QuickTips #1: IC2 Energy - will it blow?

[ShneekeyTheLost]

Hello all and thank you for joining me on my pilot episode of QuickTips, advice for the newer users of FTB and modding in general.

Today, I'm going to address one of the things that can be most confusing to new players, and that is Industrialcraft Energy.What will make a cable explode? Why does it draw more energy than it is supposed to? What is this whole 'voltage' stuff?

Industrialcraft^2 is a great mod, with a lot of fun features. One of the best is the good ol' fashioned Macerator which you generally want ASAP to double your ore output. However, it requires power. And that's where we get started with wondering WTF is going on with energy.

When working with energy in IC2, the most important metric when predicting cable melting is Packet Size, or 'how much EU is going through in one big chunk'. Copper cables, for example, can handle a packet of Eu up to 32 Eu. Notice, I didn't say 'Eu/t', there's a reason why. Right now, we are looking at how big a packet the line can hold, we don't really care how many of them are getting spit at us, nor how frequently they are being spit out.

So, if I have a bazillion water mills set up in an enormous auto-feeding system which lags out the server every time I turn it on, it will produce a bazillion Eu/t. However, it's a bazillion packets of 1 Eu coming down the line every second, so even Tin (Very Low Voltage) cable can handle it just fine.

But wait, you say, my leet system isn't producing a bazillion Eu... it's only producing something like 120ish/t. WTF?

Well, cables also have Energy Loss, which is PER PACKET. So if I have 8 packets of 32 Eu going through the Insulated Wire, then it's going to take Eu loss on each of those eight packets every 5 blocks.

Let's take the bazillion water mills example. Tin cable has an Eu loss of .025/block. Eu is always truncated (rounded down), so unless you go more than 40 blocks, you're fine. But, you hit 40 blocks, and each packet loses 1 Eu out of it. So yea, those bazillion water mills each producing 1 Eu/t? It goes 40 blocks, and takes a Eu loss of 1. Last time I checked math, 1-1=0. So your energy is gone completely.

You can look up the wiki for exact energy loss numbers and maximum packet size.

PROTIP: Higher tier wire does NOT necessarily mean less energy loss per square. In fact, it almost always means MORE energy loss per square, but they turn out more efficient over longer distances. Confused yet? Let me explain.
So why does higher voltage wire mean more efficiency over longer distances? I mean, you've got worse energy loss numbers, shouldn't distance actually make it even more worse?

Not really. Let's look at an example where we are trying to send 512 Eu/t across 50 blocks.

Spoiler: MATH AHOY!

If it is in low voltage, using Insulated Copper Wire, then you will lose 1 Eu every 5 blocks, PER PACKET. So, that's 16 packets at 10 each, for a total loss of 160

Double insulated gold wire, on the other hand, can carry MV, or 128. So there's only 4 packets. So despite the fact that it loses 20 Eu per packet (twice what the copper had), because it has four times less packets running through the wire, it comes out to around 80 Eu loss.

Now, 3x Insulated HV Cable can carry all that in a single packet. So despite having an Eu loss of .8/block, it's only a single packet getting hit by that, so your total Eu loss is 40.

Of course, Glass Fibre cable can do it with a loss of a single point, thanks to being able to carry HV voltage and having an energy loss of .025. But then, they also cost diamonds, so you'd expect that level of efficiency.

So, to recap:

• When determining if your machines will explode or your cable will fry, the only thing you should be looking at is the largest single packet going down the line. The number of packets is meaningless.
• When determining how much Eu loss is occurring, you need to count the number of packets and multiply by the Eu loss from the cable you are using. The size of the packets is meaningless.
• Use Transformers to step up or down the voltage to increase or decrease the packet size to maximize efficiency versus line compatibility.

 

[Juanitierno]

Heya!

How many packets per tick can stuff output?

It happens to me that i have a full MFE -> LV transformer -> machines and the machines dont get enough power when overclocked (with proper storage upgrades).

Are the transformers/LV limited in the packets/t they output?

 

[Aixler]

Transformers seem to be limited to 4 packets per tick.

 

[Cloud]

I don't think passive watermills give out 1eu/t but instead 1eu/p every time it accumulates to 1 eu. However, this isn't very reverent in this guide.

 

[Exe19]

Cables and transformers can hold infinite number of packets per tick. Transformers only split big packets to 4 smaller and vice versa.

