Circuit Breaker for Distict

[LKOrailroad]

I want to use Ring Engineering for my control system but don't care for their AR-1 breaker/reverser. The AR-1 uses spring clips for wire termination (I prefer soldered lugs or screw terminals) and the maximum wire size the spring clips will accept is 16 AWG. I want 12 AWG. Splicing in a short section of 16 AWG into the 12 AWG bus seems so tacky even though I know it would work. I inquired about DCC Specialties products but they won't work with the Ring DC supply.

I found this circuit (see pg 14) http://cds.linear.com/docs/en/datasheet/lt1153.pdf but would rather buy a production unit that is purpose built and ready to go.

Does anyone know of a 4A electronic automatic reset circuit breaker for 15v DC other than the AR-1?

[peteski]

I suspect that the lack of response is an indication that there aren't many users of this system who visit TRW.

I have heard about RailLynx (similar, but using IR communication) but until now, I have never heard of this system myself.

I have not encountered any DC-based automatic breakers. But then again, I haven't really looked for one either. I suspect that you will probably just have to live with AR-1 and spring clips.

[LKOrailroad]

The Ring power supply puts plain 15v filtered DC on the rails. Any circuit breaker for 15V DC will do the trick. Unfortunately, I can't find such a thing. At least not one that is fast enough.

[LKOrailroad]

Found what appears to be the ideal solution. Going to build one to test. The thermistor mounts to the MOSFET heat sink for secondary overheat protection (Ex. Ring power supply breaker doesn't trip). The diagram indicates the thermistor sensing motor heat. The 1153 datasheet says can also be mounted on MOSFET heat sink or not used at all. The diode across the motor in the diagram will have to be moved to the board. This would normally reduce its effectiveness however in the case of Ring there is a decoder between the motor and the breaker. The decoder deals with the heavy motor spikes, not the breaker. Maybe could eliminate the protection diode altogether. The chip also has low level logic input to control on/off which could be used to digitally shut down unused tracks (Ex. roundhouse) to quiet sound equipped locos. Cool feature for free.

Looks like a complete unit can be built for about $10 in parts - IC, MOSFET, sink, thermistor, and sense resistor. I have on hand sufficient quantity of the common components. I'm going to give it a try and see what happens. Nothing to lose except a little factory sealed smoke.

Datasheet: http://cds.linear.com/docs/en/datasheet/lt1153.pdf

[peteski]

Looks promising. The diode across the motor is to protect the circuit from the motor's inductive kickback voltage.  In your case, where the motor is controlled by the decoder, that diode is not needed. The decoder should prevent any of the motor's kickback voltage from reaching the breaker circuit.  But having the diode installed will not hurt anything either. I also think that the thermistor is an overkill (but it is a nice feature).

[LKOrailroad]

I know this is an old thread but that is the speed at which I work! 

After many trails and tribulations I finally arrived at a circuit breaker design that works great. The LTC1153 chip was the right answer. All that was needed was to refine the component values to get the right trip, delay, and recovery times for use with actual locos and my power supply. If you have a DC or Rail Pro system and want fast electronic circuit breakers for your power districts then you will like these.

Details on two different posts on the LK&O Railroad blog.
Working through the design phase: http://www.lkorailroad.com/circuit-breaker-and-block-detection-design/
The finished product complete with PCB mask and parts list/placement if you want to build your own: http://www.lkorailroad.com/circuit-breaker-and-block-detector-final-units/

I worked on a block detector design at the same time I worked on the circuit breaker. Here is a pic of both assembled units ready for placement into service on the layout.

[peteski]

Nice job!  I have some unfinished projects which are over 20 years old! 

[bdennis]

Nice work on the block detectors.
I have a question.. Are you feeding the status of the block detector into JMRI some how or just showing status via a LED on the facia?

[LKOrailroad]

I have a question.. Are you feeding the status of the block detector into JMRI some how or just showing status via a LED on the facia?

There is an LED on the board for troubleshooting purposes. There is an optocoupler providing a low output when the block is occupied. The detect signal from the optocoupler will drive the logic for my signaling system which is loosely based on the old TracTronics approach - dedicated logic circuits. The logic circuits and signal head driver circuits are on the drawing board yet to be built. The circuit breakers and block detectors were needed now as they are an integral part in the track wiring which is the phase I am in building the layout.

My railroad is RailPro so no JMRI. This is by design. RailPro has only clean 14 volts DC on the rails which greatly simplifies design of detection circuits compared to DCC. I enjoy building electronic stuff and so want to build the entire signaling system myself. Sort of like scratch building as opposed to RTR. I prefer to scratch build my signaling.