Buck converter problem-Solved

John, can you point to the button or circle it?  I buy the cheap ones without connectors and readout.  Use them alot.

I'm with TED too.  Got the cheap ones and all work fine.  Please point out the button so all can see.

Hi guys,

I had problems with the low cost boards where there was a great voltage fluctuation, lights would flash from dim to bright rapidly. The seller requested a video which I sent. When he saw what was happening he credited me for the units that were defective.

Ray

I put up another picture. The buttons are directly under where it says in and out for the lights. These were only $2 each and 2$ shipping for 4, so I would highly recommend these over the "inexpensive" ones.

John H posted:

I put up another picture. The buttons are directly under where it says in and out for the lights. These were only $2 each and 2$ shipping for 4, so I would highly recommend these over the "inexpensive" ones.

You’re probably right, but my last batch was 10 @ $7.71 shipped.  Still have a bunch, and, each one handles a bunch of buildings/leds.  I really like adjusting lighting to suit my taste. My tendency is lower to reduce glare and brightness.  Just more realistic IMO. 

Definitely a good price. They do work well for lighting, no matter which one.

For less than $1 (free shipping) you can get a standalone DC meter that you can attach with just 2 wires. 

To my knowledge you still can't buy a AC-input, DC-output buck-regulator that has a built-in voltmeter.  This would be for hooking up to, say, 14-16V AC accessory voltage to generate 4.5V DC for Menards, Lemax, Miller, etc.  So for now it's a $1 add-on.

Stan,

 I have two of these. 18 volts in from a KW, adjustable dc out. I’ll have to check my purchase history for more info. Ed.- Just added picture

The downside of that AC-to-DC converter module is it uses the older "linear" stepdown technology (venerable LM317 IC regulator) vs. newer buck/switching technology (e.g., LM2596 IC regulator).  The eBay listing suggests it can deliver 2 Amps.  Well, I'd be a little reluctant to power up too many 4.5V DC buildings at 2 Amps current from Accessory AC voltage.

Starting from 14V AC and converting down to 4.5V DC at 2 Amps (9 Watts outputs) with an LM317 IC, you must dissipate 9.5V x 2 Amps or an eye-popping (finger-burning) 19 Watts of heat in that heat-sink.    It doesn't look like a 19 Watt heatsink! 

A "switching" regulator such as used in the DC-DC modules shown earlier probably dissipates/wastes around 2 Watts of heat when deliver 9 Watts of output.  But you are correct that I was not precise in my claim that you can't find an AC-to-DC regulator module with a meter.

One nice feature of your LM317 module is the finger-adjustable knob of the voltage control.  Most DC-to-DC and AC-to-DC voltage regulator modules use the tiny screwdriver adjustment potentiometer - typically the blue component.  I continue to offer a gold star to anyone who can find a source of these knobs in small quantity.  These apparently snap or press-fit onto the potentiometers.  Yes, I realize most guys set a module once to 4.5V DC (or whatever) and never touch it again.  But for messing around on the bench such a knob would be most handy and I figure it's something that should be 5 cents each or whatever.  Yet, I cannot find these knobs by themselves in small quantity - rather you have to buy hundreds or thousands of a regulator module from the likes of Alibaba.com.

So I'm dumb as a stump when it comes to these things.  But if you wanted to use track power to do some home brew motor controller... which say is a tiny computer board.  Don't you think you need a lot of amps at the ready if the dc motor pull it (say near stall or motor start)?

Given that, I see big solutions like this below appeal to me because they show big numbers going in/out -- which seems to me means they have gobs of power at the ready if needed.   Now this one is DC in... but could DC in be made from AC with an appropriately side rectifier in front of the inputs?

https://www.amazon.com/dp/B071...9P&pd_rd_w=bC3zp

I found an U.S. eBay seller who sells trimmers with the knurled knob in small quantity...but with domestic USPS postage, it's over $2 for the component - i.e., more than the cost of a typical regulator module!

But you may be on to something with the knurled brass standoff with screw.  It could be the knob is integral to the threaded adjustment screw.  In other words it's not a separate/removable knob that press-fits to the tiny exposed end of the slotted adjustment screw.  Of course the problem is the blue trimmer part is molded/sealed so would be a challenge to open it to change out the screw to one with an integral knurled knob...obviously easier to unsolder the entire 3-terminal blue trimmer and replace with the $2 part.

The more I think about it, it's hard to imagine how to manufacture a solid brass part that press-fits over the exposed slotted screw.  Even trimmer screwdrivers are assembled from 2 parts.  

Photo shows a larger multi-turn trimmer and its knurled knob is permanently fastened to long adjustment screw.