 

[Bluehorazon]

Cables and transformers can hold infinite number of packets per tick. Transformers only split big packets to 4 smaller and vice versa.

The problem is that transformers only output 4 Pakets per Tick, so if you want to run a LV-Line with high traffic (like 512 EU/t) through a transformer you will only output 128 EU/t, since the transformer seems to be able to only handle that much. But well... you can easily split the line equally for 4 transformers and LV-Transformers are not that expensive.

Since Batboxes are Cheap (well with Gregs you can build MFEs with Rubies or I think also Lithium-Batteries) it is often better to supply a machine with 128 EU/t via 4 Batboxes than a MFE.

For Watermills, they produce 0,025 EU/t for each water-block that is next to them if you feed them it is 2 EU/t (although the autofeeding does not cause that much lag if you do it correctly, like with TE although this will make you lose some EU, because you have to power the liquid-transposer, the most lag is created by the timer, which has to run incredibly fast for a large system.

People should also note that most machines have an internal storage, so that machines do not waste EU providing them with energy. If a machine drops EU it will get a new paket which is stored completly inside the machine. This might cause situation in which machines overfill, since if a Batbox with 39.999 EU accepts a paket of 32 EU it will have 40.031 EU stored (shown as 40.000).

Although not shown IC2-Machines internal storages are cabable of performing a single operation (and are completly filled once by feeding the machine redstone), while GregsTech machines normally store 10.000 EU (at least the Grinder, Blastfurnace, Electrolyzer and Centrifuge behave like this).

 

[ShneekeyTheLost]

Manual feed output for water mills got nerfed down to 1 Eu/tic, which is what I was referring to with autoloading.

I wasn't aware you could hook up a LV transformer to receive HV, I was under the impression it would explode. Nevertheless, you always want to stage it down HV ->MV->LV.

Also, energy storage devices can be used to bypass Eu loss (placing a batbox after 4 copper cables, for example) and for stepping up voltage quickly (An MFSU outputs HV, even if it being fed with a solar flower on tin cables).

 

[whythisname]

Actually, Water Mills can give 2 EU/t, but they only do so when fed with Water Cells/Bottles. Water Buckets do produce only 1EU/t.

Note that Water Cells/Bottles get consumed entirely, so it'll cost you tin/glass besides the water that's in them, while buckets only get emptied.

On the topic of Water Mills, I've been experimenting and Glass Bottles are a very effective way to power the Mills. If you use Rotary Macerators and Induction Furnaces to generate glass from cobble it'll take you relatively little power. I've tried other ways of generating glass, but those seem to be the fastest and most efficient.

Also, if you make a setup with buckets, try to fill them with Dispensers. Free and fast filling. Though I have to say, an efficient system with Dispensers probably requires some (Autarchic) gates. Which I personally consider to be expensive enough that you can/should have other power sources by the time you can use them.

Anyway, enough about Water Mills.

Something funny to note, if you feed more than 32 EU/t into a Batbox it will charge up because it can't get rid of the power fast enough. The same goes for any other energy storage, so energy storage doesn't always have to be based on how much you want to store. You also have to take into account how much power is coming into that storage.

Additionally, when setting up a plant that needs energy to produce energy make sure you can't suck the whole system dry and lock it up that way. The easiest way to do that is to have your main power cable split in 2, 1 goes to storage for the production line, the other goes to storage for whatever you want to use it for. That way each output has some backup and the production line will always have power even if it gets cut temporarily.

Also, there is an EU-Reader to measure EU/t on cables and machines. Very useful to find out if your machines are getting enough juice and to see how many of those 1 EU Packets you're sending down a wire (it'll add those up, so you can see what storage/transformer you'll need to handle the total output).

 

[Bluehorazon]

Manual feed output for water mills got nerfed down to 1 Eu/tic, which is what I was referring to with autoloading.

I wasn't aware you could hook up a LV transformer to receive HV, I was under the impression it would explode. Nevertheless, you always want to stage it down HV ->MV->LV.

Also, energy storage devices can be used to bypass Eu loss (placing a batbox after 4 copper cables, for example) and for stepping up voltage quickly (An MFSU outputs HV, even if it being fed with a solar flower on tin cables).

Nope you can't transform LV directly into HV, with the LV-Transformer since it only outputs 128 EU/packet. But technically you can transform LV directly into HV with an MV-Transformer. But not sure I have never tried it.