Background: My first day on this site, first post. Any help appreciated. I am not an electrical engineer, I am in software. I have thrown a soldering iron around in the past. This is a project to help my 12 year old son. This will be a trivial problem to solve for the folks I believe frequent these forums.

I am trying to help my son fix a Safetran Hybrid railroad crossing bell. This bell was designed at the end of the company's life around 2013. Unlike nearly every mechanical railroad crossing bell which all use contacts, it uses a small circuit apparently made in-house at Safetran. That circuit is bad. The way it works is (I will attach photos): Send the bell between 9-15v DC and an electromagnet with a hollow center picks up a ferrite rod , the circuit kicks in and cuts the DC voltage and the rod drops on the bell housing making the ringing. Circuit energizes again to lift the rod, cycle repeats. This bell will ring 250 times per minute.

I would like to create a simple circuit to handle this voltage switching. For bonus points I'd love a speed control but I probably should start simple. Space where it is mounted within the bell housing is tight.

Any help appreciated! Thank you.

 

Certainly the timing requirements are within what a 555 timer can provide, even an adjustable speed rate.Then a transistor to switch the power. Use the system DC supply if there is one.

 

What is the blue module? A picture of its label would help. Is it a timer or relay module and what voltage? Also, any specs on the solenoid other than the custom label? Is there a circuit diagram/schematic available anywhere?

 

The "blue module"is the failed bell ringing timer/pulse generator. Probably encapsulated with a hard material to prevent any repairs being made.

 

Judging by the solenoid label, it looks as though the solenoid has a resistance of 7.5Ω and dissipates 21W. Its rated voltage is therefore 12V and it draws 1.75A.

 

A 555 timer and Mosfet output will work, set for approx 4Hz.

 

Given that the the operation is to pull up the plunger and then drop it to ring the bell, it should work well to use a charged capacitor to deliver that peak coil current, since once the solenoid is pulled up holding it takes less current. So the transistor will need both a higher current rating and a higher voltage rating than the system voltage would indicate. But because of the short duty cycle it should not require a heat sink. Also, avoid the spike suppression diode so that the release is quicker. Just use a higher voltage rated transistor.

 

avoid the spike suppression diode so that the release is quicker. Just use a higher voltage rated transistor.

I would add suppression to protect the transistor but add a resistor in series with the diode to reduce the turnoff time.
A resistor equal to the solenoid resistance will give a peak transient spike of about twice the supply voltage.
That will give a turn-off time-constant about 1/2 the turn-on time-constant.

 

Welcome to AAC.

Just my thoughts on this: that blue module probably stores a charge then delivers all its stored energy to the coil which then strikes the bell.

You stated the bell rings 250 times per minute. That's 4.17 times per second. That's more like a school bell than a train crossing bell. Though I don't know how frequently a railroad bell rings (per second) I'd imagine it's probably on the order of three times per second.

I agree with the 555 approach. A capacitor can be charged from the power source then its charged energy is released into the coil to strike the bell. An easy proposition for sure. Well, easy to those who know a little about electronics. I understand you've melted some solder in the past. My guess is this is a good learning project for you and your son.

 

Found this:
https://images.search.yahoo.com/sea....circuitous.ca/BellRingerSch.GIF&action=click

 

The circuit in the link could work, I did not do an in-depth analysis, though.
The electromagnet lifts the plunger and then drops it, which is when it falls and hits the bell and then is pulled away by a spring. and I am guessing that the rate is more like 25 times per minute.