Need Control Box advice on my 2007 ShopBot PRS conversion

[ShawnM]

Layout looks good but tight. Rule of thumb when building and enclosure, layout everything the way you like it, measure for the cabinet size and then buy the next largest one.

Not sure what that backplane is made of but if it's plastic you have to ditch it for a metal one. If that one is metal take it out and hit it with a palm sander, BOTH SIDES, to remove all that white paint. This is a GROUND plane and should be bare metal.

[cncsnw]

Definitely looks tight.

Remember that terminals need to have wires coming out of them. If you have terminals butted up against the wire duct, then any wire whose terminal does not happen to line up with a slot in the wire duct, has no place to go. And for those that do line up with slots in the duct, you will be unable to see and read the wire marker (the number tag that you put on the wire) because it will be hidden in the duct.

I try to have a minimum of 1.5" clearance between terminals (whether DIN terminal blocks, or the green Euroblok connectors on the edges of the boards) and the wire duct, and preferably a full 2" clearance.

[djbeede]

Layout looks good but tight. Rule of thumb when building and enclosure, layout everything the way you like it, measure for the cabinet size and then but the next largest one.

Not sure what that backplane is made of but if it's plastic you have to ditch it for a metal one. If that one is metal take it out and hit it with a palm sander, BOTH SIDES, to remove all that white paint. This is a GROUND plane and should be bare metal.
And thanks for the input!

I did create a lot more room by mounting my Geckos on edge. It doesn't feel crowded to me now? The mounting plate is metal, and it is painted. However, components are mounted to DIN rails or metal plates that are screwed to back plate with self taping screws. What would symptoms be if I were having grounding issues?

[Gary Campbell]

What would symptoms be if I were having grounding issues?

The list is varied and long. You'll probably be posting yours here in a week or two!

[djbeede]

What would symptoms be if I were having grounding issues?

The list is varied and long. You'll probably be posting yours here in a week or two!

Hey Gary - I failed to notice replies had gone to a 2nd page so thought I was on my own after 3 weeks of no response, my bad. Wasn't ignoring recommendations, just didn't see it.

But I am not very inclined to totally disassemble my control box and sand all the paint off the mounting plate without some information to weigh the costs benefits. I'm sure I missed some info, but I've studied a lot of what's on the forum and the Centroid site and this is the first time I ran into a requirement for bare metal mounting plate. I bought a version of the VEVOR NEMA Steel Enclosure recommended on Centroid's "gear we use but don't sell." I watched your box build video a few times Gary and got a lot of great tips from it. Here's something I noticed.

Is this plate painted or bare?

This plate your drilling sure looks painted to me?
So when I ask for the cons of a painted control board your
"The list is varied and long. You'll probably be posting yours here in a week or two! "
isn't nearly as helpful as a few examples of the repercussions of failing to mount components to bare metal.
I'm guessing like so many aspects of this endeavor, there are mixed opinions, so it can sometimes be challenging to find the best way through. Absolute necessities and strong opinions can sound a lot alike sometimes.

All that being said, I wouldn't be half as far as I am without the support of this forum. In fact I have doubts I could have completed my conversion without it. In my opinion it's the life blood of Centroid Acorn and I'm grateful it makes it possible for small shop guys like me to have access to industry grade world class electronics.

[tblough]

EMI is extremely hard to track down. It can manifest as random faults, positioning errors, burned out drivers, motors that appear to be connected correctly but still wont move, and a hundred other symptoms. It would be great if it were one or two repeatable things, and we could say just fix this. Unfortunately there are many causes, many symptoms, and many pieces to the solution, and those solutions take many man-hours to diagnose and attempt to fix.

Because of this, there are best practices that are employed to minimize EMI sources. Good ground planes and good electrical connections are one part of this. You may get lucky and your system may work and then you'll be able to post that you didn't bother with this or that and you didn't have problems and therefore it wasn't really needed and was obviously someone's strong opinion. Or, you could spend weeks trying to find a random fault.

The difference between absolute necessity and strong opinion is the cost of your time.

[cncsnw]

In my opinion (based on 20+ years of building controls that generally run reliably) there is no reason to strip the paint off the subpanel.

If the subpanel itself is grounded, and it is bonded to the enclosure, and the DIN rails are securely bonded to the subpanel via their mounting screws, and you have run ground wires to the ground terminals of your drives and powers supplies, then it is probably fine.

Personally, I hate self-tapping screws. I take the time to tap the holes and use machine screws. If you are using self-tapping screws, you are probably making metal chips in the cabinet while your components are present. Better to tap all the holes, clean up all the chips, then mount your components. But, your self-tapping screws probably make an okay bonding connection, at least the first time or two that they are driven in.


On open-loop stepper controls such as Acorn, one of the first and most common symptoms of noise issues is loss of position. Either the drives see steps that were not commanded, or they fail to see steps that were commanded. As you run the machine, one or more axes will drift farther and farther from the intended position. This forum is full of posts about that.

