Much cheaper servo control for points

[Robin2]

 

I would suggest the first thing to do  is mock up a semi functioning web interface , i.e. press buttons, set routes or whatever and evaluate how that works.  Then you can add the backend and with that a suitable data store 

I share most of the sentiments you have expressed but I am not completely comfortable with the piece I have quoted.

 

Perhaps it depends what you mean by "semi-functioning". To my mind that must include the ability of the server program to respond to the interface and that, in turn, means that some substantial part of the "backend" is required - though not anything as formal or complex as a "data store".

 

At the moment it seems to me that @Grinkle has managed to produce a pretty HTML/CSS picture that only functions as a picture because it is not linked to a server program. I have suggested to him several times that the Python Bottle web framework makes the development of the linkage between the HTML/CSS picture and the server program as easy as it can be.

 

...R

[Junctionmad]

What I mean by semi functional is there is no need to build the subsystem to get the servos actually working

 

Ie you can simulate that side of it with test stubs etc. The key is to build a working model of the visual interface.

 

This let’s you prototype the “ working “ interface and will generally help you determine the back end requirements and clarify the storage issues. You will also have to decide how to support concurrent access and serialising access to the servos

[Robin2]

What I mean by semi functional is there is no need to build the subsystem to get the servos actually working

 

Ie you can simulate that side of it with test stubs etc. The key is to build a working model of the visual interface.

 

This let’s you prototype the “ working “ interface and will generally help you determine the back end requirements and clarify the storage issues. You will also have to decide how to support concurrent access and serialising access to the servos

I think that is also what I have in mind.

 

My concern is that @Grinkle does not see it like that - or not yet.

 

...R

[slngshotx]

Seems like a lot of people have a downer on this project! I'd not worry about a specific design methology for this if you don't want to. I'm still a fan of fag packet designs!

 

I know you don't want to use Arduino but i recently discovered wemos D1 which may be a good fit for this project. It has an onboard WiFi module that can run a simple web server, and can run simple html code. And they will cost a couple of quid.

 

I've got one running as a WiFi throttle on a DCC layout.

Edited May 21, 2018 by slngshotx

[Robin2]

Seems like a lot of people have a downer on this project! I'd not worry about a specific design methology for this if you don't want to. I'm still a fan of fag packet designs!

To my mind the biggest obstacle has been the OP's unwillingness to learn.

 

You can design the appearance on a scrap of paper (no fags, thank you ) - I don't have a problem with that - but the OP needs to face up to the creation of the server code, whatever platform he chooses. IMHO using Python on a PC or RPi will be easier than C++ on a Wemos - even with the help of the Arduino IDE.

 

 

...R

[slngshotx]

whatever platform he chooses. IMHO using Python on a PC or RPi will be easier than C++ on a Wemos - even with the help of the Arduino IDE.

 

Nooooo there is a circle of hell reserved for Linux, it should have died before the end of the last century.

 

Just thinking there are a lot more examples and tutorials for doing this sort of thing using wemos. I've never found a pi a satisfactory platform to control hardware, gpio support has many issues.

 

Anyway just my ramblings, and would encourage anyone wanting to do stuff like this to have a go, the hardware is cheap enough and while the learning curve is steep, there are huge resources all over the web for any chosen platform.

[Grinkle]

The solution is simple whatever hardware you choose

Me being think is the problem here

what I used to do easily I seem to have a mental block on

I know how I would like it to work (to make it simple for everyone)

It's not just making the hardware work but also making the user interface really easy

 

There is other hardware that can be used but I know the RPi can run a web server and act as wireless access point (and I already had one)

[Robin2]

Nooooo there is a circle of hell reserved for Linux, it should have died before the end of the last century.

So, should everyone use Windows?

 

 

Just thinking there are a lot more examples and tutorials for doing this sort of thing using wemos. I've never found a pi a satisfactory platform to control hardware, gpio support has many issues.

If you need better control of I/O just couple an Arduino with an RPi

 

 

More important than your prejudices or mine is the fact that the OP is familiar with the RPi.

 

...R

[Grinkle]

the RPi has short comings when it comes to gpio but these are not relevant when only using the i2c bus

[Robin2]

the RPi has short comings when it comes to gpio but these are not relevant when only using the i2c bus

What will be connected using i2c ?

 

...R

[Junctionmad]

The solution is simple whatever hardware you choose

Me being think is the problem here

what I used to do easily I seem to have a mental block on

I know how I would like it to work (to make it simple for everyone)

It's not just making the hardware work but also making the user interface really easy

 

There is other hardware that can be used but I know the RPi can run a web server and act as wireless access point (and I already had one)

 

yes but its GPIO is poor and cumbersome and Linux isnt the greatest thing for an embedded solution 

[Grinkle]

sorry for being so slow

considering we are  only 2 web pages, 3 data files and a bit of python script away from a functioning system I am getting frustrated

I am working on it but things here are dificult

[Robin2]

 

I am working on it but things here are dificult

Based on that it is a bit difficult to know how we can help.

 

...R

[Grinkle]

here is my idea for the settings sub page

this will tell the system which pwm boards are connected

settingssubpage1.html

 

It's not complete yet

the first two columns are only labels the third column (switch) tells the system if the board is present

and the forth column is just a reminder to the owner/operator

 

here is the data that will need to be stored

 

settingssubdata.txt

 

the first column is the id number of the board (0-61) used as a multiplier on the settings page

the second column is the address of the board in hex (it is just text) to display on the settings page

these first two columns never change

 

the third column (true/false) tells if it should be displayed on the settings page (this is the third column from the settings sub page)

the forth column is the forth column from the settings sub page to be displayed on the settings page

 

I have put a couple of examples in the data file

[Grinkle]

still working on it

here is how it is going

settingspage1.html

[Robin2]

still working on it

IMHO it would be a better use of your time to leave the HTML as it is for the moment and concentrate on learning the part that you don't know - how to link the actions on the browser to the code that will control the servos.

 

...R

[petejones]

This looks excellent. I am going to try using my recently purchased Raspberry Pi to control my new layout.

 

Cheers

Pete

[DCB]

Having read 7 pages of comments I have  decided to try operating my outside layout points with servos.  Operated by switches, apart from the middle and top loop which would be better with a battery and Radio  Control..

For the future if I do go computer control of points and trains I will be able to run trains in the train shed and watch them on web cam even when I am hundreds of miles away. Or better still have it work autonomously even when no one is aware it is working.