# SCRs & Latching relays

#### maassmi

Joined Oct 30, 2011
81
If I choose to power my frogs, I should be able to power them by using either method! Either way I will probably run two sets of LEDs on the one set of pins and the frog using the other set of pins, just like with the Atlas # 200 snap relay. Or, am I missing something?

#### SgtWookie

Joined Jul 17, 2007
22,227
With the switch motor (solenoid coils) being fed like this, the average current will be cut in half; it probably won't have enough power to move the switch points.

I don't see an easy way to fix this.

Last edited:

#### maassmi

Joined Oct 30, 2011
81
That's okay. I haven't powered my frogs in the past and have not had any problems. Thanks!

#### maassmi

Joined Oct 30, 2011
81
How about with your method with the 2-coil relay? Would I be able to power the frog that way?

#### SgtWookie

Joined Jul 17, 2007
22,227
Yes, with the 2-coil relay, you can use the spare set of contacts to power the frog.

I have verified that the schematic I drew will work, using a simulation. There is no leakage path for current via the switch solenoids, because only one of the solenoids will be receiving power via the Atlas #57.

The switch motors (solenoids) receive full wave AC, and the relay actuators receive half wave as I'd planned.

I also verified that djsfantasi's schematic will result in the switch solenoids only receiving half wave; as a result they would have less than half the power they do now.

#### maassmi

Joined Oct 30, 2011
81
Getting back to the diodes, upon looking back at the posts, even though each end was given a name, does it really matter which way the diode is attached?

Also, I am looking on eBay and 1N4148 diode appears to have two or three bands so I am not sure what to do at this point. I am going to keep looking as I think this is a grat solution and will be much cheeper than using the Atlas relay.

#### djsfantasi

Joined Apr 11, 2010
8,663
My interpretation of the 2-coil latching relay was that there were two separate SPDT relays in the package, each with its own set/reset input. I also assumed that setting one did not reset the other. This is where I was confused, I think. It also looked to me that switching polarity between pins 6&10 would latch the relay connecting 7,8&9 in each direction. It is this assumption that has the output of the two horizontal diodes connected and where the sneak current comes form in my version.

It does make a difference which way the diodes are wired into the circuit. There should be one band at one end; it should be wired the same way it is shown in our sketches. The 1N1418 is a Zener diode; you will need a general purpose diode.

Although the switch machines will receive half power, they will likely still work, as diodes in line with the switch motors is a common configuration used in diode matrix ladder switching.

PS SgtWookie - I am honored to work with you on this circuit.

Last edited:

#### SgtWookie

Joined Jul 17, 2007
22,227
Getting back to the diodes, upon looking back at the posts, even though each end was given a name, does it really matter which way the diode is attached?
Yes, it does - and you'll need to be consistent.

Both diodes need to have their cathodes (the end with the band) pointing towards the switch control box.

I suggest that you build the circuit on a breadboard first. That way, you can simply plug in and unplug things until everything is working properly; then you can document exactly what worked, in the way that you best understand it. Print out or sketch out the schematic I posted, and mark it up as needed to help keep yourself straight.