[djbeede]

Thanks Tom -
Though still pretty vague I hear you trying to express the potential importance of doing everything as in line with best practices as possible to minimize those potential future EMI headaches. That's my approach generally. I'm very close to the end of my conversion and it hasn't been a walk in the park by any means but progress is steady. Even with all my components wired up in a pile on a plastic side table, I was able to surface my spoil board and do some test Vcarve files. The kind of care I put into building the control box felt more than adequate, but now I'm being warned I missed something extremely important. I did a search of the forums for bare metal mounting plate and found 3 occurrences, all in my build thread.
From what I've seen it does appear that even people who build control boxes for hire don't always use bare metal mounting plates? It's hard to understand why something seemingly so critical is given so little coverage on the forum? Though I tend towards caution I don't tend to jump through hoops to solve problems I don't and may not have. And given that my box is fully built - it will take the exact amount of time in the future to take it apart and sand the mounting plate and reassemble it, as it would to do it today.
     Here's the thing. I've been building musical instruments for 50 years and using a Shopbot CNC for 20 years. SB had its com errors and I had some grounding issues to work out along the way, but generally pretty dependable with decent tech help. I know Shopbot is not that highly thought of on the forums, and I can understand why, but it was good to me and it paid for itself many times over.
     I found my way to Centroid because the ShopBot controls couldn't reverse direction to a spindle, which I need to realize my goal to implement the RapidChange ATC system. It looks like I'll be able to pull that off. I may be living some version of your "Confidence" quote Tom - though I can't say I've felt much confidence so far. Like life, it's a bit of a roller coaster. Sorry, I'm rambling a bit. I do know the folks on this forum are all trying to help. So thanks again for your input.     
Cheers...David
     

[tblough]

I don't use bare metal plates. Mine are all powdercoated. However, if the component I'm mounting is metal, I sand the powdercoat off down to shiny metal underneath at least one mounting screw. "Metal components" includes DIN rail, the Acorn mounting screws, ground connections, etc.

It's hard to be concise because the EMI problem has many causes. Are you having problems because of a painted backplane, using unshielded wire, ground loops, signal cables too close to power cables, noisy cheap components, lack of snubbers on inductive components, or some combination of these?

When you build one panel for a hobby, and you skip a few steps, you could have a 90% chance that it will work without problems. That's probably a good tradeoff. When you build 100 panels for a living, 10 panels that don't work could be the end of your business especially if the problems don't show up until the equipment is in the customers hands.

As a musical instrument builder, can you explain in concise terms why everyone can't build violins that sound like a Stradivarius? It seems fairly simple. Cut a few pieces of wood on router and glue them together.

[djbeede]

I don't use bare metal plates. Mine are all powdercoated. However, if the component I'm mounting is metal, I sand the powdercoat off down to shiny metal underneath at least one mounting screw. "Metal components" includes DIN rail, the Acorn mounting screws, ground connections, etc.

It's hard to be concise because the EMI problem has many causes. Are you having problems because of a painted backplane, using unshielded wire, ground loops, signal cables too close to power cables, noisy cheap components, lack of snubbers on inductive components, or some combination of these?

When you build one panel for a hobby, and you skip a few steps, you could have a 90% chance that it will work without problems. That's probably a good tradeoff. When you build 100 panels for a living, 10 panels that don't work could be the end of your business especially if the problems don't show up until the equipment is in the customers hands.

As a musical instrument builder, can you explain in concise terms why everyone can't build violins that sound like a Stradivarius? It seems fairly simple. Cut a few pieces of wood on router and glue them together.

Thanks for the added insight into the contrast between pro panel builders and a one off in a one man shop.

As for your Stradivarius question. The sonic qualities of wood are varied and elusive. It doesn't just vary by species, tree or grain direction, but within the same board. Any luthier that has ever built a pair of twin instruments from the same wood knows this. They don't sound the same. The variables that need managing are nearly endless. (Which keeps it interesting.) Science has gotten better at testing and tuning resonant frequencies but it's not the whole ball game.

As for your Stradivarius question, it contains a hidden assumption. That Stradivarius made magical instruments surpassing all others. He lived in Italy during a 2nd wave of the plague which took out some of his competitors, then he made a family of instruments for the royal family. Some say it may be PR. Marketing..?.!

In double bind listening/playing trials Stads don't fare too well. Here's a brief quote from an article on one of those tests.

"one of the most telling results in 2012. They had 21 experienced, blindfolded violinists play Stradivarius, Guarneri del Gesù (another old-instrument-maker of comparable repute) and high-quality modern instruments. Blind testing showed that the violin judged best by the players was a new one, and that the instrument judged worst was a Strad. And the musicians couldn’t reliably tell whether an instrument was new or old."

A sign on a fellow luthier's workbench reads:
"Stradivarius made NEW instruments!" As concise as I can be on my favorite subject!

Link to article
https://tinyurl.com/3hj7h63x