Posted on 02/05/2024 9:14:32 PM PST by Paul R.
I am modestly(!) familiar with 4 types of DC voltage output power supplies that offer some degree of regulation to a load.
Those are:
Zener regulated supply: This has a resistor in series with the incoming / source unregulated DC voltage rail, and a zener diode to ground that fixes the output voltage at the zener voltage. Usually a capacitor to ground at the output is added to help stabilize the voltage, esp. if the load tends to vary a bit.
Zener-pnp transistor regulated supply: This adds a transistor and a couple more resistors for more current capability and better regulation. Darlington configurations are also available.
Linear regulator supply using, say, a LM317 IC, or any one of a number of fixed voltage regulator IC's. This can drive an output transistor(s) for higher current.
The above configurations are not particularly efficient, somewhat dependent on current drawn, etc.
PWM (Pulse Width Modulation) power supply. This is a type of "switching supply". These are very efficient and a quite physically small supply can cork out a few amps without generating much heat. Such supplies sometimes cause problems with RF radiation / interference that needs suppression.
MY application (a personal / hobbyist project, BTW, of no commercial intent or value) is to power a LED(s) load with a constant 13 volts, BUT, the number of LED's (actually 3-LED modules) in parallel may vary. (Long story.) So, I don't want to use a typical constant current LED supply. The source voltage is a 12v nominal(!) automobile battery system, which system can on occasion reach 14.5 or 14.6 volts with the IC engine running.
The LED modules appear to have minimal current limiting if any "onboard". Brightness without excessive heat is good at 12-13 volts. (Very good at 13 volts.) As one approaches 14 volts the current goes up quickly, and the modules get quite hot. Brightness begins to fail to increase comparably to current drawn and heat. (It may actually decrease if the module has time gets hot enough for that -- I've not tested this as I didn't want to blow up a module!)
The LED module consists of 3 LEDs and (I suspect) a resistor, all (I suspect) in series. The resistor provides modest, but not a lot, of current limiting. The vendor basically just tells buyers to hook up to a "12v" source in an automobile. But, some buyers report overheating and burnouts, even though most do not. Ergo, I'm looking to limit the voltage in this lighting circuit supply to ~13 volts.
Best I can tell, PWM supplies need to be supplied with a voltage source 2.5 to 3 volts higher than the output. Since the car's TYPICAL battery voltage is 14 volts or less (esp. if the engine is not running), if I try to set the LED supply at ~ 13 volts, I'm guessing the PWM supply's output will "drop out" with an input of 13-14 volts?
I believe the linear regulator needs a differential of around 1.5 volts?
The zener "regulator" output voltage will basically track the supply voltage as it falls below the zener voltage. Thus the LED's will dim but not go out until the voltage the LED module sees is well under 12 volts.
The Zener-transistor circuit will also "gradually fade out" as source supply voltage decreases below 13v (given a 12v zener diode -- I think???
Noted!
The one disadvantage (besides cost & size) I can think of for a buck-boost convertor is that if running off the battery only, I might kill the battery before the lights dim at all...
If the input voltage drops, the behavior is unpredictable but it will be nothing that you want. You have to give the regulator more volts than its output, that is immutable.
Is there some reason why you can’t run your LEDs from AC? Or switch to a half-wave rectifier ahead of the regulator? If your supply transformer simply does not give enough input volts, you can use a small transformer as a “bucking” transformer and boost the AC line input by 6 or 12 volts. At some point you may have to heatsink the 7812 because of how much power it has to get rid of.
If you don’t give your 7812 enough input volts, it will not cooperate in the most irritating ways imaginable.
Making the regulator out of discreet parts may work better, but the 7812 is made up of transistors and zeners anyway, so it might just turn out the same. I’d focus on getting the regulator input volts really high, aim for 18-20 volts and heatsink your regulator. Get it high, move to a 7815, then use some series diodes on the output to drop the overall supply voltage by .6 or .7 volts each.
If indeed you are only powering 3 LEDs at (figure) 20 mils each, you should be able to give the 7812 (in TO-220 format = 1 amp) a solid 20 volts and should not sweat.
Yes, DC-DC Buck Converters for automotive, heavy equipment, and aircraft systems. I don’t know how tight of a package you are trying to end up with or if room is an issue but did you see those I listed in #15? They are pretty cheap nowdays and other companies make similar with similar spec parameters.
We have been running two 150 watts each 38 volt 7 amp solar panels through just one of those Buck Converters extremely reliable so far. We set these panels up in pairs through one BC on it’s way to the charge controllers. We can input either 14 volts or 28 to the charge controller depending on if the system is a 12v or 24v system. And the heat sinks barely get warm in full charge conditions.
But the range capabilities of these and being adjustable is great. They can be used for just about anything you need them for and are extremely efficient.
Thanks for the reply - the modules each draw about 100 mA*, and anywhere from 1 to 8 might be switched on at any given time. Given that the current drawn increases rather quickly at and above a 14 volt input, I believe each module consists of 3 LEDs and a limiting resistor, all in series, something like the “12 volt” example here:
https://www.waveformlighting.com/pcb-designs/led-strip-light-schematic-and-voltage-information
*Exactly what LED is used I am not sure. It is apparently something a bit “less” than a 2835, as the 2835 is rated at a maximum 150 mA, not 100 mA. (At 150 mA these modules get awfully hot awfully quickly. But they seem fine, just a bit warm, @ 100 mA.)
More, in a bit...
Then I would go on eBay and look for what is called an “open frame” power supply. There are about fifty companies that make these, they are ubiquitous. Sola, Condor. Power-one, Lambda. Super reliable. I would prefer a linear supply. If you can handle that exposed wire format.
