Hi,


I have seen this great project developed by a very talented guy which provides a nice easy way of driving a slot car around a track while getting a visual indication of the quality of the power supply at every point. Would be great to identify poor connections or ideal points for additional wire taps. Problem is he doesn't go into any detail on how to create what he did. I have included some info, anybody in a position to join the missing gaps for me?

 

Sounds like basically he is just using a voltmeter and dummy load wired into a slot car chassis that he rolls along the track to detect voltage drops due to high resistance in the track connections..
The voltmeter with LED indicator could be made with an LM3914 dot/bar circuit.

You would have to decide what voltages are appropriate to light the green, yellow, orange, or red LEDs and then design the LM3914 circuit accordingly.

You might also consider using a panel digital voltmeter if you can find one small enough for your purpose, which would be simpler.

 

That sounds exactly correct. Looks like he was using the slot car motor "blocked"?? to apply 2 amps of load to the track power at 14vdc. Also seems to suggest at the bottom of his page that he was thinking developing another method of applying load so the motor was free and car could be driven around the track.

I had a look at the LM3914 and it looks like it would do the job. I checked out the link you provided and although a lot of what they are talking about is above me I get the idea.

 

Personally I would look for a small digital voltmeter such as one of those I referenced.
It would be much simpler to build and will give you more information.

 

There are a lot of ways to approach this. It could be as simple as a 2A dummy load and a voltmeter, forget the car. I'm a big fan of light bulbs as dummy loads, since the brightness the lightbulb is good feedback. The voltmeter would just be quantitative backup.

On the more complicated end of the spectrum, you could have a car set to travel slowly around the track carrying the dummy load and voltmeter. Just start it going on the track and watch for spots where the bulb dims. Better yet, have it clean and polish the rails as it goes. Kind of like a Roomba for slot tracks.

All that stuff with microprocessors, wireless and bluetooth seems overly complicated, in my opinion.

 

There are a lot of ways to approach this. It could be as simple as a 2A dummy load and a voltmeter, forget the car. I'm a big fan of light bulbs as dummy loads, since the brightness the lightbulb is good feedback. The voltmeter would just be quantitative backup.

On the more complicated end of the spectrum, you could have a car set to travel slowly around the track carrying the dummy load and voltmeter. Just start it going on the track and watch for spots where the bulb dims. Better yet, have it clean and polish the rails as it goes. Kind of like a Roomba for slot tracks.

All that stuff with microprocessors, wireless and bluetooth seems overly complicated, in my opinion.

Years ago I saw a magazine project for an electric railway track cleaner.

Basically its a blocking oscillator with a thick wire secondary. The secondary is in series with the feed to the tracks and has a low impedance to the rough DC from the controller. Normal current draw quenches the oscillator - but grot on the track allows the oscillator to start up, the secondary voltage is high enough to burn up any crud.

There's a special name for this circuit - if I could remember what it was; there's probably loads of examples out there............

 

There are a lot of ways to approach this. It could be as simple as a 2A dummy load and a voltmeter, forget the car. I'm a big fan of light bulbs as dummy loads, since the brightness the lightbulb is good feedback. The voltmeter would just be quantitative backup.

On the more complicated end of the spectrum, you could have a car set to travel slowly around the track carrying the dummy load and voltmeter. Just start it going on the track and watch for spots where the bulb dims. Better yet, have it clean and polish the rails as it goes. Kind of like a Roomba for slot tracks.

All that stuff with microprocessors, wireless and bluetooth seems overly complicated, in my opinion.

I think the appeal of having a board on a car running around the track indicating supply voltage quality under load is the visual simplicity of the LED indication. I can see it not been easy to actually obtain readings from a volt meter on a car that is taking turns ect. Running a car around the track with a green led indicating all good, Amber ok but could be better and red something is wrong is ideal. Tracks like mine are not easily accessed on all sides where we use grabbers to collect cars that have crashed hence my attraction to something that can go around the track on its own

 

So if you want to do the LM3914 circuit, tell us what voltage range you want for each of the LEDs to light and we can help you with the design.

