Howdy. I'm looking for an electronic circuit design that accomplishes the following: using a DC voltage source first charge a capacitor to a certain voltage level and then (automatically) discharge that capacitor through another circuit, then start charging the capacitor again so the process keeps repeating. So basically I need a circuit that turns a constant DC voltage source into a series of current impulses (because I want both the cyclical nature of the impulse and also a higher current in the impulses than I would be able to get from just the DC power supply I'm using, which will probably be a not-too-large battery). One more thing, I would like this circuit to run at a reasonable high frequency (maybe 50-200 Hz) so I'd prefer not to have any mechanical components such as relays in the circuit since it seems like they could wear out easily under those conditions.
So, any help I can get with this would be appreciated. I'm actually an electrical engineer, but in control theory, so while I'll be able to understand a circuit diagram I haven't done any real circuit work in ages. Thanks.
What voltage? What capacitance? What load? (Is it resistive, inductive, what? What sort of peak current are you looking for? How long do you expect the output pulse to be? How much energy per pulse?) Are you thinking to essentially just have the circuit connect the capacitor to the load and let it discharge, then disconnect? How did you have in mind charging the capacitor? (That is, would you just put a resistor between the voltage source/battery and the capacitor, or do you want to do something more complicated? -- A resistor will dissipate half the energy provided by the voltage source, if the capacitor is allowed to discharge to zero volts.) The simple form of what you describe is just a relaxation oscillator, and there are some easy ways to do it, but the easy ways usually have some limitation or other, and their usefulness would depend on just what you're trying to accomplish.
You are pretty much describing a plain old switchmode power controller.
--
Many thanks,
Don Lancaster voice phone: (928)428-4073
Synergetics 3860 West First Street Box 809 Thatcher, AZ 85552
rss: http://www.tinaja.com/whtnu.xml email: don@tinaja.com
Please visit my GURU\'s LAIR web site at http://www.tinaja.com
When the timer goes low the PNP charges the cap and the NPN is cutoff, preventing current through the load. When the timer goes high, the PNP is cutoff, disconnecting the cap from the battery, and the NPN allows the cap to discharge through the load.
The load is going to be mainly inductive (an electromagnet actually). Not sure about my voltage source yet. Probably be in the 5-10V range. Figured I'd adjust the capcacitance to get a frequency-pulse power combination I could live with.
Basically, I'm just looking for a general circuit design that I can play with to get what I want after I get some idea of what I'll be using for voltage source and load.
Yeah, I think so.
Well, if I wind up using a battery with a small maximum current, I might just connect it straight to the capacitor. If there's a better way to do that, I'd be open to suggestion.
--
Many thanks,
Don Lancaster voice phone: (928)428-4073
Synergetics 3860 West First Street Box 809 Thatcher, AZ 85552
rss: http://www.tinaja.com/whtnu.xml email: don@tinaja.com
Please visit my GURU\'s LAIR web site at http://www.tinaja.com
ha. a 555 is good choice if all you need is a simply timer. why waste time and effort with such a sensitive device for such a simple job. Now, i suppose if one wanted to get fancy with in making a motor controller? then a pic would be a good choice how ever.. have fun.
--
"I\'m never wrong, once i thought i was, but was mistaken"
Real Programmers Do things like this.
http://webpages.charter.net/jamie_5
I don't think a 555 or PIC is going to do me any good.
What I want is to be able to take a relatively small constant current (i.e., from a battery) and generate large amplitude/short duration current pulses. That is to say I want to build up some energy for a period of time, release it all in a burst, and repeat that process ad infinitum.
This sounds like a typical unijunction transistor application. GE semiconductors used to h ave lots of application notes on these, typically for delivering a stiff turn-on current to low sensitivity SCRs.
On a sunny day (Mon, 20 Nov 2006 11:30:47 GMT) it happened Jan Panteltje wrote in :
And I forgot to mention I in R1 should be below the hold current of the thyristor. Else you will have to use a transistor of sorts and not a thyristor! Oops.
