Hi, not sure if I have a problem or just a complete lack of understanding of signal masts.

?

I have a very simple panel with 3 signal masts and two blocks

?

Sm1????? ???????????????? ? ? ? ? ? ? ? ??sm2? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?sm3

? #-------------------------------------#--------------------------------------#

????????????????????? B1???????????????????????????????????????? B2

?

The signal mast logic for SM2 is be green if B1 is unoccupied.

The signal mast logic for SM3 is be green if B2 is unoccupied.

?

If B1 is occupied and B2 is unoccupied ?SM2 and SM3 both go red. Is this correct (I was hoping that only SM2 would be red and SM3 would remain green).

?

Thanks

?

Jim

开云体育

(I’m assuming that the signals are visible to trains moving from right to left)

With B1 occupied, B2 unoccupied, and nothing else relevant happening, I’d expect sm2 to be red and sm3 to be yellow.

sm3 should be protecting B2 _and_ sm2: it’s appearance depends on both of them.

Is there any chance that you configured sm3 to protect both B1 and B2?

Bob

On Dec 17, 2019, at 8:21 AM, James Anderson <james_anderson_999@...> wrote:

Hi, not sure if I have a problem or just a complete lack of understanding of signal masts.

I have a very simple panel with 3 signal masts and two blocks

Sm1 sm2 sm3
#-------------------------------------#--------------------------------------#
B1 B2

The signal mast logic for SM2 is be green if B1 is unoccupied.
The signal mast logic for SM3 is be green if B2 is unoccupied.

If B1 is occupied and B2 is unoccupied SM2 and SM3 both go red. Is this correct (I was hoping that only SM2 would be red and SM3 would remain green).

Thanks

--
Bob Jacobsen
rgj1927@...

Many thanks for the replies. Once again my lack of understanding. I was looking for a simple red/green system on the lights. I disabled the approach aspect on the mast and so when it could not go yellow it went? to red (your mention of yellow highlighted that to me so thanks ). I think I might have to rethink what I want to do and perhaps put virtual masts in between physical masts to hide the yellow.?

Thanks?

Jim

If you don’t want the Approach (yellow) to show, you could

1) Not have each mast protecting the following one. Just delete those references.

or

2) Configure your mast definitions so that the Approach is mapped to “light the green output”

Bob

On Dec 17, 2019, at 9:51 AM, James Anderson <james_anderson_999@...> wrote:

Many thanks for the replies. Once again my lack of understanding. I was looking for a simple red/green system on the lights. I disabled the approach aspect on the mast and so when it could not go yellow it went to red (your mention of yellow highlighted that to me so thanks ). I think I might have to rethink what I want to do and perhaps put virtual masts in between physical masts to hide the yellow.

--
Bob Jacobsen
rgj1927@...

开云体育

Hi, found the XML files I think. I have two :-?

Hi James,?
Bob's option number 2 would be the more complicated ways of going about setting up your signals.? Implementing option one would not require changing any xml files.? All you need to do is to remove the signal mast pairs.? Delete any reference to SM2 in SM3's signal mast logic dialog box.? Only configure the signal mast logic to look at the block occupancy.? That way the signal is red if occupied and green if not.

Signal mast logic is designed to work in pairs.? In your example SM3 and SM2 are a pair, SM2 and SM1 are a pair, and so on.? SM3's aspect is based on SM2's aspect as well as B2's occupancy and configured speed.? Since SM2 was at stop, the only correct aspect for SM3 would be approach.? However if approach was removed from the possible aspects when setting up the signal, a danger state would be created because no other aspects would apply.? Danger is a special appearance usually mapped to stop which is why SM3 was red.

Might I inquire why you want to eliminate approach?? Once upon a time I could understand why one might want to eliminate aspects due to hardware limitations and difficulty in configuration.? Today though it seems like you would have to work harder at eliminating aspects as compared to just using them.? With the variety of affordable products like NCE Light It decoders or Team Digital SHD2s and pre-built block signals, it is practically turn key to get three aspect ABS signalling up and running if the detection is there.

Greg McCartney

Greg wrote: "Might I inquire why you want to eliminate approach?"

For those who have an interest in prototype signaling, let's take a quick trip to "Prototype Signaling Class". Those uninterested may ignore the remainder of this post.