 

So if you want to do the LM3914 circuit, tell us what voltage range you want for each of the LEDs to light and we can help you with the design.

Anything above 12.5vdc would be great, 12.5 to 12vdc would be everything ok, below 12vdc would need attention. A cool feature would be to have these set points adjustable.

 

While the original description mentions a uC, the output is only 4 LEDs. A uC can do sexy things like average out very small perturbations in the track voltage, retain the lowest voltage indication after a complete lap, etc. But at its most basic level, this is 4 level comparators each driving one LED, and can be achieved with a single LM339 quad comparator. No uC, no programming, no 555.

ak

 

Anything above 12.5vdc would be great, 12.5 to 12vdc would be everything ok, below 12vdc would need attention. A cool feature would be to have these set points adjustable.

If you want the trip voltages adjustable then you could use the four comparators in an LM339 package as AK mentioned.

To only light one LED at a time you could use a priority encoder, such as a 40147, to simplify the logic to do that.
You would connect the 4 comparator outputs to inputs 1, 2, 4, & 8 (lowest comparator voltage to highest comparator voltage) of the encoder and the LEDs to outputs A, B, C, and D (also lowest to highest voltage).
That way only the LED with the highest voltage is lit.

 

For the record, the LM3914 has both the single LED mode and a bargraph mode.

 

For the record, the LM3914 has both the single LED mode and a bargraph mode.

Its called; "dot mode" in the datasheet.

 

Hoping to bring this thread back to life, never followed it through to a conclusion.

 

Anything above 12.5vdc would be great, 12.5 to 12vdc would be everything ok, below 12vdc would need attention. A cool feature would be to have these set points adjustable.

Easy enough. With three states we're now down to one dual comparator, LM393.

Question about the display - when the track conditions change from great (green) to OK (yellow), does the green LED stay on or go off? Same question for the transition from OK to attention (red).

This is a comparator application where you do *not* want hysteresis. For marginal conditions, flickering between two display states is a good thing. The question is, which is more attention getting:

a) green stays on and yellow flickers
b) the LEDs flicker back and forth between green and yellow

ak

 

I think changing from green to amber to red based on "supply" would be best.

 

First pass at a monitor schematic. The 4.096 V reference is the reference for the Great-to-OK transition, which we'll call T1 for Transition-1. The wiper of R3 always is less, and sets the reference for the OK-to-Trouble transition, which we'll call T2. With the values shown, the voltage at U1pin3 is 4.096 V when R6 is centered and the input is approx. 12.5 V. Adjust R6 to set T1. The range is approx. 11.9 V to 12.9 V. After that, adjust R3 to set T2. The range is 100% to 80% of T1. You can play around with the resistors and Ohm's Law to create other detection ranges.

When the input voltage is Great, both comparator outputs are high and only D1 is lit. When the input drops to Good, U1A goes low. D5+D6 has a lower combined forward voltage drop than D1+D2+D3, so D5 comes on and D1 goes off. Likewise, when the input drops to Trouble, D4 comes on and D5 goes off. Since only one LED is on at any time, only one current limiting resistor is needed.

ak

 

Great work AK. Am I correct in thinking it would be best to place a load on the circuit to check for voltage collapse while testing?

 

Total for the monitor is around 20 mA. Any real load current iss external to that. I would have more, but I don't know what the norms are. For any load resistor, calculate the power dissipated and double that.

ak

 

A simple solution would be to simply monitor the current drawn from the power supply.
However, I suppose it is more effective to keep your eye on the car as it goes around the track.

Here is an even simpler solution.


Wire this circuit across the motor on the car. The LED dims as the supply voltage drops. Red or yellow LED dims to about half brightness at about 9.5V. You get different effects with different LED colours.

LED is off at about 8.0 to 8.5V depending on red or yellow LED.

Experiment with different zener voltages to get the desired trip point.