You should be able to find something for under twenty five dollars delivered, Possibly brand new in the box. 100x easier than building one.
Those are pretty cool, but, way more than what I need / want in terms of output capacity. And, I’m not sure about the constant current part: Since different loads will be presented, ie., I’ll switch in (”on”) anywhere from 0 to 8 LED modules, I don’t really want something that tries to push a constant current.
Then there’s also the bit I mentioned above about the lighting being its own indicator if the source (battery) voltage gets low. Conversely, here, if the battery voltage drops, the boost function will draw even more current from the battery, trying to maintain the converter’s output. Great way to really kill a battery.* The LEDS stay on nice and bright until suddenly you are plunged into darkness.
I could add something as simple as a relay circuit to disconnect the battery if battery voltage runs low, or, sound an alarm, but, this is getting much too involved...
I just want something that acts more or less like it wasn’t even there when the battery / source is at 13 volts or less, and drops the voltage to the load to 13 volts when the source (car charge system and battery) is greater than 13 volts. The zener regulator does that, but very inefficiently and requires big components to have much current capacity. The zener-transistor regulator MAY be ok as voltage drops, certainly with better current capacity than the zener regulator (which is really merely a voltage limiter), and the linear regulator I just don’t know.
*Different app, but I’ve discovered that a sharp decline at the end of a battery’s discharge cycle can be a serious disadvantage in many cases. A case in point is a smoke detector or a moisture detector (often used to tell you your sump pump has failed), usually powered off a 9v alkaline battery. Try to use a rechargeable battery and usually the battery is quickly “too low” and the low battery alarm sounds. (My “9 volt” [actually more like 8 volt] rechargeable lithium batteries work great with good “life” in many devices, but, not in smoke detectors.) Try to use the new rechargeable batteries with tiny boost converters in them to provide a stable / regulated 9 volt supply, and the battery voltage falls off a cliff toward the end of the discharge cycle. Instead of being away a couple days and coming back to a low battery alarm that may well have been chirping intermittently as designed since 2 minutes after you drove off, you come back to silence. Maybe that fire happens the next day, but the alarm never alerts you because the low battery alarm only sounded briefly while you were gone, and then IT died too. So poor resident dies too, clueless until too late. (Sounds like Biden voters, eh?)
That’s why smoke detector instructions always say to only use alkaline batteries — they have a “soft” gradual “death”.
Eh, took me a bit longer than “a bit”. (eye roll)
I did just post #27, paragraph 4 applies in particular. :-)
Indeed, there are cheaper linear regulated supplies on Amazon, eBay, etc., that with good heat sinking of the regulator chip and output transistor could easily crank out a couple amps, especially given that the voltage drop when limiting the output will be modest. BUT... same problem — if the input is 13 volts, what does the output do?
It occurs to me that SOMEWHERE around here I should still have my old audio op amp prototyping breadboard which had either a 7815 and a 7915 in the power supply section, or possibly a 7818 and a 7918, as I sometimes worked with +/- 18v supply rails and NE5532’s and NE5534’s. I could muck around with the 7815, for example, to see how the output behaves as the input decreases. If below a supply voltage of 18 volts or so, the output pretty much follows the input, no problem. I don’t really need true regulation, just voltage limiting. But if the regulator turns off when the input falls under 17v or so, that’s no good.
Then I would go on the assumption a 7812 would behave much the same, just at 3 volts less all around.
I still think the zener-transistor circuit MAY do what I want, so, I have parts ordered and will see what happens... If it doesn’t work I’m out ~$15 and will have several extra 13v zeners and TIP31C’s to add to the parts bins. Plus I got up to $35 for my Amazon order, finally! :-)
As per lengthy discussion, my source (would feed whatever circuit or module I use) is a variable DC voltage. There is no AC input to be had...
A fellow over at All About Circuits is assisting me. He’s come up with a couple pretty interesting, novel (to me) circuits. Essentially, as the input (source) voltage ramps up, the output voltage ramps up too, but levels off above 13v. This is not quite analogous to audio amplifier clipping, but could be crudely thought of in that way.
The parts count is a little higher than I was hoping, but still cheaper than if I had small regulators on each LED module.
I may still purchase a small PWM buck regulator just to see how it behaves as it’s input voltage sags. Or if it requires RF suppression. They can be had for well under $5 on eBay or Amazon.
Thanks!
See my post #31. :-)
Thanks!
Sorry Paul. I meant to catch up with you on that. Got hammered with real life stuff here. I understand your smaller need for the application you are after. I just wanted to share that they do make these of all different ratings that are smaller but adjustable as you like.
But I don’t see how any regulation of any kind can be used without the LED circuit being switched somehow “before” your custom voltage control to prevent continuous draw such as a momentary contact or on/off switch? Or a small relay to close that LED circuit when wanted such as only when an ignition switch accessory output is on?
It is the age old dilemma of the dash clock eventually running down the battery months later because it is not switched.
Sounds like you are making some sort of meter with the LEDs so they light at different voltages? Something I found with LEDS is no matter how they are rated they are pretty forgiving with voltage inputs. You can have 10 in a series rated for a load of 15 volts and short around two of these in the series circuit and they will still light normally without being overloaded because there are only 8 in the circuit.
Same in reverse, you can put only 10 volts instead of 15 volts to a circuit of 10 and they will still light up same as if they were getting the full 15. The rating range where they will work is kind of broad and forging. So I don’t know how much trying to regulate the input to them from 18 or even 20 volts down to 15 volts will make much of a difference to them. A series circuit of LEDs rated for 15 volts will probably also handle 18 without any problem.
I’m going to go back and read your OP to make sure I am not sending myself down the wrong track with what you are after. :)
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.