Danijel Gorupec

I was looking for long-term energy consumption trends. The first graph I found here http://oilprice.com/Energy/Energy-General/A-Look-at-World-Energy-Consumption-Over-the-Last-200-Years.html Of course, I am not sure if this page is correct, but 10 times energy consumption increase over 100 years sounds quite real to me. Of course, the same trend does not have to continue.

Working further from swansont's 0.01%... If we only cover 10% of Earth area (supposing that more would make deep scars over natural world), then our current consumption is more like 0.1% of what we can obtain from solar in a "green" way. In 100 years, it will be 1% (no change in consumption increase rate) in 200 years, it will be 10% in 300 years, it will be 100% Now, 300 years is a looong time, but not that long that one would not be interested to ask - and what then? My guess is that it is simply not possible to continue our current growth rate for more than few hundred years. If we want to stay green, we will either have to stop advancing (either by loosing interest or by war) or we will have to miniaturize ourselves.

I actually already released the software having only a statement, without legal text (whatever it means). I decided to do only that much after Acme suggested that there will be no automatic prosecution anyway... I suppose the only thing a potential user should still worry about is am I going to change my mind in future (That is, if there is no automatic prosecution, then I suppose a legal text only has meaning to ensure users that copyright owner cannot change his/hers mind any more).

I guess... if you start transferring really large amounts of energy toward the Earth (say, and I am just making a wild guess, 100 times more than we currently use) then you will start heating the Earth surface noticeably. So, I guess there are practical limits on how much additional Sun energy we can deflect toward Earth (in any form). On the other hand, there are probably not practical restrictions on how much energy can be harvested from Sun (at least not for some 1000 years in future, possibly much more). There is enough energy there to vaporize our planet many times. (lol, I once had a silly SF idea to make huge, super-light paragliding wings that will float near Sun on its light and solar wind, and at the same time reflect light and focus it on Mars to heat it up.)

I see... but wold not then this other person need to somehow prove he/she is the author (copyright holder)? Okay, I suppose it is not hard to change one single byte in the code and become its author.

Thanks for this explanation... In this case I only need to sound honest and make sure that end-users understand that I will never move against them. I don't need to be concerned about legal matter.

Are you saying that there is no one else than author (copyright owner) that will start a legal action against a copyright offender? Does it mean that those permissive licenses are only wordy to ensure (comfort) the end-user that it is all right to do stuff that are usually illegal? I was under impression that these wordy licenses are needed to protect the end-user from some automatic prosecution.

I just read about the CC0 license, thanks for pointing at it... Yes, this is what I legally want (only it is way too long to make it effective as a marketing tool). In the process of reading, I also found the WTFPL license that is much shorter, but its legal strength I cannot estimate. I find it funny that it takes so much effort to abandon your own rights.

Yes, I considered the 'copyleft' but found it too restrictive for my users (I want to be outrageously generous because this is a part of my marketing strategy. This is also why I would like to restrict my license wording to plain and striking English.) Still, I would be interested to learn if the copyleft sign (or the 'copyleft' word) has any legal meaning?

As I understand, copyright is obtained by mere creation of an original work - including software. There is no need to request it... However, I want to abandon this right for my software (if this is at all possible). I understand that there are some licenses available that will do this, but I find all of them way too wordy (yes, I am serious). The question is... can I do it in one short sentence? Here is my try and I wonder if it has any legal weight: "This is a free software and this particular copy is now completely in your ownership."

Amazing! Obviously crazy!... I suppose the Argan fruit is a laxative that makes them vicious too.

I see your point, Swansont, but you are probably exaggerating: lawn mowers need to be made, fueled and disposed; are not particularly efficient; are not particularly 'natural'; and you still need to walk behind them. But what I wanted to note is that when people are using their own physical power then people have innate sense for efficiency. This is lost when we use machines (and slaves, I suppose). This is why I think it is not realistic to expect that general population will use machines to max of their efficiency (you will let the lawn mower idle for extended periods just because you are lazy to optimize the mowing path - but when you need cut the grass using your own power then you will think all in advance).

I am sorry to hear this, studiot... this cashier girl was very rude.

Thanks for clarification.... I was confused because here we can sell solar-generated power to grid at higher price than we buy from grid. No market for Power Wall in Croatia. Still I would be interested to see the pavelcherepan calculation if no Power Wall is used (while I do understand that some 'battery balast' is probably beneficial to your home equipment). Just sell/buy the power difference from grid in real time.

So, when you say you can get $20 for feeding into the grid, you are only talking about eventual excess energy you gathered?

I don't get it.... If you can feed to grid and get paid, why do you need the Power Wall? (Is it because grid pays more during night? Hmm...)

