 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Electronics : Capacitor Charge time.
| Author | Message | ||||
| Davo99 Guru  Joined: 03/06/2019 Location: AustraliaPosts: 1577 |
I would assume this is a question with a lot of variables but.... In near enough terms... If I have a 1000Uf Cap and put 500W into it, how long should it take to charge up? Would it be 1-2 Sec or 5 sec or 10? Something ball park is close enough for what I want to do. Is there a way to determine how long a cap bank takes to charge with a certain power input? Any online calculators? Conversely, If I present said charged cap with a multi hundred watt resistive load, how long should it need to be connected for a full discharge? Would 1 or 2 seconds suffice when connected with decent ( 2.5- 4mm) short cable? Thanks. |
||||
Gizmo Admin Group  Joined: 05/06/2004 Location: AustraliaPosts: 5012 |
It all depends on the current. If you tried to charge it with a battery that could supply 1000 amps, it would charge very quickly. If you tried to charge it with a 9v radio battery, it would take longer. The current is represented by a series resistor in most calculations. Its a RC circuit. See https://www.omnicalculator.com/physics/rc-circuit Glenn The best time to plant a tree was twenty years ago, the second best time is right now. JAQ |
||||
| Davo99 Guru Joined: 03/06/2019 Location: AustraliaPosts: 1577 |
Thanks for that. The setup I am thinking of is charging from an array of 4x 250W solar panels. That would have a resistance of around 14 Ohms. I calculated the power conservatively at 500W so I would round off if the factor was constant. Using that Calculator, it seems the charge time would be well within the Sub second. Dose that sound Ballpark Correct? Not worried about fractions of a second so if 1 sec is enough, that's good enough. I would take it the discharge time is the same if not faster. If I then went to say 4000UF, I should still be able to charge that within a sec and if I short charged then discharged, that should not matter either. |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
Its very difficult to calculate a precise answer, because as the voltage rises (or falls) the current usually does not remain constant. But you are quite right, its going to be fairly quick in human terms, less than a second. Cheers, Tony. |
||||
| Davo99 Guru Joined: 03/06/2019 Location: AustraliaPosts: 1577 |
Thanks Tony, that's precise enough for what I need! :0) What I want to do for my own curiosity and a YT vid is to set up a side by side test comparing Ohm matching panels to a 240V heater element and just Pulsing a capacitor bank. I plan to set up 2 arrays of 4 panels each and have both on a 3600W element in a bucket of water. One will be direct and the other will have a pulsed capacitor bank, one sec on, one sec off going to the element. I can't really measure the power so I thought I'd just measure the temp in the buckets and see which one comes out ahead. I was also thinking it would be good to see how much more efficient your controller comes out ahead as well later on. It won't be scientific but I think it should be real world enough. I am expecting to see an improvement with the cap bank and it may be enough to make a simple pulsed setup like this worthwhile. I'm not sure how much Capacitance will be worth while. I think I have some 1000Uf's around so I'll use a couple of them and see how that works. From what I figure, there is no going to big, if they aren't fully charged they will dump all their power anyway so the only thing to avoid giving their rapid charge time, would be not to have them full and sitting there as that would cut into the power put into the element. Maybe a second on time would be too Long? .5 sec may be better going by that calculator? |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
Back to back buckets of water temperature rise is an excellent way to compare power. But I think you may get some pretty confusing results by trying to set up the switching points arbitrarily. At the very least an oscilloscope would allow you to manually set the repetition frequency, and percent duty cycle so that the voltage on the capacitor zigs and zags up and down between two appropriate switching point voltages. My little circuit does that automatically and it self adjusts for every switching cycle. Cheers, Tony. |
||||
CaptainBoing Guru Joined: 07/09/2016 Location: United KingdomPosts: 1985 |
Ayup Davo. This stuff is not my strong point but in your position I would go along the following track... As already pointed out, it depends on the current available (which depends on the internal resistance of your battery). This gives your input impedance from Ohm's law. My electrical principals college work taught me it would take C*R seconds to get to 63% of your voltage (assuming constants everywhere), another C*R second to "add" 63% of the remainder and so on... After 5CR seconds, you are close enough to consider the cap fully charged - theoretically it *never* is but rarely bothers practice. If the load is heavy and your impedance "high" you might have to cater for voltage collapse and recovery, but for back-of-a-fag-packet calculations you can ignore those unless they are large. Edited 2020-07-15 17:46 by CaptainBoing |
