Adding Latching Relay to Seep Point Motor to Change Frog Polarity

[Jeremy Cumberland]

Do the relays have HFD2/012-S-L2-D printed on them?

 

9V might not be enough to switch the relay (the data sheet says 9.6V), although the click is pretty much a giveaway.

 

How are you testing the switching contacts (pins 4, 6, 8, 9, 11 and 13)?

 

With a bistable relay such as this, you cannot tell the initial state. If you apply 12 V to pins 1 and 16, the relay should "set". If you apply 12 V to pins 2 and 15, the relay should "reset". Once the relay is "set", you can apply and remove power to pins 1 and 16 as much as you like and the relay state won't change. It will only change if you apply power to pins 2 and 15.

[Brian]

How many pins has the relay got.  8 its a bidirectional latching relay.  10 pins and its a twin coil latching relay.

If it has 8 it will need reverse polarity applied to pins 1 and 16 to unlatch it i.e. apply positive to pin 1 and negative to pin 16 to latch the relay. Then apply positive to pin 16 and negative to pin 1 to unlatch it.

IMO a twin coil latching relay is far easier to wire into a solenoid motor circuit

Edited March 28 by Brian

[kevinlms]

1 hour ago, RFS said:

 

Same relays as I have. Their input must be DC which is why I fitted diodes. Check the output of your CDU as I suspect its output might be AC. 

A CDU cannot put out AC. But yes relays are usually DC, but you can have AC ones.

[grob1234]

1 hour ago, Jeremy Cumberland said:

Do the relays have HFD2/012-S-L2-D printed on them?

 

Yes they do.

 

In short, I think the CDU has supplied too much voltage and blown the relay.

 

The CDU outputs DC voltage. It is this exact model:

 

https://www.layouts4u.net/other-products/track-related-items/block-signalling-cdu1c

 

When testing the other relays with 12v DC, there is an audible click, and you can feel it inside the relay, and the output changes as you'd expect.

 

I'd imagine 24v relays would work well, I just don't want to spend another £24!

 

Oh well, live and learn.

 

The relays have 10 pins, and testing on the bench has shown them to work exactly as advertised, so I am confident I have the right type, just the wrong coil voltage.

 

 

Edited March 28 by grob1234

[kevinlms]

More details here.

 

https://www.tme.com/Document/cb545971aaf7e641179e4750a1704dc9/HFD2_en.pdf

 

What voltage did you say the CDU put out? You do have a multimeter?

[grob1234]

21 minutes ago, kevinlms said:

More details here.

 

https://www.tme.com/Document/cb545971aaf7e641179e4750a1704dc9/HFD2_en.pdf

 

What voltage did you say the CDU put out? You do have a multimeter?

I wasn’t sure how to measure the output of the cdu, as it’s momentarily on, I couldn’t get an accurate reading with my multimeter. Is there a way to measure it?

 

Edit, I have measured across the terminal. 23.4v… there’s my answer confirmed… if only I figured this out before hand. Doh!

 

Even the 24v relays have a max set/reset voltage of 19.2….

 

Edited March 28 by grob1234

[kevinlms]

17 minutes ago, grob1234 said:

I wasn’t sure how to measure the output of the cdu, as it’s momentarily on, I couldn’t get an accurate reading with my multimeter. Is there a way to measure it?

 

Edit, I have measured across the terminal. 23.4v… there’s my answer confirmed… if only I figured this out before hand. Doh!

 

Even the 24v relays have a max set/reset voltage of 19.2….

 

Measure the resistance of the coils - power removed totally of course! That will tell you if they're cooked - plus the burning smell of course!

[grob1234]

Well, thank you all for your help and replies.

 

I’m obviously at the limit of my understanding of electrical stuffs, so I’ll retire gracefully and put the idea of latching relays for changing frog polarity to bed.

 

12v latching relays now available on EBay!

 

 

[Jeremy Cumberland]

20 minutes ago, grob1234 said:

Even the 24v relays have a max set/reset voltage of 19.2….

That is the highest the setting voltage will be (the setting voltage is no higher than 19.2 V in a 24 V relay). Quite likely the actual setting voltage is lower, and you were able to set your relay with a 9 V battery, when the setting voltage listed in the datasheet is 9.6 V. The maximum voltage the coil can withstand is in the right hand column: 27.7 V.

