Comment by tomas789 | Hacker News Reader

Comment by tomas789 | Hacker News Reader

Wasn't this settled during Tesla and Edison? Their argument is that renewables are mostly DC. It looks like they mostly care about combination of SPV and battery storage.

Case 1: SPV 12 kWp @ 200V -> Inverter (eff=96.2%) -> 200m of 8 AWG wire -> AC-DC switching power supply (eff=92%)

Case 2: SPV 12 kWp @ 200V -> 200m of 8 AWG wire -> DC-DC buck converter 90%

If first case we get 9.16 kW. In second case we get 7.596 kW. In Case 2 you need the buck converter because the DC voltage will vary with temperature of the cable itself. Voltage loss in this case will be roughly 56V. In both cases I disregarded the effect of MPPT as it will be the same in both cases.

This suggests that when the cables are short enough (probably about 50m) then it might make sense. For anything else we are probably better off with AC networks.

One remark is that an industrial warehouse with rooftop SPV will probably need AC too. This might work for single family houses though.

With AC you can step up to higher voltage with similar efficiencies which will lead to lower transmission losses. I kept the voltage at 200V for simplicity. But going from 200V to 800V will lead to transmission loss of 3.43% instead of 13.73%. This way AC is clear winner.

The "war of the currents" was between Edison and Westinghouse. Tesla worked for both of them, but had very little influence on this issue. In the 19th century, using AC power was necessary because transformers were the only way to generate high enough voltages for long-distance power lines. This is no longer the case in the age of high-performance semiconductors.

In your calculation, you are comparing apples to oranges. An "AC-DC switching power supply" is just a rectifier followed by a DC-DC converter. There is no way for your "Case 1" to be more efficient than your "Case 2" because it simply contains two more conversion steps. This is exactly what the Fraunhofer project is aiming to eliminate.

Feel free to re-calculate the numbers and point at the mistake. I calculated everything to the best of my knowledge.

At the same voltage, DC will in all usual cases have higher losses than AC. This is basic physic.

Non-constructive critique helps nobody.

I think with new(er) GaN(or SiC) DC/DC buck conversion, 380V systems can get up to 98.2%, typically around 94% efficiency now.

Edit: And voltage of these type of DC grids are often higher (600V - 1000V, or industrial even up to 1500V), so losses in cables are less.

Yes. The original commenter is a bit behind the times. There is also reliability advantages.

> new(er) GaN(or SiC) DC/DC buck conversion

Can you share guide lines about costs?

For comparison, utility scale AC inverters for PV are ~$5-$10/kW

Original commenter might be a bit behind on many things but this is actually my day job.

Do you have a source for the reliability claim? There are so many good old transformers deployed around the world. Many of them working for 10s of years without a replacement. I'd be suspicious that a buck converter will have longer MTBF than oil submerged transformer.

Also, I don't see why it is mentioned here but NREL estimates the cost of grid scale inverters (for installations of 100 MW) to be 5-10x more than what was mentioned. [1]

[1] https://www.nrel.gov/docs/fy22osti/83586.pdf

> but this is actually my day job.

You are probably more knowledgable than me. I am just here for the cake.

However, AFAIK the most fault prone components in PV are inverters and (where applicable) transformers.

> 5-10x more than what was mentioned.

This was my mistake. I started with 5c/W and missed a decimal place.

None of this addresses the key issue - you mentioned 10% loss for a simple DC converter, and the next commenter [1] mentioned the newer generation of far superior alternatives.

[1] https://news.ycombinator.com/item?id=37260441

AFAIK PC power supplies are actually one of the most efficient power supplies out there so I'll use them as a reference. The Titanium class of power supplies is around 94% efficient at 100% load and 90% at 20% load [1]. On the other hand plain old transformers start at around 95% and can go even higher. Random internet source claims that it can be 98.5% efficient [2]. Of course you cannot run your computer at AC. I had mostly grid-scale equipment in mind. I just wanted to use something relatable as an example.

[1] https://www.velocitymicro.com/blog/what-is-psu-efficiency-an...

[2] https://www.electricaleasy.com/2014/04/transformer-losses-an....

They didn't have frequency converters / fast high power transistors back then. With AC, AFAIU, it's easier to change voltages with primitive equipment. But then you have to have the frequency the same everywhere.

What about capacitive and inductive losses. These are significant in AC transmission lines, depending on their construction. I believe that's why high voltage interconnects between different countries are usually DC and use rotary convertors at the ends.

The main reason for HV DC historically was that it doesn't require grids to be synchronous, or even of the same frequency. This is often desirable because a synchronous grid requires deep organizational integration. HV DC also lets you use mono-pole systems fairly easily, since they're single phase; sea or ground becomes the return path.

They are both taken into account in the calculation. I assumed power factor of 0.85.

AC has the great advantage of being well supported. However, scratchpad calculations can't prove that.

Reply on news.ycombinator.com