||||
| Davo99 Guru Joined: 03/06/2019 Location: AustraliaPosts: 1577 |
Could you elaborate please Tony? I was going to use an arduino on a mosfet for the switching. I can get sub second times with that on and off. I undserstand the limitations and realise this won't be anywhere near perfect because of the fixed switching points but do you see a way of optimising this within it's limitations? My - thoughts- were a cap presents a small load to the panels and shouldn't pull them down too much. Is this correct or will they load the panels like any other resistance? The second though was if the cap bank is big enough to hold all the power the panels can generate within the switching time, then fully charging them is irrelevant because they will dump the power they generate within that time anyhow. That comes back of course to the presumption, perhaps inaccurately observed with inadequate equipment, that the caps won't pull the panel voltage down too much. Also, do you think there is any "reactive power"/ Capacitance within the panels themselves? By this I mean when a load is dumped on them, is there any worth while power/ time before they are pulled too far down off their curve or is that instant? I'm wondering if I'm basing this idea on observations of meters that are far too slow to show the real situation? |
||||
| Davo99 Guru Joined: 03/06/2019 Location: AustraliaPosts: 1577 |
I take it in this case as there will be no battery, the parameter would be the the impedance of the caps? Is that likely to be more than a few ohms for regular high voltage caps? Spose I could meter them and see. The elements I am going to use are about 15 ohms so I would think the caps would be less than that? Yes, I expected that, much like a battery and read up on the same. I -figured- that the charge level was not that important as all the power would be dumped but maybe I over looked the idea of Tonys board which brings the caps up to the voltage point that the element works best at and THEN dumps the power? Think I'm starting to see the problems here but at the end of the day, do you learned Gentlemen see any advantage to a very Dumb, non variable setup like this over straight ohm matching? |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
Captain, You are quite right. Charging from a constant current source, or through a fixed resistance gives a very predictable result. Dave will be charging direct from solar panels, which have a very non linear voltage/current behavior, and which also varies greatly with the available solar energy. At dawn/dusk there will be almost nothing, a clear blue sky a lot. So its not possible to calculate any meaningful capacitor charge time. Dave, We are trying to optimally load the solar panels with a constant heating load, so that maximum power is transferred into a fixed heating element. If the solar panel loading is insufficient, the panels will be too lightly loaded, and the voltage across the panels will be high, and current and the power less than it otherwise could be. Optimum loading is when there is just enough loading to pull the panel voltage down to the specified maximum power voltage, which is listed on the ratings plate. Usually around 32v per panel. Excessive loading pulls the solar panel voltage right down. Although current will have increased slightly, the power transmitted into the heating load will be less than at the peak. Now the resistance of our heating element is fixed. And it may be spot on for one set of solar operating conditions, but completely wrong for different solar operating condition. We can come very close to having a perfect system that always transfers the maximum available energy into the heating element if we can hold the voltage across the solar panels within a very close range of the maximum power voltage. We can do that by charging up a capacitor which draws an almost constant charging current from the solar panel, and intermittently switching the heating load on and off. The heating element power pulses on and off, power drawn from the solar panels remains fairly continuous and constant. We allow the capacitor to charge to just above the maximum specified power voltage (on the ratings plate). When it reaches that maximum voltage, we switch the heating element load across the capacitor, and that discharges the capacitor and pulls the voltage on the capacitor down. The heating load sees solar panel current + capacitor discharge current which will be a lot higher than just the incoming solar panel capacitor charging current. When the voltage has fairly rapidly fallen to an allowable minimum, a bit below the maximum power voltage, we abruptly switch off the heating load. The capacitor voltage rises and falls in a zig zag manner between the two chosen voltages, set above and below the maximum power voltage. This works very well if you can detect the rising and falling capacitor voltage and use that to turn the heating element on or off directly. Its pretty simple to do in hardware, you can use a microprocessor if you want, but I cannot imagine why that would be needed or any better. Edited 2020-07-15 20:08 by Warpspeed Cheers, Tony. |
||||
 Print this page |
