast detection of issues and fast isolation/resolution with redudant lines. If you have one rogue smart control point, you could end up electrocuting the engineer.

m4x 13y ago
If the grid supply goes down for some reason (natural disaster, car hits transformer, significant sustained overload blows pole fuse etc) then everything connected to the grid in that area will see some or all of the following effects: * sudden change in supply voltage * sudden change in frequency (and rate-of-change of frequency) * sudden change in phase angle You can't mask these signals and they are particularly pronounced on the section of grid that was affected (i.e. the part that is now blacked out). These (and more) are the specific effects that inverters monitor for, and they will disconnect if they notice anything unusual. Inverters aren't checking for the presence of a grid supply, they're monitoring it's health. Your local lines company dictates the trip limits based on your address and size/type of installation. As for safety, compare some scenarios that a lines worker could actually find himself in: 1) he has turned up to a fault and found that a HV line is on the ground and the auto-recloser is still trying to re-liven it at periodic intervals. The voltages are so high that walking anywhere near the fault could result in him being electrocuted due to a step voltage. The power levels involved are high enough to cause a major explosion. 2) he has to open or close a circuit breaker in a switch yard. He has no way to know whether the breaker is going to explode when he racks it in. At least he probably won't know about it if it does blow up. 3) he turns up to a fault on a 230V line, tests it and finds that there's still voltage present despite the earlier fault. He visits the houses connected to that part of the circuit who have a grid tied inverter and manually isolates them before investigating further. That last one is hypothetical, since I've never heard of an inverter failing to turn off and even if one did stay on initially it would very quickly trip on overload after trying to supply your house and your neighbors all by itself. Testing the safety of a circuit before you work on it is an everyday part of the job for electricians. Finding that a circuit is still live is very common, and having one re-liven unexpectedly happens too from time to time. You just take precautions, wear protective gear, and do your best to minimize the danger.