 

If you have connected the relay across the CDU on its own, without the point motor, you might possibly have fried it, but if you had the point motor in circuit as well this seems very unlikely, particularly if you wired directly to the point motor terminals rather than the switch terminals and the CDU. Given the low resistance of the SEEP coils, it seems more likely that you had too low a voltage going to the relay than you have too high a voltage going to it.

 

You haven't said exactly how you bench tested the relay. What did you do after hearing the "click"? Did you alternately apply the voltage across 1 and 16, then 2 and 15, then 1 and 16 and so on, to set and reset the relay, or did you just apply and remove the voltage from pins 1 and 16, for this will do nothing. Bistable relays aren't like ordinary relays and won't change each time you apply or remove the power to a coil. They only change if you apply or remove the power in a certain way, depending on the relay type and how it is already set. For your relay, applying 12 V across 1 and 16 sets the relay one way, and applying 12 V across 2 and 15 sets it back again. Only a momentary connection to a power supply is required - according to the datasheet, the switching time is 4.5 milliseconds.

 

Rather than listening for the click, to test the action of the relay you should either incorporate it in a circuit - a battery and a lamp/LED, for example - using pins 4/6, 4/8, 13/11 and/or 13/9, or else use a meter to measure continuity or resistance across one or more pairs of pins (4/6, 4/8, 13/11 or 13/9).

[grob1234]

2 minutes ago, Jeremy Cumberland said:

You haven't said exactly how you bench tested the relay. What did you do after hearing the "click"? Did you alternately apply the voltage across 1 and 16, then 2 and 15, then 1 and 16 and so on, to set and reset the relay, or did you just apply and remove the voltage from pins 1 and 16, for this will do nothing. Bistable relays aren't like ordinary relays and won't change each time you apply or remove the power to a coil. They only change if you apply or remove the power in a certain way, depending on the relay type and how it is already set. For your relay, applying 12 V across 1 and 16 sets the relay one way, and applying 12 V across 2 and 15 sets it back again. Only a momentary connection to a power supply is required - according to the datasheet, the switching time is 4.5 milliseconds.

 

Rather than listening for the click, to test the action of the relay you should either incorporate it in a circuit - a battery and a lamp/LED, for example - using pins 4/6, 4/8, 13/11 and/or 13/9, or else use a meter to measure continuity or resistance across one or more pairs of pins (4/6, 4/8, 13/11 or 13/9).

 

Yes I did exactly this to test the relay, and it worked perfectly. Now, having used it connected to the CDU etc, its broken.

 

Not to worry, it was only an attempt to improve reliability. I'll list the remaining items on Ebay and try and recoup my money!

[cliff park]

The output of a CDU can NEVER be AC. It comes from a charged capacitor. (CDU =Capacitor Discharge Unit). If it is being charged from a nominal 16v AC it may well be charged to about 24v , which may be too much for a 12v relay coil. By the way RFS your reference to 30v, 3amp is it's capacity to switch, not the coil

[GrumpyPenguin]

I would suspect that the peak voltage output of the CDU is/was far in excess of what the realy coils could handle or possibly the back emf from the solenoid coils did the damage.

 

The peak voltage output may not register on the cheaper end multimeters.

 

What sort of switch are you using to operate the points ?

(I have an idea - don't sell the relay's yet).

[Suzie]

I suspect that you may have a faulty relay here. Those you have are good for up to 27V so you cannot cook them from the 24V output of your CDU.

 

The most likely thing that has happened is too much current through the contacts welding them together due to a shorted load when testing. Sometimes a good beating with the wrong end of a big screwdriver can free the contacts but it might be permanently dead. Carefully check that you have wired the contacts correctly, they are not always connected intuitively.

 

I would not give up yet - plenty of people have achieved excellent reliability so stick with it.

[meil]

There's nothing stopping you adding a resistor in series with the relay coils to reduce voltage across them.

[grob1234]

8 hours ago, GrumpyPenguin said:

I would suspect that the peak voltage output of the CDU is/was far in excess of what the realy coils could handle or possibly the back emf from the solenoid coils did the damage.

 

The peak voltage output may not register on the cheaper end multimeters.

 

What sort of switch are you using to operate the points ?

