Locomotive Elevator

Posted by: @will

↑

The lead screw should be installed with a thrust bearing to support the axial load and a gear or belt system to link to the stepper motor. That way there's no axial load on the stepper and it won't grind internally.

Ok... so we're accomplishing the same thing by two different methods.  My suggestion was the pillow block take all axial load, and the spiral collet only transmits torque.  I believe, Phil's rig accomplishes the same thing.  The lead-screw rests on the hard floor taking all the axial load while the stepper is attached by the hopefully a spiral collet, taking no load.

Posted by: @byron

↑

It would seem that it would be best for the platform to either have the lead-screw in the centre, one at each end, or maybe to have one in each corner (and four motors of course) and do away with the rods supporting the four corners. 

The current design and the code to support it assumes one at each end.  I'm encouraging him to follow up using some kind of level sensor like you suggested... oh... about a hundred pages back to error out if one stepper starts loosing steps.  He already has the safety factor @will suggested about wiring such that if any wire breaks, the whole thing also stops.  I'm also encouraging him to only set the torque of the motors to only be able to lift the elevator, not to scissor a train car or finger.  That we want it to miss steps.  IOW a built-in load limiter.

@inq This is exactly how it is, as per my posted picture above. I have ordered a couple of the DRV8255 items you mentioned in the post above when they get here I will give them a try and report back. Hopefully they will give it a bit more oomph! When I tested the motors on the bench I was a bit afraid of grabbing them too hard and thought they were more powerful than they actually are, I tried them up the train room today and I can actually stop them with my hand so maybe I should have seen this sooner.

Cheers Phil

Hi Everyone, just a quick update to let you know the status on the project. The 8255 units came and I just plugged them straight into my home made circuit board, and .... the smoke came out! I should have checked first because the pinouts on them was different to the 4988 units, anyway destroyed two of them and the ESP8266 unit as well. So moving on I bought a couple of these,

https://www.ebay.co.uk/itm/314730981733?hash=item49476da965:g:pWMAAOSwmAFkvjU8&amdata=enc%3AAQAIAAAA4OU6OqBGFXcpO9gxNnJdIbxqzoaA0wmuCWsqzF4I5Bmr9kt1Jdmv6WOVCj%2FjzLhnXrJLE5aYc44Nw4yARVyugJYYcn9GUHryu2U7OYMybhGPeVJwx7IA1QtcGcjqoMFsmHZ%2BMGJSr4KOwTigrwrRQEdu%2FB3yY7EAk2T1Bbg1JjAtJGF%2Bs4i3ZO8AUuFlIGm2h%2BtIe37hOeylIZFRKh33VP3Qe0RleV%2BpCAXDBziF%2FaXXBdRPfGGJUh7NFkXbXilQ6bMzSVKDVrtO%2BZfGGh%2FmDhLZpRqmVwqHnj3iZNSYuvei%7Ctkp%3ABFBMwNaUucJi

wired them up and they are fantastic, no slipping now or skipping steps, got them set at around the 2 amps position using the dip switches, and the real bonus is they work directly on the 3.3v outputs from the new ESP8266 unit, so no more making up circuit boards, waiting on some aluminium material now and then I will post another video so you can see it running.

Cheers Phil

@thephilnewman 

Sorry to hear about the untimely demise of your drivers, but I'm glad to hear that you've found even more suitable new ones and have progressed to the next phase of your implementation.

Posted by: @thephilnewman

↑

Hi Everyone, just a quick update to let you know the status on the project. The 8255 units came and I just plugged them straight into my home made circuit board, and .... the smoke came out! I should have checked first because the pinouts on them was different to the 4988 units

I'm glad you found something you are satisfied with, but this statement is flat out wrong and misleading to anyone else that might read it.  A4988 and DRV8255 are PIN for PIN compatible.  They are replaced universally in 3D Printers all the time.

@inq 

I've almost blown up several of them because the drivers have the pinouts on the bottom and so I've looked at the bottom and then spun it 180 degrees to mount it and almost put them in backwards because I didn't account for the rotation 🙂 I now have to check my plasticized pinout chart instead of using the markings on the PCB 🙂 

Posted by: @will

↑

@inq 

I've almost blown up several of them because the drivers have the pinouts on the bottom and so I've looked at the bottom and then spun it 180 degrees to mount it and almost put them in backwards because I didn't account for the rotation 🙂 I now have to check my plasticized pinout chart instead of using the markings on the PCB 🙂 

I think that falls under being a positional joke? 🤣 You had to be there.  And you had to look in the mirror at the same time while standing on your head.

@inq 

In the 'old days' when I used to do a lot of FR4 PCB and wiring from underneath, I'd take a picture of the parts laid out on the top, then take flip that picture right-to-left. That left me with a perfectly matched layout for wiring so I could more easily identify what pin belonged to what thingy.

@inq hello still working on the elevator, only it does not seem to want to connect anymore? do you have any ideas? I am using

http://192.168.1.41/Administration.html

Just noticed the 41 has now changed to 34 have tried that as well?

Cheers Phil

@thephilnewman I certainly can't speak for Dennis, but the normal strategy to eliminate that kind of issue is to use host names, not numbers. Either a hosts file or a DNS. In my case for instance, all my 192.168.x.x addresses have been changed to things like Pi4Dev1, PiDev2, MiniWin, NAS1, NAS2 etc now they can change address all they want, but your code stays the same.

@zander hello I dont really know enough about all this but the thing that I have noticed that has changed and not for the first time either is the http://192.168.1.??/ where the two question marks have been 47 then 41 and now are 34 like I said I dont know enough about this but it is messing me up when I am trying to get things to work on the elevator because the address has changed and I have no idea why. This time where I spent so long looking for the reason the pages would not load and trying different things I have now lost all my elevator settings and I am in a bit of a pickle trying to get things back how they were.

Cheers phil

@thephilnewman It is nothing unusual to have addresses change that are not 'nailed down' by one of the mechanisms I mentioned. I am sure Dennis will know the best answer for his setup, I was giving you the basics. If nobody involved is not network knowledgeable, I am afraid you will have ongoing issues unless you do something about it.

IIRC, the code you are using from Dennis is a software router, amongst other things. AFAIK all routers have the ability to hard code a hostname/mac relationship under something like 'reserved addresses.'

I live in an old age apartment, and everyone has one or more routers where the addresses come and go. At the moment, I have 19 devices connected, and none are named 192.168.x.x; they have names like Living -Room or Rons-Air etc.

@thephilnewman Is it the address of the esp8266 that keeps changing? If so, is there not an Admin page in the Inq App? I would be surprised if there was not a Hostname entry on the Admin page.

Looking at his code, parameter #5 of the svr.begin call is the Hostname. Put something in there like TrainElevator but that may be too long, so maybe just Elevator?

@thephilnewman 

admin.html