Hmm... this magnet piece might not be a straight bar magnet with two poles (N at one end, S at other end). Instead it might be a Halbach array or something... So, some of above mentioned tests might not work... Also, it is hard to break a metal road with no metal tools on hand... So I guess the best answer is the one that says you should take a large stone and hammer one of the pieces heavily and check if magnetic force decreases (or if you have a pile of coke & wood on hand, you can try to heat one piece above its Curie temperature). You have 50% chance that your test will be non-destructive, lol.

I suppose the simulator tool shows all those spikes because of parasitic capacities in transistors (while I don't know how detailed transistor model is used). If so, then by increasing R3 resistance the spike will become shorter, as you said, but also wider - is this what is happening? Note that in switching applications you usually deal with those parasitic capacities by, funny enough, decreasing resistances (usually by decreasing resistor value that leads the current to/from transistor base). This way you make spikes higher but thinner - making your transistor switch faster. If indeed these spikes are from parasitic capacities, then you don't need to worry. A 200ns-wide spike is not a problem in your application.

I am using, for small drivers (about 2A), BYV27 diodes because I have them in the local store. These work fine. BYV28 are larger... Anyway, there is large selection of fast silicon diodes. Schottky I never use (cannot find them here). Only now I see your picture with spikes on R1.... This is simulation.... can you see the width of the spike? This could be only parasitic capacity of the transistor (base-emiter junction). Edit: Yes, if possible find diodes with 2.5A, but this is not necessary because normally even smaller diode can handle 2.5A for small fraction of time. So, my guess is that even 1.5A diodes would work without problem in your case (but, because in electronics we always add considerable margin, using 2.5A diodes is a smart move).

As Spyman nicely showed, resistors come in various standard power ratings. Most often you will find 1/4W and 1/2W (every electronics store)... It is also good practice, as Sensei mentioned, to have some power margin (not necessarily double, but some 50% would be nice). For your simple drive circuit you don't need Schottky diodes... just use some that are faster than typical diodes used in 50/60Hz rectifiers. (If you made a high performance PWM driver then Schottky might be recommended, but in your case is probably overkill).... Anyway diodes are must and if you don't have them you risk damaging your transistors (even motor insulation, maybe). I am not sure what kind of current spike do you see on R1 (during switch-on or during switch-off)? Do you use oscilloscope to see these spikes? BTW, if you don't use diodes all kind of spikes are possible, but I cannot tell for sure if this is the reason.

If you can decrease your 24VDC power supply to, say, 15VDC (and re-calculate resistor values accordingly) you should decrease resistor heating a lot. Or you can keep as it is, and use more powerful resistors (or use parallel/serial combinations of several resistors). Or you can program your PLC never to keep single output switched on for a long time (when motor is stopped, switch output on/off with, say, 50% duty cycle, as fast as you can).

Yes... and I can add: - you will want to know how hot gets your motor when its temperature stabilizes at some winding current (preferably through both windings simultaneously). It can be quite warm, but not so hot that you cannot hold it in your hand nor that it starts to smell funny (it is too late then). It takes loooong time for temperature to stabilize (say 15 minutes for larger motors) so take your time before you decide to increase the current further. - you will check other motors in catalogs that have the same physical size and construction and about the same winding resistance... this will give you a good estimation about the current your motor can use.

Yes, at first look your design seems ok. Basically, only diodes are needed (make sure you use faster ones). If your power supply is dynamically slow, you might add some capacitors in parallel to the power supply. Your resistors might overheat. I think you can safely double values for R1 and R2 and still had enough base current (modern transistors have high gain). Also place transistors on a heat sink (this is always the hardest part for me). Now... you use 12VDC power supply and 5ohm winding resistance. Do you expect your motor can work with 2A constant current or do you plan to use PWM?

Hmm... if you use a more complex power supply (with +V, GND and -V) then you can permanently connect one end of a winding to GND and then use two switches to bring the other end of the same winding to either +V or -V.... This would be solution if you care not to have many switches (but the power supply is then more complex). Edit: I see now you have the same solution It would be the same with a H-birdge... This is a typical thing you have to consider when you build a bipolar drive. Your switch-timing must be carefully considered (when transistors are used, for example, care is taken to always close a transistor few nanoseconds before opening the other one). (Also, and we didn't talk about it yet, you will have to add some diodes to protect transistors/switches from high voltage generated by motor coil when it is switched off)

This is normally done wiht a H-bridge (read Wikipedia)... ask if you have questions about a H-bridge. There are maybe simpler ways than H-bridge, but are ridiculously inefficient.