HandCab WiFi Throttle Build...........

[Sumner]

This is a WiTcontroller throttle similar to one I built ( HERE )  but with a stalk and controls on the top of it that represents the 'barrel' control stand that was found in GP7/GP9 cab's. I have a number of GP7'S and GP9's and after looking at....



…. the control stands in them decided that ….



… I'd work at copying them to some degree.

 

I couldn't copy them exactly but went for a 'kind of looks like' one approach. I could of gone somewhat smaller but needed the room for the pots I was going to use and didn't want mini-controls on the throttle.



Since it will be 3D printed it won't weigh all that much more than the one on the right and it is very light. The control stalk raises a touch over 3 inches above the top of the case. Looks like more in the images.


I wanted the control handles to feel like you are actually using something larger than what you use with most throttles. I won't know if that was the right move until I get this printed and together.

 

I am about a day away from printing what is shown above. If I'm happy with that I'll modify the bottom of the case for the other one to fit this one.

This project is still a long ways from being completed. I'm hoping to use all the parts that are used in the WiTcontroller throttle with the addition of the 3 pots. The code for that throttle, written by Peter Akers (EngineDriver), will work but not for the three levers/pots as written. I could write some of that but for sure couldn't add it to the rest of Peter's WiTcontroller code.

I'll report back when there is more progress.

Sumner

[lyncher]

Sumner, I've been working on a diesel-centric withrottle as well.  I've prototyped the controls which includes potentiometers for throttle and brake. The throttle uses a pot with indents that gives me the 8-notch + idle function. Brake is regular pot but the software gives me 4 steps + off. I've been focused on the software at this point, the hardware is in the breadboard stage though I do have a 4x4x2 utility box holding everything, including a small OLED screen.  Re: the software.  I'm using a RaspberryPi PICO-W board and writing everything in Micropython, which is going remarkably well. I'm a day or two away from a full test. But, a family emergency has just come up that I have to deal with.  I'll be able to respond to any messages from you - I'll be traveling - but testing my throttle will have to wait until I return in perhaps a week.

I am really intrigued by your design and would love to collaborate. If you're not interested in using or testing my PICO-based software, I understand, but I'd still like to test your cab design with my software and PICO-W board.   If you're interested, let me know.

-lyncher

[Sumner]

The first print of most of the top part of the throttle. Before finishing up the design of the case top, middle and bottom I thought I'd better see how I liked the control stand at the top. Designs can look quite a bit different on a computer vs. printed and in your hand.



I need to make a couple tweaks to the design but minor ones as I'm happy with how this is coming out. Somehow I printed two brake handles so the reverser handle that is shown needs to be printed.



The controls work well and I like the feel of them. The throttle handle has 9 positions total, idle and 8 throttle steps. I designed it with a ball bearing in a round channel pushed by a ballpoint pen spring and a set screw for pressure adjustment. Works as good or better than I hoped it would. Easy to move the handle from detent to detent and easy to adjust the feel.



The brake and reverser have no detents but also have fewer positions so not the need for such precise positioning.



The reverser at the bottom is either centered or to one side or the other so it will be easy to read the pots position and know where the lever is. The brake will be either off or a couple positions before being full on, so more of a 'feel thing' vs. precise steps like the throttle needs to be.



All of the controls only read about 120 degrees of the pots travel/resistance. I believe that is plenty even for the throttle to determine where it is and what to do based on that position. I put the multimeter on it and the throttle is very repeatable for the detents. I won't use the exact reading at any detent position but a range on both sides of there.


Will write some simple code next to read the pots and if all is good after that print the rest of the case out and the top for this section of the throttle.

A link to the build on my website ( HERE ).

To be Continued …...

Sumner

[Sumner]

...... but I'd still like to test your cab design with my software and PICO-W board.   If you're interested, let me know.  -lyncher

Yes, real interested in what you are doing.  Hope to get Peter's WiTcontroller software working on the throttle as I've already build two throttle using it.  He will have to add code to read the pots.  I've done that once with my turntable project( https://1fatgmc.com/RailRoad/Turntable/Turntable-Index.html ) but would not have any idea how to add it to his current code.  He has expressed interest when time permits for him.  His software is based around an ESP32.

If you are interested in the print files I could send them.  I'll probably have the rest of it printed out by this time next week and the files pretty much finalize.  I'm about done with the rest of the case as I mainly using the same basic case I'm using on the WiTcontroller throttles I've already built.

https://1fatgmc.com/RailRoad/DCC/WiTcontroller-Index.html


Eventually if this all works I'll put them up on thingiverse.com.

Sumner

[Mike C]

@ Sumner , I an guessing with 9 notches you will want to use a lot of momentum to smooth out the accel / decel speeds ?

[Sumner]

Yes, one notch is idle.  Some of the people that use the Proto Throttle have momentum set to where it takes about 25 seconds or more from idle to notch 8 (full throttle).  Set deceleration where you have to use the brake basically to slow down.  Not hard to do with ESU and a couple other decoders but not easy to do with a lot. 

I'm working on a way where the throttle software handles that not the decoder.  Early stages of that.  Just sent the ideas to Peter who would have to implement it.  I'm using ESU so not a problem for me but think that the option to set momentum in both directions via the throttle would be nice for the other throttle (  https://1fatgmc.com/RailRoad/DCC/WiTcontroller-Index.html  ) also that uses an encoder.