Here's one prototype-based reason: It turns out that in early prototype signal systems, some signals could only display two aspects. In order to get a third aspect, one needed to add a second two-aspect head to the mast. The early so-called "banjo"-style signal heads, including the "Hall Disc" signal, typically could only display two aspects. See for a description of the Hall Disc signal. That page reports that Hall "Disc" signals were in use from the 1870s to the 1950s, and includes some photos of this type of signal still in-use in the 1940s.

The web site states that "Disc"-type signals were the most common signal type until the 1910s when the number of installed semaphore-type signals exceeded the number of "Disc"-type signals. In some cases, two-aspect Disc signals were replaced by two-aspect semaphore signals.

Some (later) Disc signals were able to display three aspects. I have a 1903 railroad rule book which shows aspects used in interlocking applications where one "Disc" signal had three aspects - Red ("Stop"), Green ("Caution"), and Clear ("Proceed" ).

Regards,
Billybob

Thanks again for the replies.

Confession time. I am very new to the world of Trains and model railways and know very little about it at the moment. Having just retired, I decided to teach myself electronics and was introduced to this world by my father-in-law. To this end I have designed and built all my own hardware which is controlled using I2C driven from a raspberry pi (JMRI jython scripts talk to python code, which manages the hardware and feeds back to JMRI). In my eagerness and ignorance the boards that I mage that manage the signals are for dual aspect only (and I have dual aspect signals). In the long term, as I learn more, I will redesign these for three aspect signals. Bob's option 2 worked well for me and only required one word change in the XML file and a little change in my python code.?

I tried your method this morning and it also worked so thanks - learning all the time.??

As a matter of interest when going with Bob's option two I thought I would need to change the? xml file in?/home/pi/JMRI/xml/signals/basic/appearance-one-low.xml when I thought it would have been the one in??
/home/pi/.jmri/TestTrack/JMRI/xml/signals/basic/appearance-one-low.xml.

Thanks to everyone for their help.?

Jim

Jim,

Your local path should be:
/home/pi/.jmri/TestTrack/xml/signals/basic/appearance-one-low.xml

That JMRI was extra for the path. That way you don't have to touch the other
path which will get replaced when you apply any JMRI update.

-Ken Cameron, Member JMRI Dev Team
www.jmri.org
www.fingerlakeslivesteamers.org
www.cnymod.org
www.syracusemodelrr.org

Thanks that has sorted me out.?

Jim

That's awesome Jim.? I wish I had more time to devote to learning electronics.? I kind of figured you must be building your own equipment when I saw the two aspects.? I picked up a bunch of ardiunos and used them to set up C/MRI which was pretty neat.? It was a cobble job but still fun.? Good luck with your project.

Greg McCartney

sorry, another question. All is running well now. The train does what I want it to do and the signals all change as expected. With the signal mast logic running it is clearly managing the masts which is great. Is there a way to set a signal to Stop by clicking the signal on the panel without the signal mast logic immediately overriding it. I can set it to held from the signal mast table and test for that condition (getAspect) but wanted to check if there was a better way to do it from the main panel.

Effectively I am trying to run the train but want to be able to stop it manually by setting a mast to Stop from the main panel and testing for that condition.???

Thanks Jim

Jim,

The 'held' concept is intended for the case of holding a signal to stop and
override the normal signal logic. Your use is what it is intended to do. The
other way would be adding some sensor to the mast logic that would also set
the stop, just like the block occupied does.

-Ken Cameron, Member JMRI Dev Team
www.jmri.org
www.fingerlakeslivesteamers.org
www.cnymod.org
www.syracusemodelrr.org

Do a control-click or right-click on the icon, and see if you can set it so that clicks set held/not-held. I forget the name of the option, but it should be in the context menu that pops up.

Bob

On Dec 21, 2019, at 12:06 AM, James Anderson <james_anderson_999@...> wrote:

sorry, another question. All is running well now. The train does what I want it to do and the signals all change as expected. With the signal mast logic running it is clearly managing the masts which is great. Is there a way to set a signal to Stop by clicking the signal on the panel without the signal mast logic immediately overriding it. I can set it to held from the signal mast table and test for that condition (getAspect) but wanted to check if there was a better way to do it from the main panel.

Effectively I am trying to run the train but want to be able to stop it manually by setting a mast to Stop from the main panel and testing for that condition.

--
Bob Jacobsen
rgj1927@...

Jim,
You need to change the action on click from changing aspects to hanging the held status
Bob Bucklew

Thanks.?