 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 : Ultra Capacitors "Boost-Caps"
  Page 3 of 4   |
|||||
| Author | Message | ||||
yahoo2 Guru  Joined: 05/04/2011 Location: AustraliaPosts: 1166 |
The Maxwell paper I quoted was published in April 2009 and was an update from 2007. I doubt they would have imagined the variety of uses and the fact that individuals are buying cap banks now. My guess is that it was aimed at larger companies that are changing out expensive custom made batteries very frequently, as a lower cost option. Phill has hit the nail on the head, the reduction in capacity by lowering the voltage is just too much of a sacrifice. I am glad it's you guys that are experimenting, capacitors seem to commit hare-kari around me. I got a bit exited about LiPo batteries 12 months ago and found a similar thing, most of the positive info I found on the internet was outdated stuff that was rehashed. the forums on the other hand seemed to be full of tales of battery management modules not doing what they are supposed to do, and banks getting out of sync. I'm confused, no wait... maybe I'm not... |
||||
MrDelanco Senior Member  Joined: 12/11/2011 Location: United StatesPosts: 101 |
Hello everyone; This is an update ;I found the resistance in the system coming from the store bought controllers causing the build-up of current up the line in the system overheating the power resistor's. They are not designed to work with a positive grounded setup. I found the cause of the build-up when voltage was increased and the current increased on the positive side into the charge controllers was caused by the store bought charge controllers not allowing the current to pass without restrictions. The balance circuit stabilized the negative side of the system to a constant 2.5 voltage. On the scope I do see the tunneling start and the readings are normal the reverse voltage stays at 2.5 volts even during the increase in voltage on the pos side and the operating current remains stable. I installed some used sealed cell electric lawnmower batteries into the line after the ultra-capacitor's before the charge controllers too act as a buffer for the controllers. When the current increases on the positive side the batteries absorb the increase so the controllers don't over regulate the flow. They now have a stable constant current to work with. The lawnmower batteries were a quick-fix in the circuit. The system is working as it should. When I get some extra time I will design and build a new controller to replace the store bought ones and remove the shunt batteries. Cheers Robert. MrDelanco:Project Videos It is not only too know what it does but to understand how it does what it does. |
||||
fillm Guru Joined: 10/02/2007 Location: AustraliaPosts: 730 |
Robert , which balancers do you have, as am a bit confused , it now seems like you have a active balance system that is store bought and not as you claimed as quoted below in your first post that showed the passive balance LED / Diode diagram that you built and modified to suit your application. |
||||
Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
Bob, the problem with using "zener" diodes (and the fact that Robert does not seem to grasp) is that the lowest available zener voltage is 2.2V. Now, when you put a LED in series with it, and the LED itself requires about 2V across it to light up, its never going to light when connected across a 2.5V super capacitor. Just an ordinary silicon diode has a 0.5V drop across it so combined with a LED it would do as expected. I just wish the word "zener" would disappear from this topic. Years ago ( about 30  ) I built a digital clock that used a circle of 60 LED's for the minutes and 12 LED's for the hours. I quickly found out that all LED's are not equal and there was a significant brightness difference even when they had lo tolerance series resistors. I ended up matching LED's for brightness individually against a chosen sample and required about a hundred LED's for a neat brightness match. LED's were expensive back then  .
Anyway, to build a LED equalisation/ charge indicator its worthwhile to match the LED's and series silicon diodes on a breadboard first, something I will definitely do when I build one. I might even incorporate it in that DMM idea I mentioned above. Klaus |
||||
| VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539 |
Hi Tinker I have never used them for anything this low in voltage before, so I see what you mean now. If I decide to go ahead with these caps I was thinking on a balancer like in my model battery lipo balancer, for final top up it sends a controlled voltage to each cell for 30 minutes at reduced current. All the best Bob Foolin Around |
||||
| MrDelanco Senior Member Joined: 12/11/2011 Location: United StatesPosts: 101 |
Phill I am talking about the charge controllers after the ultra caps that regulate the input into the 24 volt battery bank not the charge balance circuit. Regards Robert. MrDelanco:Project Videos It is not only too know what it does but to understand how it does what it does. |
||||
| MrDelanco Senior Member Joined: 12/11/2011 Location: United StatesPosts: 101 |
|
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
How about a TL431 ? 
