Folks using Nemo in a makerspace environment?

108 views
Skip to first unread message

Nicholas Bentley

unread,
Aug 19, 2024, 4:01:37 PM8/19/24
to nemo
Hello all,

We (University of Maryland College Park) are curious if there are any Nemo users out there using the platform in a makerspace environment. Oddly enough, this is where our first use case resided and since then it has been adopted by our Nanocenter as well, which I believe is more aligned with Nemo’s audience.

We’ve found that many of the features also work well in our makerspace environment and are looking into implementing more advanced functionality, particularly in tool lockouts as we grow our use. We would love for others to share their experiences implementing Nemo’s more advanced functionality in their spaces, if you are willing! Maybe there is an opportunity for a group Zoom in the future?

Lastly, Mathieu — as always — thank you for supporting such a great piece of open source software and for establishing this community. We hope to be able to contribute code directly in the future and are actively working towards this goal.

Best,
Nick


Nicholas Bentley (he/him)
Sr. Systems Engineer
A. James Clark School of Engineering
University of Maryland College Park


THE WORLD GOES TO MARYLAND

Travis Massey

unread,
Aug 19, 2024, 8:26:57 PM8/19/24
to Nicholas Bentley, nemo
Hi Nick,

LLNL is using it for both. Our facility has a nanolab cleanroom at its heart, but we also use the same NEMO instance to manage all of the non-cleanroom tools in the building, including those you’d find in a maker space. It works well. I’ll add also that we’ve found value in “tools” that aren’t tools, allowing users and staff to create tickets related to materials, computers, appliances, furniture, lighting, wishlist requests, utilities, you name it.  Granted, we use NEMO primarily (for now) as a ticketing system and info repo, and we underutilize some of its access control and billing features. It meets our present need and is equipped to grow with us. 

One feature we’ve found valuable for 3D printers and the like is the ability to track filament/resin consumption with post-usage questions, and to track what’s installed with configurations.  Not sure if this fits your definition of advanced. 

My understanding of interlocks for tool access control is that they almost all work by cutting power to something — a display, a specific power supply on an instrument, etc.  There seem to be two standard/accepted types: NCD.io ProXR and WebRelay from ControlByWeb.  Other vendors’ options are unlikely to ever be allowed on our networks. The former can handle larger loads, but the latter is UL listed (hard requirement for us) and can do PoE, so I’m planning on prototyping the WebRelays later this year.  Happy to compare notes when we get there. 

Best,
Travis Massey

--
You received this message because you are subscribed to the Google Groups "nemo" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nemo+uns...@list.nist.gov.
To view this discussion on the web visit https://groups.google.com/a/list.nist.gov/d/msgid/nemo/CA%2Bo2QCXDtwdGzYN60ba9p%2Bs%3DNGAVL0eKXZN%2BREKtozKsrTSxgg%40mail.gmail.com.

Nicholas Bentley

unread,
Aug 22, 2024, 2:02:48 PM8/22/24
to Travis Massey, nemo
Hi Travis,

This is very insightful, thank you! I appreciate hearing how you're using it for 3D printer filament/resin consumption – that's an area we struggle with at the moment so this insight is timely.

We're playing around with the WebRelay from ControlByWeb and it seems to work well thus far. UL listed is a hard requirement for us as well. Would love to connect more when you start prototyping the WebRelays. We identified that they are great for most use cases where we're controlling a peripheral but some of our folks have expressed interest in doing even deeper integration into equipment (e.g. 5-axis CNC) and we're not entirely sure how we're going to handle that.

Best,
Nick

Nicholas Bentley (he/him)
Sr. Systems Engineer
A. James Clark School of Engineering
University of Maryland College Park


THE WORLD GOES TO MARYLAND

mathieu...@gmail.com

unread,
Aug 22, 2024, 4:04:45 PM8/22/24
to nemo, nben...@umd.edu, nemo, Travis Massey
Hi,
I just wanted to chime in regarding interlocks and ControlByWeb relays.
We recently published a hardware manual for NEMO that includes bill of materials and schematics for building boxes for those relays, which allow different ways to interlock tools. At the moment we only have one option to interlock keyboard/mouse but more could easily be added to this flexible system.
There is also contact information for a company that can build those for you as well.

You can find that information by going to https://nemo.nist.gov/public/ and selecting "NEMO Hardware Manual" or clicking on this direct link: https://nemo.nist.gov/public/NEMO_Hardware_Accessories.pdf

Let me know if you have any questions

Thanks,
Mathieu

Travis Massey

unread,
Aug 22, 2024, 9:49:35 PM8/22/24
to mathieu...@gmail.com, nemo, nben...@umd.edu
Mathieu, this hardware accessories document is a fantastic resource, thank you!  This will cut down the prototyping effort considerably. 

Nick, if the CNC mill can handle having its controller rebooted every time someone uses it, I’d wire its power supply through the web relay.  Or perhaps you can splice into the power to the spindle if it’s a small one. If it’s a large mill, WebRelay units are meant to be DIN mounted, so it may be able to go inside the same UL enclosure that all the other DIN rail mounted electronics live in.  A third option would be to use the interlock to deliver a specific voltage to one of the limit switch input pins on the controller, so the mill thinks it’s continually up against a hard limit and refuses to move.  Along similar lines, the controller may also have a dedicated pin for a kill signal (a stop-motion-and-hold-all-motors-still-with-full-torque button can sometimes be preferred over an emergency off button to kill a CNC if no human flesh is in danger). 

Best,
Travis

David Barth

unread,
Aug 23, 2024, 11:05:38 AM8/23/24
to Travis Massey, mathieu...@gmail.com, nemo, nben...@umd.edu
Hi all,

I've done a lot of messy one-off interlocking with various kinds of hardware, so I'll add a few thoughts. One common and pretty straightforward strategy is to use a dry contact relay to interlock something like a door sensor. In the case of a CNC, there is likely a door that must be closed for the tool to run. Interlocking the door is essentially just adding another switch so the tool thinks the door is open unless the interlock is closed. It's not hard to build that into the sort of box that the hardware accessories document describes. You can also do a much simpler implementation with off the shelf hardware like the ones below that just plug into the wall. I've used the Web Power Switches with NEMO before and they're about the quickest and simplest possible way to get started.


Best,
David

Sergi Lendínez

unread,
Aug 27, 2024, 10:40:33 AM8/27/24
to nemo, dsb...@seas.upenn.edu, mathieu...@gmail.com, nemo, nben...@umd.edu, Travis Massey
Hi all,

I just want to mention the solution to interlocking we use here at LSU which is, as far as I know, one of the cheapest around. 

We are using a raspberry pi (~$30) to listen to the TCP address and port to which NEMO sends the interlock signal order. Currently we are using the "Stanford" type but, as far as you know how to decode the signal (and this is easy since the code is available on the interlocks.py file), any type can be used; I have used the Modbus type, starting a Modbus TCP server on the raspberry pi, and it works without any issues. Once the signal has been decoded by the Raspberry pi, we send it to an Arduino (~$30), which has an 8-Channel Relay Shield (~$22) that sends a 24 VDC signal to activate the tool. Depending on the tool, we use that signal to activate power via another relay, or sometimes to close the contact of the interlock chain of the tool in question.

Adding in the power supply, some cables and an enclosure, that's probably less than $200 to interlock up to 8 tools.

Best,
Sergi
Reply all
Reply to author
Forward
0 new messages