Chuff switch for fan driven smoke

I must admit, I was initially confused by the circuit, it's upside down!   I had to go back and take a second look to analyze exactly how it worked. 

I was also curious about the pencil notation that crossed out the original Zener value and substituted the 13 volts.

I'm a bit surprised they transition at that high a voltage, you'd have to be really cooking down the rails to have a track voltage of 14-15 volts I would think!  Sounds to me like it mostly runs on one resistor.

Don't forget that the 13 volt transition is AC that has been rectified to DC; so the equivalent track AC value is more like 9 volts.

Or did I miss something?

Rod

Rod,  AC volts would be higher, the full bridge rectifier drops about 1.4 volt turning the AC to DC.  2 diodes worth.  The smoke unit runs on the rectified DC.

You also have to remember the QSI/PS-1 boards have several diodes in series turning AC to DC for the motors.  So track voltage is higher than DC volts applied to the motor by 3 to 4 volts.

I agree though, my PS-1 engines move pretty well at 10V.  14-16 volts and they are screaming.

I traced an actual smoke unit circuit and the diagram is correct.  The larger Q is a PNP, I could not find a reference on the other one, but I think it is a PNP also.  What got me was looking how the resistor was in series effectively cutting current in half.

I charted the current and Watts at 2V increments.  You get low resistance at low volts which is good for smoke output, but as volts get higher and you kick in the 2nd resistor the current drops in half.  You make the power up as voltage increase.  I am sure this helps with the resistor longevity.  These are such good smoker they don't need to be messed with, but if you lowered that second resistor to about a 8 ohm, it would really be cooking.

I am going to run a test with one of my engines and run up the voltage in neutral slowly and see if I can see any difference at the cut over threshold.  G

I don't think this is correct.
Any full bridge rectifier will produce DC that is about 1.4 times the rms AC value. That is what I have always heard and proven in practice.

I stand by my original point that the 13 volts DC transition equates to about 9 volts AC track power.

Rod 

Rod,  all these can motors are rated at 12V DC.  So a starter set engine with any of the manufacturers conventional reverse units with bridges would have 14 volts at the motor when track voltage is 10VAC and at 14VAC the motor would have 19.6VDC?

I don't think a can motor can handle 20VDC.  G

Under load, the DC voltage will not be 1.4 times the peak AC value, that's only open-circuit.  Try a simple experiment on your workbench with a full wave bridge, a cap, and a load, you'll see what I mean.   Since this DC is under load, the 1.4 figure doesn't apply.

John;

I like your idea of putting the magnet on the drive rods.

This way you can run the fan and shut it off with the switch in mid stroke, thus making the smoke puffs sync with rod ends travel like the real engines.

And running lo speed with burst at chuff time also works well. that way even if the engine stops with the reed switch active you won't burn out the heat resistor.

(if you use a one shot triggered by the switch, it would also eliminate the problem)

Also, to convert a PS1 to PS2, they give instructions on modifying the smoke unit, you might want to check into that. (I have a PS2 kit that has this info if you want a copy)

I'd like a copy of those instructions, perhaps you could post them here?  It would be interesting to see what they do to the unit.

Yep, I don't think I want to run the two 16 ohm resistors in parallel, that's going to be too much of a good thing!

What does the Lionel ACREG unit take in as it's trigger?  Is there any hope of interfacing this to make the job simpler?  Got any reference to the part number and where it's used?  I would like to see if it's cost-effective and fits the application.

It takes in Serial data from the R2LC.  This is the unit on Lionel's "Super Smoke fan driven units.  Some of these units also control other functions.  I think it is like MTHs MUX board in concept.

I will have to do some research, but I think the Lionel Techs could help.  I have seen this on the newer TMCC Diesels and such.  G

If it takes serial data, I'm assuming that it requires some support from the R2LC to do it's thing.  That may be out of reach with the C08 code that comes with the ERR Cruise Control.