This is a proper shunt voltage regulator with an accurate 2.5v internal voltage reference. No need for either R1 or R2, just connect Vref direct to Vi and it will regulate Vi to 2.5 volts. A series resistor is required to limit the current through the TL431 to below it's maximum safe rating of 100mA when it turns hard on. Maybe 27 ohms 1 watt should do it. Only two components wired directly across the cap, enclosed inside a bit of heat shrink tube is about as simple as it gets. Cheers, Tony. |
||||
| sjh7132 Newbie  Joined: 27/12/2011 Location: United StatesPosts: 13 |
I think there are a few problems with this circuit: 1) The zeners are in the wrong direction. The zener break down voltage is for the reverse direction. In the forward direction their voltage is about 0.7 like any other diode. 2) All the discharge current goes through LEDs? I guess this is okay as long as you just trickle charge these caps, but if you are going to dump serious power into them I'd think the LEDs would be quickly toasted. (or are these the huge 100W LEDs?) Steve |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
My thoughts exactly. For any serious application of Ultracaps, where the charging current is high to very high, it will take more than bleeding a very few milliamps to hold down the voltage across the weakest individual Ultracap in a series chain. A TL431 by itself will sink 100mA, but if you need to sink more, a TL431 and a big arsed transistor with some wire wound resistors should be up to it. Cheers, Tony. |
||||
| Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
A timely posting as just today my belated Christmas present arrived  , a box containing 12 used ultracaps of 3000F @ 2.7V rating. I could only afford used ones but they appear to be in good condition and arrived promptly from the USA.
I was looking at the application sheets of the TL431A and saw a circuit for a high current shunt regulator using a PNP transistor. I will try that with a 1A rated transistor and a diode/LED indicator as well in the TL431 loop, sinking an additional 30 mA or so. I'll set it up so that the shunt turns on when the capacitor reaches 2.65V on its terminals. This will only happen when the batteries are completely full with my 11 caps in series connection on a 24V nominal battery bank. It would not be difficult to replace the transistor with a higher rated (and more expensive) type should it be required. I'll report here how that goes once its all fixed up, meanwhile I'll have an excuse to draw up a 11 regulator PCB and make the board - good fun .Klaus |
||||
| VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539 |
Hi Tinker Sounds like a good solution,I have been thinking along these lines and will order a set of caps as finances allow, will you have a bleed down circuit as well? I follow with interest. All the best Bob Foolin Around |
||||
| Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
Hi Bob, I'm not quite sure what a "bleed down" circuit ought to do at a capacitor bank that's permanently connected in parallel with my battery bank - please elaborate. Today I have been looking into active balancing systems, the one Maxwell Capacitors have in their notes does only 3-400mA per cell. Its a bit hard to guess just how far these caps go out of balance, time will tell. I'll make provision for a darlington bypass transistor shunting off excess charge up to 4A, might be an overkill though .
Klaus |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
If your battery is in deep discharge, maybe near the end point of ten volts (for a 12v system) and you suddenly start very high current charging, the voltage might rise to fourteen volts pretty quickly. Automotive systems do this all the time after engine cranking and starting. What actually happens with Ultracaps fitted may be much less dramatic, but I can still see the possibility of some over voltage under particularly unfavorable conditions. I really don't know how much current needs to be bled off each Ultracap to hold down voltage, but I doubt a single LED would be quite up to the task. Cheers, Tony. |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
Another totally different "high tech" way to solve the unequal voltage problem, without throwing away any power, would be with an active charge equalization system. This would transfer power from the highest voltage capacitors to the lowest voltage capacitors until all capacitors had exactly equal voltages. The idea here is you connect four transistors (or Mosfets) in a bridge type power output stage across each capacitor, to drive a small 1:1 pulse transformer. This can be run at a fairly high switching frequency to generate a five volt peak to peak square wave output on the secondary of the pulse transformer. The clever thing about this, is that it will also work backwards as an active rectifier. If you feed a five volt square wave into the transformer, and synchronously switch the four transistors (or Mosfets) it becomes a very efficient active rectifier, driving dc current back into the capacitor. If you have (say) ten Ultracaps, you build ten bridge power switches all optocoupled from a single master oscillator, so they all switch synchronously exactly in phase. The ten isolated pulse transformer secondary windings are all connected to a common ac bus rail. Power will flow in whatever direction it needs to flow until all the capacitors end up having exactly equal voltages, all essentially with zero power loss. Small pulse transformers rated at an amp or two are commercially available for driving the gates of large SCRs, and should be ideal. This active charge equalization idea is nothing new, it has been done before and definitely works. Cheers, Tony. |