Severn posted:

...Now this one is DC in... but could DC in be made from AC with an appropriately side rectifier in front of the inputs?

A (bridge) rectifier itself only provides pulsed-DC from AC-track-voltage.  In other words, the DC input voltage to the module would collapse to 0 volts twice for each 60 Hz cycle.  Hence you also need capacitor(s) to maintain a DC input voltage to the regulator electronics during these zero-crossings.  There's some geeky math involved but if Amps of current are flowing, the capacitor can get physically large - perhaps larger than the size of the module itself!  The input capacitors shown above would not be large enough for an AC-to-DC application of more than, say, 1 Amp - your mileage may vary.  It's not that capacitors are expensive -maybe $2-3 for the module shown - point is there is some additional homework to be done.

To your point about driving motors.  Yes, if I were using such a module for DC motor control I'd want it to handle a sustained motor stall current.

I can desolder those and solder in new much large capacitors -- this is the suggestion?  This would maybe easier than trying to add a battery in to the picture to even things out -- an off hand thought.

Severn posted:

I can desolder those and solder in new much large capacitors -- this is the suggestion?

Much easier to add the additional capacitor(s) to the DC-input terminals - leaving the existing capacitors in place.  The new capacitor(s) will simply add to whatever is already on the board.  I'm just answering your original question at face-value.  

If you're asking my suggestion as to what I would do, I'd want to know exactly what you're trying to do with actual voltage ranges, actual motor power requirements, how you adjust/set speed, etc.  For example you have that tiny screwdriver adjustment on the DC-DC regulator module that sets the output voltage.  Does this mean you are going to set the voltage one-time and never touch it again?

Virtually all DC-motor control today - including within our O-gauge engines - is done using a so-called PWM or pulse-width-modulation electronics.  It is a similar digital-like technology as used in the buck DC-DC regulator module.  It would be effectively redundant to use the buck module to generate a constant-voltage DC ... and then follow that with a PWM module to further alter the DC to control motor speed.  I'm not saying the following module would work for your specific application but gives you an idea of what a PWM controller module looks like.  This one happens to operate up to 5 Amps over a reasonably DC large voltage range.   Very economical.  PWM controllers come in all shapes and sizes.

If powered from AC track voltage, you'd still need the (bridge) rectifier to convert AC to DC.  You'd still need large capacitor(s) to smooth out the DC.  

Separately, is this on a piece of rolling-stock such that you need to power this from AC track voltage?   If this is a layout accessory I'd suggest using a DC-output wall-wart, surplus computer-power supply, etc. to provide smooth DC voltage (12V, 18V, whatever).  5 Amp, 10 Amp, 20 Amp, etc. DC-output supplies cost a fraction of similar AC-output supplies (like train transformers).  You could then use this DC-output supply to power the original DC buck converter or a PWM motor speed controller module...no rectifier or additional capacitors.

I have a raspberry pi b with one of these -- https://www.pololu.com/product/2756

... Connected to a DC power supply and 2 mabuchi motors out of an old engine.  I can make they them go reasonably well and thought I'd put them on the track less a long wire (on the mounting plate of the old engine less the outer bits to make it all fit if necessary), and watch it go around a few times.  (i'll provide a simplistic control interface)

Track power is ultimately appealing but a battery could work too -- I haven't looked into either deeply except I thought "What if I search ogrforum on track power...."

And this thread looked sort of close-ish to the track power solution.

So you already have a PWM motor drive module.  You don't need the DC buck regulator module.  Feed the output of a bridge-rectifier into capacitor(s) into your PWM module.  A PWM module does not need smooth, tightly regulated DC...it is going to chop this voltage before applying it to the motor!

There is more homework to be done.  Selecting a suitable bridge and capacitor is a function of the range of voltages you will encounter...which depends on your application.  For example, it appears the Polulu module operates up to 28V DC input...which might be a tad under-sized if your AC track voltage is more than 20V AC (or so).

Also, if using a RPi or Arduino or the like, you need a steady source of DC voltage (e.g., 5V) that must NOT be interrupted.  In rolling stock applications, you must protect against fraction-of-a-sec track voltage dropouts when going over a turnout or piece of dirty track.  A brief voltage dropout can be disastrous to a microcontroller as it will likely "reset" the software.  On the other hand the motor control module should be able to tolerate brief voltage interruptions. Perhaps this is all old-hat so I don't want to drone on. 

You might want to start a new thread about this as I don't think a DC-DC buck regulator module is needed in your application.

Ok but this is just my little fun project and since I don't have anything solid to give out or show at the moment, I hadn't considered further discussion.  BUT I roughly think I understand and will try it.