Schematic For PS-1 Smoke Unit; Board Modified

Here you go.

If the fan is not working, the usual suspect is the 3-pin 5V regulator.  If you don't measure 5V DC at the 2-pin fan connector (fan connected/disconnected), try ebay for a wallet-busting 2-cents each (free shipping from Asia) for a life-time supply!    Or 15-cents from a U.S. seller.

John and Stan;

Thanks for that info; just what I needed!

Rod

The reason the 5V regulator goes is without any heatsink, the TO-92 package is only good for maybe 20ma running from track power.  If you have 12V on the track, the regulator is seeing around 17 VDC on the input, and it has to drop 12V to output 5V.  That results in dissipation of .25 watts at 20ma, but if the smoke motor draws 40ma, that rockets up to half a watt!  That's more than the little package can deal with, so it shoots up to more than 100C.  Even if it runs for a spell, it just cooks the regulator.  For the same reason, I replace a lot of those on Lionel smoke units.

There are off-the-shelf heatsinks of course, but search "DIY TO-92 heatsink" 

The advantage of rolling your own using thin sheet metal is you can bend and fold to fit the unique space around your smoke unit.

Good info again guys thanks. That RYO heat sink is pretty neat.

So actually the smoke unit in question operates just fine; it's a NOS spare one I had in the parts box. But I want to modify it for a friend who runs RC boats. One of the boats is steam powered (simulated), and he figures a smoking stack would be a nice touch. The problem is the boat is powered by a standard 8 VDC high amperage battery  , like those used in modern RC aircraft. So instead of 12 VAC track power, we have 8VDC battery power to work with. Of course the smoke output is anemic on 8 VDC.

So looking at my markup below, I figured to jumper the bridge diode as shown, and cutout C1, not needed. The 5 V reg circuit can stay intact to power the fan. I would also remove zener diode D2, resistor R2, and transistors Q1 & Q2, whatever all those components do. Then I would jumper the center point of the two 16 ohm heaters to hot as shown, and cut the trace powering the top of R1, and jumper it to ground of R3, which on the PCB is right next to it. Fairly straightforward.

When this is all said and done I should have an effective resistance across R1-R3 of about 8 ohm, which should draw about an amp on 8 Volts, and use about 8 watts of power. From my experience this should produce a pretty good volume of smoke. So I just wonder if you guys think my logic is OK, and whether this will work as intended?

Rod

It costs nothing to leave C1 in place.  The regulator benefits from capacitance on its input for electrical stability.

Update! I got the mod done and it smokes like a trooper. I could only run it for 10-20 seconds at a time in my shop otherwise the smoke got too thick. It draws .95 amps at 8 vdc, so about 8 watts of power. I think when my friend gets it installed in his boat it's going to look great.

In the end I left C1 in place, as well as the D2/Q2/Q3 stuff. There was no good reason to remove them, so it's just easier.

Anyone looking to use a smoke unit for a building stack or other trackside structure could do the same mod. And it would be a snap to power it up using one of the mini ac or dc to dc buck converters that have been mentioned here before.

Rod

One caveat Rod.  If it runs out of smoke fluid, it'll cook things in short order with 8 watts inside that small smoke chamber!  One reason I'm a bit reluctant to suggest this kind of thing for stationary buildings is the issue of keeping fluid loaded.

Good point John, and I plan to warn my friend accordingly. I beleive he mentioned he plans to be able to switch it on off from his rc console. I figure it should be able to run 10 minutes or so on a refill before smoke dwindles, so we shall see.

Thanks to both you and Stan for assitance and suggestions on this small project! 

Rod

A "Delay-OFF" relay module could provide some protection from an extended fluid-out condition.  I've seen some that use FET switches instead of relays so they are smaller if space is an issue.  5V and 12V modules seem the most common, but as shown there are those with wider DC range for your 8V application.

There have been sporadic discussions on DIY fluid-out detectors such as sensing the resistance of the heater elements as a proxy for temperature but can be tricky to implement so that goes to how good of a friend are we talking about.   I've pondered application of washing machine and vacuum cleaner sensors that sense the "dirtiness" of the flowing water or air to adjust operation.  I figure such a mechanism could detect when the flowing air in the smoke stack becomes clear.

I was thinking about maybe something like a high temp switch like that on the plenum of a furnace, but smaller so it could be attached to the outside of the fluid chamber. Maybe adjustable set initially to about 50 C.

Not sure where you would find something like that? 

Rod

I don't know if there are economical adjustable versions but you can experiment with a fixed temperature 2-wire thermal-cutoff-switch that would go in-line (in series) with the power to the smoke unit.  I suggested the concept in this OGR smoke-related thread.

Should be less than $1 in your choice of cutoff temperature.  For example, on eBay:

Expect to do a lot of experimentation to find a suitable temperature if you can only sense the chamber's outer wall...as opposed to somehow detecting the heater-resistor temperature itself.

Stan, thanks for the idea and the link on the KSD switches. I kinda think that experimenting with these would be way more time consuming than it's worth, so we will just go ahead without for now. If we burn out a smoke unit, well, so be it. Lesson learned. Easy enough to rebuild.

Once installed I will try to get some video of the "sea trials" (well, pond trials really) to share. Should be interesting. 

Rod

If you go for the KSD9700, go for the normally-closed version or you will never get smoke!