||||
| MrDelanco Senior Member Joined: 12/11/2011 Location: United StatesPosts: 101 |
Hello Happy holidays too all; Tony your concept is a good idea, post a drawing for others , I have my system working well but I like your idea as a much safer approach for others to use and it will be less costly. The transformer can be a recycled unit and the transistors or mosfet's are easily obtained. Good concept Regards Robert. MrDelanco:Project Videos It is not only too know what it does but to understand how it does what it does. |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
O/k Robert, I will rise to the challenge. Here are some basic ideas for the experimental design of a synchronous bi-directional chopper/rectifier connected to each Ultracap. There are two P channel, and two N channel bridge connected Mosfets, all will be initially held off, with zero gate voltage via the two 10K gate biasing resistors, and the dc paths through the two pulse transformer windings. The circuit will draw essentially zero current from the Ultracap with no drive signal applied. The common driver bus can then be energised with a high frequency square wave signal of 50/50% duty cycle, of a frequency that best suits the characteristics of the particular pulse transformers used. The amplitude needs to be sufficient to drive the Mosfet gates well into conduction, maybe 20 volts peak to peak (+/- 10v at each gate). One feature of this circuit is that no dead time is required. No cross conduction can ever occur where the upper N Mosfet, and the lower P Mosfet can conduct at the same time. That is because the gates are tied directly together, and the common gate signal to both upper and lower Mosfet on each side will be either positive or negative, so both Mosfets can never be on at the same time !! The chopper/rectifier is isolated from the common power bus by a second pulse transformer, so each Ultracap is completely dc isolated from all the others. It is not necessary to match the Rdson of the N and P Mosfets, because on each half cycle an N and a P will be in series, so each half cycle will be the same, even if the Mosfets have very different 'on' resistance. Power flow is bi-directional and will be out of the highest voltage Ultracaps into the lowest voltage Ultracaps via the common power bus, until all the Ultracap voltages equalise. The stronger Ultracaps aid the weaker ones, with essentially no power loss. The common drive signal could be switched on and off as required, as part of a further power saving strategy, or it could be run all the time. Power could conceivably be continuously drawn from the common power bus with this system, to run some external control electronics maybe..... It could even power it's own bus driver oscillator that way, but it would need a small dry cell startup battery, but once running, the startup battery would no longer be loaded.  Cheers, Tony. |
||||
| VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539 |
Hi Tony A very interesting way of doing it, just one question, what voltage are you holding the caps at, this may sound a silly question, but I am thinking of the caps as an extra energy sink when there are peaks of solar activity to absorb some extra energy rather than just dumping it, also with the windmill to absorb those extra short term wind events without dumping as well. All the best Bob Foolin Around |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
This circuit just keeps all the capacitors at the same voltage, whatever that is. The Ultracaps all go up and down in voltage together, from zero up to whatever total voltage is applied to the series string. This is essentially a low power circuit just to constantly balance the Ultracap voltages during charge and discharge. It is not intended as a load dump or total voltage limiting system. Cheers, Tony. |
||||
herbnz Senior Member Joined: 18/02/2007 Location: New ZealandPosts: 258 |
Hi I have read through this thread and gone off to the various links but still can not get why any application mentioned here needs these ultracaps. They may have less leakage and lower internal resistance but a standard electrolytic is not far behind. looking at prices I see a standard 8000ufd 100volt is $16. By the time you series up Ulta caps you are getting only 300ufd 25 volt (approx)I dont know cost. I have been over the years playing with my windmill solar and hydro feeding into a commercial MPPT (tristar solar ) I have used an old cap from my junk box 5000ufds 400volt to take the windmill output before feeding on to the controller. I think this came out of a inverter welder. Herb |
||||
| Page 3 of 4 |
|||||
 Print this page |