--
You might want to download Linear Tech\'s free simulator and that way
you can try any circuits anyone might propose without having to
actually build anything, at first.
Their simulator page doesn\'t seem to be available right now, but
when it comes back up you can get to it easily from:
http://www.linear.com/designtools/index.jsp
OK, I've looked over the other suggestions, and there seem to be some good ideas. I'd avoid a PIC: the 555 is a lot better at handling a range of voltages gracefully, especially ones above 5V. And, it's easy to use. It can generate narrow pulses easily. I'd recommend the CMOS version (with due care about zapping the CMOS inputs). You didn't say what peak current you want, but at such low voltage, and running from a battery, certainly a power MOSFET can be found to do what you want.
Because you want to have fairly fast recharging of the capacitor and you're planning a small (?) battery for power, I'd recommend you consider a couple different things: first, do like one poster suggested and disconnect the recharge circuit when you're conducting current in the load. Second, provide some much larger capacitor across the battery, so you don't draw relatively high current pulses from the battery. A capacitor likely will be much better at delivering the AC part of the recharge current.
And finally, if you want the best battery life, by all means use a resonant recharge circuit. If you connect the battery "directly" to the capacitor, you will lose the same power in the battery's internal resistance and the circuit resistance as you would if you put in an external series resistor. The net resistance changes the time it takes to recharge, but not the energy lost in heating the resistance. A typical resonant charging circuit would be: battery, with large energy storage capacitor, C2, in parallel. Negative side to common/ground. Positive side to a P-channel power mosfet source. Pull the gate of the power mosfet low to turn it on when the load is disconnected. Power mosfet drain to the anode of a diode (Schottky nice, to keep the voltage drop low). Cathode of the diode to an inductor, L1. Other end of the inductor to the capacitor you're recharging, C1. Pick the inductance of L1 such that the resonant frequency of C1 and L1 has a period LESS THAN two time the length of time you have to recharge C1. For example, if C1=1000uF and you want a 200Hz rep rate with a discharge time of 1 millisecond, that leaves 4 milliseconds to recharge. You need to pick L1 to give a resonant frequency with C1 of at least 125Hz, which is 8 milliseconds for a full cycle. The recharge takes place in half a cycle. So L1 should be less than 1.6 millihenries. Since a 1000uF capacitor will have a pretty loose tolerance, I'd suggest 1.0 millihenries. NOTE: the capacitor will charge to a voltage such that the average of the final voltage and the initial voltage (just before you started the charging) is equal to the battery voltage less a diode drop -- not accounting for a little additional drop in resistances. So if you have a 9V battery, C1 could charge to nearly 18V with this circuit. The peak current in the inductor (important that the inductor can handle this current without saturating) will be such that Ipeak^2*L1=0.25*Vpeak^2*C1. (That overstates the current, if you don't discharge C1 to zero each cycle, but it's a safe limit.) In this case, note that it's a pretty high current, about 9 amps! Why? Because 1000uF charged to 18 volts represents (18^2)*0.001/2 joules = .162 joules, and at 200 discharges per second, that's over 32 watts. From a 9V battery, that's an average of 3.6 amps. -- In other words, be sure to set you expectations correctly with regard to how much power it will take to run your little experiment, before you start. And...C2>>C1; I'd say C2 >= 10*C1 Be sure to size the parts right for the job.
The simplest and most rugged might be a thyristor, they're swell at switching high current and voltage. You can trigger it with a diac. Two components! Hard to beat that.
The 555 does all of the timing you need, and the output can deliver (current) pulses as you (not so accurately) describe; the current available is specified on the data sheet; the output is a voltage and can be converted to a (reasoable) current via a (erasonable value) resistor in series to the unspecified load. Instead of dealing withvague generalities concerning the requirement, be eXplicit and state what you really need to drive (LED? inductor? relay? spitzensparken?) as well as eXplicit parameters (2uSec pulse every millisecond? 100nSec pulse at 300 per millisecond? fratzensprats on the bumblelator?).
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.