It’s been a few months since I got my Nomad and I wanted to give some feedback to the Carbide 3D folks but also share my experience so that others in a similar position to me who are considering the machine can get a realistic idea of what it’s like (also, if anyone has solutions for the problems I’m bringing up, I’m all-ears).
First, I’ll state my assumptions, because I think a lot of what I’m about to say is based on them:
- The Nomad is intended for people who:
- Do not intend to use it professionally in the manufacturing industry (i.e. this isn’t a replacement for a Haas).
- Prefer subtractive manufacturing over additive manufacturing because of the precision or flexibility of materials it offers, in particular, it’s ability to machine metals (otherwise they would buy a much cheaper and less complicated 3D printer).
- Choose the Nomad over the Shapeoko for one of the following reasons:
- It is more precise.
- It is more compact.
- It is more rigid.
- Are possibly newcomers to the world of CNC.
- Want something that is able to mill projects out of the box, if they wanted the machine itself to be a project, they’d do a CNC conversion.
- Will use the machine on a desk in a residential or office environment rather than say a garage. If a garage or other industrial workspace was available, a larger machine like a Tormach might make more sense.
Now, to the actual feedback.
I personally wanted to get into CNC because after using a 3D printer for a while, I became frustrated with the weakness of plastics and wanted the ability to make stronger things from metal. I was heavily influenced by ThisOldTony on YouTube.
I chose the Nomad because:
- I wanted a machine that worked out of the box. I didn’t want the machine itself to be a project (I considered other machines like a Sieg CNC conversion or CNC-converting locally available machines), I wanted something that would Just Work™.
- I live in an apartment and
- I wanted something that was small and compact enough to sit on the space I had on my desk.
- I asked around on Reddit and it was suggested that the machine was quiet enough to be used in an apartment.
- People said the Nomad was built like a tank and could potentially be upgraded later if I wanted (this is msot of why I chose it over Sherline).
- It was fairly affordable.
Despite living in Switzerland, I ended up ordering it from the US because EU pricing is disgusting, even with VAT removed, and because none of the EU sellers would ship to Switzerland.
The experience immediately after opening the box was fairly decent. I got my machine opened, loaded up the provided material and got my wrench cut without any problems. The biggest issues here were:
- The machine was loud. I wasn’t expecting it to make anywhere near that much noise. The noise levels to expect should be called out somewhere explicitly (e.g. “imagine a high-pitched lawnmower”).
- The Aluminium-bonded-to-black-gunk material got everywhere and stuck to everything it touched. I think there’s still some in my machine somewhere. Another material would be great here.
After that, I wanted to get straight into cutting Aluminium, so I used the double-sided tape to tape down a 12mm thick plate of it, put the feeds and speeds that I thought I read into Fusion 360 and let it rip! This is where everything started to fall apart:
- I didn’t understand that the feeds and speeds for the starter project weren’t appropriate for Aluminium plate. I crashed the machine into the stock and it made very loud, very scary noises.
- When I looked up proper feeds and speeds, all the ones provided for Nomad were Imperial and as a non-American, I only use metric and was a bit confused at first trying to convert them. It’d be great if the tables could be updated with metric conversions.
- I ended up opening up the MDI and doing very shallow cuts into my stock to trial-and-error my way to something that worked.
- The noise was still far too much. I had to do something about this.
At this point, controlling noise became my new project, as the machine just wasn’t viable in an apartment as-sold.
First, I thought it was just vibration from the bottom of the machine, so I added some anti-vibration foam pads. The desk stopped shaking but the noise was still horribly loud.
I Googled around and it looked like an enclosure was the best option. I checked other people’s builds and YouTube videos and settled on a box-in-box design made of MDF with noise absorbing foam on the sides and top and vibration-isolating foam pads on the bottom. This took quite a while to design and acquire materials for, as I’m in Switzerland and can’t easily just buy whatever I need from Amazon.
While I was waiting for the box, I started looking around forums (like this one) and YouTube. I discovered Winston’s fantastic videos, which gave me actual useful feeds and speeds for Fusion 360. I also read feedback that the ZrN-coated cutters were great and might be quieter, so I ordered a ton of them.
Once the box was finally assembled and my cutters arrived (a few weeks ago now), I tried out the machine in its enclosure and it was finally working and able to cut, but I wasn’t able to see into the enclosure since it was made of MDF. I added some LED strips and a Raspberry Pi with a camera and that problem was resolved.
I had a couple of problems with Carbide Motion apparently ignoring my zeroes but the forum helpfully suggested that that might be because the machine lost steps when I was zeroing and had no idea where it was.
And that brings us today. My recent problems are:
- Due to the size of the bed, it’s difficult to hold large, flat workpieces. I bought the threaded table and clamps but had no idea how to use them. A few weeks later I saw Winston use them in a video and finally figured it out and was able to hold my stock.
- Due to the limited stickout of the endmills and the thickness of the parts I want to produce, I’m forced to use multiple setups a lot and need to accurately re-zero. There isn’t a good solution for this:
- The flip-jig isn’t nearly big enough
- The paper method for zeroing isn’t precise enough
- Most of the off-the shelf options don’t work
- I’m yet to see a single “wiggler” edge-finder that’s compatible with the Nomad’s spindle speed range. There used to be one in the store I see but it’s long gone.
- More expensive alternatives like Haimer’s Zero Master have shanks that are too thick for the Nomad’s ER-11 collet.
- Due to the sound-isolating enclosure, the machine gets hot after a few hours of cutting.
- There’s no built-in way to clear chips.
- Chips tend to get everywhere, like on the lead screws and Y-axis rails.
And my next project is likely to be adding a fan or compressor to clear chips.
To conclude the feedback, I’m okay with the machine at this point but I don’t feel the machine as-sold meets the needs of the people I see it as being sold to (including myself):
- It’s too loud to be used in a residential/office environment.
- Despite the software and website appearing to be very beginner-friendly, aside from the starter tutorial, you’re kind of thrown in the deep end. There’s very little provided documentation on how exactly to use the thing. Perhaps the intent is that people will just open Carbide Create and use the default settings and maybe that works but I gave up on it very quickly because as a CAD solution, it’s just not very good. Fusion 360 is free.
- Features I consider basic essentials like chip-clearing are completely missing and you’re left on your own to provide them.
- The uncoated #102 endmill was (for me) prone to getting clogged with chips.
What I’d really like to see in order to resolve these issues is:
- Carbide 3D should provide, either stock, as an upgrade that can be purchased in the shop or at least as a recommendation listed somewhere:
- A relatively soundproof enclosure with some kind of cooling, made from a transparent material and/or with a webcam and lighting. It could be shipped IKEA-style. I imagine it’d end up being some Aluminium profile, some thick but transparent plastic sheets and some foam padding.
- Some kind of chip-clearing solution.
- Some kind of Nomad-compatible probe for zeroing (I’ve been eyeing this)
- Something to keep chips away from leadscrews and rails (e.g. covers)
- Winston’s videos should be featured prominently in the onboarding process, to help people get started.
- Realistic expectations for noise should be set prior to purchase.
- ZrN-coated endmills should be strongly pushed during the purchase flow, or one could be included in the basic package, or at least the starter set of endmills.
I’m sure I’ll get the machine into a state where it meets my needs in the not-too-distant future but I wanted to give this feedback so that maybe one day it’ll be able to Just Work™ out of the box, as I hoped it would.
.