Also, I am looking on eBay and 1N4148 diode appears to have two or three bands so I am not sure what to do at this point.
Your local Radio Shack sells a 10-pack of 1N4148/1N914 diodes for about $2. They also sell a 25-pack assortment of 1N400x diodes for about the same amount. You are working with very low voltage, and very low current - so almost any common type of diode will work. I am going to keep looking as I think this is a great solution and will be much cheaper than using the Atlas relay. That was the whole idea. Thread Starter #### maassmi Joined Oct 30, 2011 81 Thank you very much for all your help and assistance. You folks have been great. This has been a fun and interesting learning session for me, not to mention that it did not take long for you come up with a solution and communicate it in such a manner I could understand it. Thank you! #### SgtWookie Joined Jul 17, 2007 22,227 My interpretation of the 2-coil latching relay was that there were two separate SPDT relays in the package, each with its own set/reset input. Yes, that can be quite confusing. However, the datasheet states "2 form C" (also called "changeover" or "CO"), which means that it works like a DPDT switch; the two sides are electrically isolated but mechanically connected. If the two sides were mechanically isolated, it would say "1 form C x 2" There is some verbiage in the datasheet which tends to confuse the issue, as well as the way the relay was drawn - however, I'm "sticking to my guns" here. I also assumed that setting one did not reset the other. This is where I was confused, I think. It also looked to me that switching polarity between pins 6&10 would latch the relay connecting 7,8&9 in each direction. It is this assumption that has the output of the two horizontal diodes connected and where the sneak current comes form in my version. Well, you can switch polarity on either coil, or just use one coil for one direction, and the 2nd coil for the other direction as I'm doing. Pin 8 is a common. It can connect to either 7 or 9, but not both at once. They are electrically isolated from the coils, as well as 2, 3 & 4. It does make a difference which way the diodes are wired into the circuit. There should be one band at one end; it should be wired the same way it is shown in our sketches. The 1N1418 is a Zener diode; you will need a general purpose diode. I believe you misread what he typed; it was 1N4148 instead of 1N1418. Big difference! Although the switch machines will receive half power, they will likely still work, as diodes in line with the switch motors is a common configuration used in diode matrix ladder switching. Well, I've never been "hands-on" with an HO scale set. I know that the old Lionel electric switches wouldn't work worth beans unless they were getting full AC power; I had no reason to believe that HO would be any different. A quick Google search turned up this interesting and related page on diode matrix ladder switching for a yard: http://users.rcn.com/weyand/tractronics/articles/smartcl/smartcl.htm But that brings up yet ANOTHER caveat; if the diodes were used with the single coil, it might preclude the use of diode matrix ladder switching for a yard as well, particularly if the diode placement/orientation was not consistent. It could also throw a monkey wrench into the design I put up. PS SgtWookie - I am honored to work with you on this circuit. My pleasure. #### radiohead Joined May 28, 2009 508 SGT Wookie, this is regarding your post (#30 on page 3). Shouldn't those diodes be turned the other way around? That way they will be forward biased and let the positive swing of the AC voltage pass and provide a positive voltage to the relay coil? #### djsfantasi Joined Apr 11, 2010 8,663 Sgt, Thanks for clearing this up for me. I didn't quite catch this on the datasheet. SgtWookie said: ...the datasheet states "2 form C" (also called "changeover" or "CO"), which means that it works like a DPDT switch; the two sides are electrically isolated but mechanically connected. If the two sides were mechanically isolated, it would say "1 form C x 2" There is some verbiage in the datasheet which tends to confuse the issue, as well as the way the relay was drawn - however, I'm "sticking to my guns" here. Thanks, this is the part that I missed and was causing my confusion. Great circuit! I believe you misread what he typed; it was 1N4148 instead of 1N1418. Big difference! I'm ton dyslexic! Pin 8 is a common. It can connect to either 7 or 9, but not both at once. They are electrically isolated from the coils, as well as 2, 3 & 4. This part I did understand. Thread Starter #### maassmi Joined Oct 30, 2011 81 All of this continued discussion on the diodes is getting me worried. Fortunately I have not purchased the diodes yet. I am waiting for the relays to arrive first. I am looking on the Radio Shack webpage at "Silicon Switching Dieodes (50-pack), Model: 1N914/1N4148 ~ Catalog # 276-1620. Earlier in this thread, these were listed as being ok to use for this project. Now I am reading that these are a zener diode and that I will need a general purpose diode. A little help here please. Which diode should I be looking at? #### djsfantasi Joined Apr 11, 2010 8,663 Thos Radio Shack diodes are ok; I misread what you had typed. #### SgtWookie Joined Jul 17, 2007 22,227 All of this continued discussion on the diodes is getting me worried. Fortunately I have not purchased the diodes yet. I am waiting for the relays to arrive first. Don't worry; be happy. I am looking on the Radio Shack webpage at "Silicon Switching Dieodes (50-pack), Model: 1N914/1N4148 ~ Catalog # 276-1620. Gee, I didn't know they carried a 50-pack. That's a better deal than the 10-pack. Earlier in this thread, these were listed as being ok to use for this project. Now I am reading that these are a zener diode and that I will need a general purpose diode. They are NOT Zener diodes. djsfantasi misread 1N4148 as 1N1418. The former is a fast switching diode; the latter is a Zener diode. They are completely different part numbers. 1N914 diodes are basically the same as the 1N4148 diodes. They are both extremely fast switching diodes, rated for 100mA continuous current, and 100PIV. They switch on and off so fast that it's difficult to measure the switching times. A little help here please. Which diode should I be looking at? Don't be confused. That 50-pack of 1N914/1N4148 that Radio Shack carries will work just fine for your purpose. Like I mentioned already, you can burn components up when soldering if you don't keep the heat away from them. Hemostats work great for keeping the heat away; you grip the wire lead near the body of the part and the hemostats wick the heat away. Here's an image of a hemostat: I like the hemostats that have curved jaws, but the straight ones work too. Northern Tool carries the above hemostats for$1.50.
http://www.northerntool.com/shop/tools/product_478861_478861

#### SgtWookie

Joined Jul 17, 2007
22,227
SGT Wookie, this is regarding your post (#30 on page 3). Shouldn't those diodes be turned the other way around? That way they will be forward biased and let the positive swing of the AC voltage pass and provide a positive voltage to the relay coil?
It's AC! The sine wave swings both ways. In the schematic you're referring to, the upper diode causes the relay to set (which would light the red LED), and the lower diode causes the relay to reset (which would light the green LED).

If the diodes had their ends swapped (cathodes towards the relay instead of the switch control box), the actions would be reversed; IE - the upper diode would reset instead of set (lighting the green LED), and the lower diode would set instead of reset (which would light the red LED).