too many volts? fried buck boost

24 volts is marginal,especially if using a PW type transformer. 18 volts peaks at 25 with a capacitor. 20 volts from a ZW peaks at 28.  To get peak multiply pulsed voltage by 1.41.   The buck I use has a capacitor in it.Yours does not seem to and needs filtered input.   Some modern transformers with altered sine waves can peak at even more. Not sure what type wave the Z1000 puts out.

This all assumes it is wired correctly and the capacitor or leads are not backwards or something.

Here is the type I used,already has a bridge and cap

LINK

here is a post on the buck converter used for accessories

LINK2

If you want to use the 24 volt board, I would put a string of voltage dropping diodes in front of it. 1n4001 or any 1 amp ones should work.

Link3

Dale H

It said DC to DC is that one good with a AC input?  G

I used a link a guy provided for those boards. I thought that 24v max sounded OK. I was wrong, again!

Thank you!

You need a bridge and capacitor. The ones I linked have that built in so they are AC or DC to DC.

dale h

Thank you!

I will use a different board. I do have another one that I ordered for my G scale. It is much larger but says it can take 40v input.

 I'm putting in my own bridge and cap. I should have got them with those installed! 

Can you post a photo of the toasted board?

I don't see a transistor on that 99 cent DC-DC board from the eBay photo so I'm assuming it's the IC regulator chip you're referring to.  In which case your assessment of excessive input voltage makes sense as it's the generally the IC regulator chip's (RT8272) input voltage limit that determines the voltage handling capability of the module.

In my opinion, if you're going to make a change, use a LM2596 DC-DC module.  These modules have higher input voltage capability (35V DC or more).  While you may plan to only use this car with your Z1000's 18VAC max output, why not give it the ability to handle worst-case O-gauge transformer voltage.

Yes, the LM2596 DC-DC modules are bigger.  But they are 99 cents with free shipping on eBay for qty 1.  They have better set-ability of the output voltage using a multi-turn potentiometer.  The ones I use seem to change about 1 Volt per turn...while the smaller DC-DC modules seem to span the full range or effectively more than 10V per turn which can be tricky.

I suppose you could also use a LM2596 AC-DC module (with the bridge rectifier) but it sounds like you already have the DCS inductor and bridge wired up.  Plus those AC-DC modules run about $3-4 in small qty on eBay.

I don't have the exact DC-DC module you have but here's a side-by-side of a typical LM2596 module vs. what I think is a similar small DC-DC module...along with the added components to make these work on O-gauge AC: (a) 22uH inductor for DCS compatability, (b) bridge rectifier, (c) increased input capacitance for flicker reduction.

BTW, these modules are not "buck boost" but only "buck".  In this context, buck regulators only reduce voltage.  A buck boost regulator can both reduce or raise voltage.  A buck boost would be useful for operating in either conventional or command since the incoming track voltage can be below or above target 12V (or so) needed to drive your LED strips.

Thanks Stan. Yes, I should have said it was the regulator chip, but I didn't know that!

The picture you show of the other board, are the ones I am using in my larger G scale cars. I will now use only them in all the conversions.

 Mine say 40 volt max input.

Next issue will be that I used your diagram and bought small 220u caps. I'm not sure if they'll be big enough to stop flickering?

 I'm using 4700u caps outside in my G scale! They fit in the tanks under the cars. The lights stay on for a few seconds when I kill the power!

that board smelled so bad I had to get rid of it!

The one side of the reg chip had a few of it's terminals burned clean off.

http://www.ebay.com/itm/DC-DC-...;hash=item1a028a0a7c

220uf will prevent flicker, larger caps obviously hold the lights up longer.  As you observe, the 4700uf keeps them on for seconds, so for killing flicker it's probably overkill.

When I was testing the circuit in my G scale car, it seemed that the voltage at the board's output was much higher than what was going into the car. I assumed that the board was providing a boost. Reading one of your posts on another subject, I learned that the cap raises the voltage. I'd have to re-test knowing that to see what was going on exactly.

The capacitor will charge to the peak voltage of the input waveform, which is 1.4 times the RMS value of the input AC.  I try to use components rated at 35V or more for O-gauge model train applications.

Yes, I'd re-check your measurements - specifically the DC voltages measured right on the board at the + and - input and output points.

Of course first check to make sure you indeed have a buck regulator module.  There are eBay sellers mis-labeling or incorrectly identifying their modules as "buck" "boost" "stepdown" "stepup" etc.  and using incorrect photos.  I've sent countless messages with only partial success (language barrier I think) and have since given up.   If the black IC chip has the letters LM2596 printed on it, that's probably the best indicator that it is indeed a "buck" or "stepdown" regulator.  The "boost" and "buck-boost" modules have a similar looking IC but will not have LM2596 on it (or none that I've seen).

As Dale and GRJ point out, the capacitor located on the input of the module will charge up to the peak voltage on the track.  I don't know of any hobby-grade meter or common hobby device that can measure this voltage.  So proxies are made such as using the AC mode of a meter and multiplying by 1.4.  This works for O-gauge command mode using "pure" sine voltages, but as you mention G-gauge it can get tricky.  Are you running AC track voltage on your G-gauge layout? 

As you know some G-gauge DC supplies use pulsed-DC and of course DCC is popular which is no-where near a "pure" sine voltage!  In these cases, the 1.4 rule-of-thumb does not apply.  And of course even in O-gauge if using a Z-controller (so-called "chopped-sine") as you mentioned, the 1.4 rule-of-thumb also does not apply.

I switched over to DC and I use Bridgewerks outside right now. I'm thinking of going back to AC as the TIU's variable channels won't work with DC. So I either have to buy a few more TIUs, or swap back to AC like I used to use.

 I prefer the new version L TIUs outside as they make a big difference in signal.

The picture in the bay link seems to be correct. I don't know what the reg chip is on this one? It is tiny and hard to read. Maybe I can use these boards in buildings on the layout?

It's this chip.

RT8272 3A, 24V, 1.2MHz Step-Down Converter

You can tell a buck/boost regulator as they normally have two inductor coils.

The XL6009 chip is a boost / buck-boost Inverting DC/DC Converter.  It's used on the following board.

That's smart thinking since power sources for accessories are generally less than the 20+ volts from track transformers/controllers. The 12V DC LED strips work great for larger buildings or station platforms.  Lemax buildings and Miller animated signs use 4.5V DC.  And a generalization it seems as more lighting products switch to LED technology, the required voltages have become lower (and DC) rather than the traditional 14-16V AC accessory power.

But specific to your module, it is a DC-DC converter and "expects" DC on the input.  The "DC" you get out of a AC powered bridge rectifier needs to cleaned-up or smoothed for proper operation with these converters.  That's what the module's input capacitor does.  Your module has a relatively small capacitor - I'd guess maybe only 1-10uF or so based on the size in the photo (arrow).  It sounds like you already have 220uF/35V capacitors so use them (or 10 for $1 on eBay with free shipping).

OTOH if your DC input is coming from a relatively smooth DC source then this extra capacitor is optional. These would include the DC accessory voltage on the back of a Bridgewerks Magnum supply, a DC-output wall-wart, a DC battery-pack, and so on. 

Good info Stan. I would have skipped the caps to lower the input voltage. I can use the caps, and adjust for them, with a known target input voltage.