Sumner

[freedj]

What a cool project!  I built an ESP32 wifi throttle based on geoffb made over on MRH.  How does the brake pot work?  Do the decoders support multiple brake values, or does the throttle essentially use PWM and lots of momentum to smooth out the decel?  How does that work with sound equipped locos?

[ Guests cannot view attachments ]

[Sumner]

....  How does the brake pot work?  Do the decoders support multiple brake values, or does the throttle essentially use PWM and lots of momentum to smooth out the decel?  How does that work with sound equipped locos?

(Attachment Link)

Working with Peter Akers on this for the software.  He is trying to decide if the software additions for the pots will be added to his  WiTcontroller software
or maybe best to use some of it but make this a new project.  Braking with the Proto Throttle as far as I know works with ESU's brake function and a TCS Wowsound function.  Both differently.  Not sure if you can have braking with it with any other decoders.  Anyone have info on that?  Since their throttle and this one uses 8 notches + idle for speed control momentum one way or another is needed to smooth things out.

We hope to build acceleration and deceleration into the throttle so that there will be momentum with any decoder.  I think I have a way to implement that fairly easily just waiting to hear back from Peter on his thoughts on what I sent to him.  I'll post a video later on how the throttle part hopefully will work.  The brake will be somewhat the same, but will only work on slowing the loco down.  The pot will be read and either the whole range of resistance values will be mapped to slow the loco down with very little stepping or there will be steps like the throttle and momentum to take the steps out.

I'd assume the throttle will work fine with sound decoders.   Throttles built around Peter's WiTcontroller software work great and you can see all the functions on the display and trigger any of them.

Sumner

[Ed Kapuscinski]

This is rad

[Sumner]

Put up a video of testing how repeatable the resistance readings are on the pot connected to the throttle handle on the HandCab throttle..

....  Turned out they were very repeatable in ....



... both directions moving up and down through the 8 speed notches (detents).

More info on the throttle build to this point here......

https://1fatgmc.com/RailRoad/DCC/page-7.html

 and a link to the YouTube video here......

https://www.youtube.com/watch?v=3t_6mOx_JVQ

Sumner

[lyncher]

Sumner,

Any progress to report on the HandCab?

I've forked my (micropython) code base and added the ability to read a keypad, which I did not initially include in the design, so it should now be compatible with the HandCab. I had originally used a 5-way navigation switch in lieu of a keypad. Regardless, I'm still very interested in your physical design and would love a chance to try my code with your HandCab. If the files for printing a copy are available, I'm ready to give it a try.

-lyncher

edit: meant to include a link to the 5-way switch I'm using:  https://www.adafruit.com/product/504

[Sumner]

Any progress to report on the HandCab?

Yes, some...been mainly working on the software.  Peter will be also working on that when he is off vacation and catches up with other projects.  I'm done with the software that will capture the resistance values for the 3 pots and pass that on in a way that his software can see along with some that I've started to write.  I  am working on another part of software that hopefully lets the throttle control momentum vs. the engines decoder.  Having problems getting data from the first program to the other.  I had it yesterday after a ton of time trying to learn how to do it.  Then the file got screwed up and I had already forgot how I had done it.  I programmed a lot with Basic 40+ years ago but then nothing until writing/coping/borrowing code for the Arduino's for the turntable  project 2 years ago.  Now I forgot about all I learned there so starting over again.  The syntax part is killing me.

So after your post above I spent the day on the throttle putting it together.  I'd printed all the parts a week or so ago but hadn't assembled it so wanted to check fit before putting any print files up.  It all ....



.... seems fine with no fitment problems so if you want to try them let me know.  I might not put them up on thingiverse.com at this point unless there is more interest so PM me with an email address to send them to if you still want them.  I'll be interested in what you do with them.  Are you running an ESP32 (I can't seem to find your project---have you posted links?)?

I'm editing images now for assembly and hope to have them up later today/tomorrow.

Sumner

[Sumner]

HandCab WiFi Throttle – Part 3...

Assembly images......



.



.



.



.



.



.



.



.

https://1fatgmc.com/RailRoad/DCC/24-04-01%20Ohm%20Test-9.JPG

.

To be Continued …...

For this whole build go ( HERE ).

Sumner

[samusi01]

@Sumner

In lieu of tapping, consider using Fusion's built-in tools to generate threads as part of your print. Bryan B. pointed this out a while back and it's been a huge assett in my various flatcar models. For 2-56, this is what I do:

[ Guests cannot view attachments ]

Hopefully the settings on the right are legible. If not, hole tap type is threaded, and I use #2 side, 2-56 UNC, class 2B, right hand threads. I found this to be too tight on my Elegoo Saturn so after some experimenting I have settled on the following step: select all the faces of the thread (four, as depicted below) and offset them 0.0016".

[ Guests cannot view attachments ]

[Sumner]

@Sumner

In lieu of tapping, consider using Fusion's built-in tools to generate threads as part of your print.

Thanks, I had experimented with that some without much success.  It is easy to create them with fusion but I wasn't having great luck getting them to come out consistently.  Maybe the resin or the printer or ???.  I might give it a try again.  The threading goes quickly and I'm not doing lots of these so it hasn't been that hard or time consuming to do.  If it was something I needed to print lots of then for sure I'd try and get it to work.

I'll give the off-set a try next time.

Sumner