(I have an idea - don't sell the relay's yet).

This is my theory too.

 

At “rest” the cdu output is 23.4v or thereabouts, but I don’t know what it spikes to upon operating the switch. More than 23.4v I’d have thought, and yes I only have a basic multimeter, so I’m not surprised it can’t register the momentary voltage spike.

 

The switches are (on) off (on) sprung to centre type SPDT type recommended for solenoid point motors.

 

58 minutes ago, Suzie said:

I would not give up yet - plenty of people have achieved excellent reliability so stick with it.

This is the annoying thing here… the test with a stable 12v supply produced exactly the results I was hoping for! So the theory and wiring was correct.

 

With regards to relays, could it be possible that a relay with a 24v coil might be more appropriate? Would the 24v relay be able to handle a higher number of amps?

[melmerby]

20 minutes ago, grob1234 said:

With regards to relays, could it be possible that a relay with a 24v coil might be more appropriate? Would the 24v relay be able to handle a higher number of amps?

In what way?

The coil will be higher resistance and will draw less current at it's rated voltage

 

[kevinlms]

49 minutes ago, grob1234 said:

This is my theory too.

 

At “rest” the cdu output is 23.4v or thereabouts, but I don’t know what it spikes to upon operating the switch. More than 23.4v I’d have thought, and yes I only have a basic multimeter, so I’m not surprised it can’t register the momentary voltage spike.

 

 

The CDU cannot give out a higher voltage than that across the capacitor.

Where the higher voltage comes from, is after the point motor has thrown, the Back EMF causes a large spike (100s of volts) in the reverse direction. This is what the diodes are for, to shunt those volts from doing damage.

 

Even a more expensive multimeter won't tell give you that information accurately, so don't bother buying one. Just take the advice that there is indeed a large spike of reverse voltage.

[grob1234]

3 minutes ago, kevinlms said:

The CDU cannot give out a higher voltage than that across the capacitor.

Where the higher voltage comes from, is after the point motor has thrown, the Back EMF causes a large spike (100s of volts) in the reverse direction. This is what the diodes are for, to shunt those volts from doing damage.

 

Even a more expensive multimeter won't tell give you that information accurately, so don't bother buying one. Just take the advice that there is indeed a large spike of reverse voltage.

 

OK that's good to know about the CDU.

 

And part 2 also makes sense as well. As you can tell, its all a learning process for me - I'll take your word for it that there's a large spike of reverse voltage. For me, in order to understand why things are as they are, its important to figure out where the excess voltage comes from, which you have explained. Thank you.

[GrumpyPenguin]

1 hour ago, grob1234 said:

The switches are (on) off (on) sprung to centre type SPDT type recommended for solenoid point motors.

What you could do (& this depends on the amount of wiring required) would be to replace the switches with DPDT ones & replicate the point motor wiring to the relays (as your diagram) but supply this circuit from your12v supply.

 

As suggested, it may be a faulty relay but I've never had one fail yet without it being something external. I would still avoid trying to operate them using the CDU.

[grob1234]

1 hour ago, GrumpyPenguin said:

What you could do (& this depends on the amount of wiring required) would be to replace the switches with DPDT ones & replicate the point motor wiring to the relays (as your diagram) but supply this circuit from your12v supply.

 

As suggested, it may be a faulty relay but I've never had one fail yet without it being something external. I would still avoid trying to operate them using the CDU.

 

Unfortunately, that would require too much wiring for the benefit gained.

 

I've definitely buggered the relay, it worked fine before hand.

 

It does seem the CDU/Relay combination is perhaps not the one. Never mind, it was worth an exploration.

[Suzie]

What is the resistance of the coils on the failed relay? they should be between 864 Ohms and 1056 Ohms and if they are you probably have not damaged the coils.

 

If the coil is damaged it is probably down to not using an inverse parallel diode to divert the back EMF from the SEEP motor (1N4148 diodes are ideal for this and very cheap).

Edited March 29 by Suzie

[melmerby]

9 hours ago, grob1234 said:

I'll take your word for it that there's a large spike of reverse voltage.

That's how the ignition coil on a motor vehicle works.

You only have 12v but the resulting back EMF when the field collapses (after the points open) is many thousands, enough to generate a healthy spark.