CDU Servo Point Motor

[AndyID]

This is a way to use a servo to drive points. The servo is modified (hacked) so that only its gears and motors are used. The electronic bits go in the spares bin (or more likely the waste bin ).

 

It's similar to a CDU except that each servo motor has its own capacitor. When the cap is charging the motor goes in one direction and when the point control switch is changed the capacitor discharges and the motor goes in the opposite direction. No current flows while the capacitor is fully charged or discharged.

 

The SG90 servos I have rotate through 90 degrees for the values I've shown. More capacitance or voltage will increase the angle. If the motor hits a hard stop and stalls most of the energy stored in the capacitor is dissipated in the resistor.

 

The switch indicates the turnout direction. It could be on a control panel or it could be one set of contacts in a slave 2-pole relay where the other set of contacts control frog polarity.

 

 

 

Edited April 16, 2022 by AndyID

[Phil Himsworth]

Very clever, I like that a lot. I've just spent ages integrating a load of servos with a servo controller controlled over I2C by a raspberry pi hosting a home made route control web app that includes a function to record servo setpoints and store them in a database; you've done it with just two basic components and no need to worry about setpoints!

 

What happens when the power is turned off when the capacitor is charged? Does the capacitor stay charged, discharge in such a way as to move the servo, or discharge slowly without moving the servo?

 

 

 

[AndyID]

6 hours ago, Phil Himsworth said:

 

What happens when the power is turned off when the capacitor is charged? Does the capacitor stay charged, discharge in such a way as to move the servo, or discharge slowly without moving the servo?

 

 

Hi Phil,

 

Good question! That will depend on the power supply and any other loads connected to it but it's most likely to drain the cap without turning the servo.

 

It's probably a good idea to make the "normal" turnout position discharge the cap and set all the turnout switches to normal before powering down. That should ensure that the switches correspond with the points at power up. Another way to do it is to exercise all the switches at power up to sync them with the points.

 

Cheers,

Andy

[Barclay]

Looks good - hacked servos seem a great idea, though it does seem to offend people who like playing with electronics !

 

My own experiments, based on advice found on RMWeb (of course !)  are based around the servo driving a SPDT switch which changes the polarity of the supply to the servo, so it effectively switches itself off. Control is by another similar switch.

[AndyID]

On 04/08/2021 at 02:02, Barclay said:

My own experiments, based on advice found on RMWeb (of course !)  are based around the servo driving a SPDT switch which changes the polarity of the supply to the servo, so it effectively switches itself off. Control is by another similar switch.

 

That might have been me.

 

Did it look a bit like this? (I tend to prefer the thruppence-ha'penny solution rather than the ten bob solution )

 

 

 

 

 

Edited April 16, 2022 by AndyID

[Barclay]

9 hours ago, AndyID said:

 

That might have been me.

 

Did it look a bit like this? (I tend to prefer the thruppence-ha'penny solution rather than the ten bob solution )

 

Yes I believe it was  - thanks for the idea. I have a prototype up and running and it's very reliable. 

[DCB]

Handy for semaphore signals as well,    Simplicate and add more cheapness to mis quote Sir Sidney Camm.

[AndyID]

5 hours ago, DavidCBroad said:

Handy for semaphore signals as well,    Simplicate and add more cheapness to mis quote Sir Sidney Camm.

 

It's probably not a good idea for me to express my real opinions on the misapplication of technology