[hoss2006]

The new free head spacer plans are available on my site now.
Projects
Hoss

[Signal9]

Hey Hoss,

How many snap locks did you use on your large enclosure?

[hoss2006]

10 iirc.
Hoss

[Signal9]

Thanks Hoss!

[diyengineer]

Just stopping by to say hi, see this thread is growing massive! Largest I think I've seen so far!!! Hope all is well Hoss.

[hoss2006]

Goin ok buddy thanks for asking.
Hoss

[tomdchi]

I'm about to dive into this whole process, and I got Hoss' full set last week, which is great, and he's been nice enough to answer some strategy questions as I'm plotting out the conversion.

I'm still working my way through this massive thread (and these answers are probably in there somewhere, but I haven't found them yet) but I thought I'd throw them out to the folks here rather than just bugging Hoss:

DC power for the drivers/steppers:

I'm currently planning on using NEMA 24/570 oz/in for X and Y and NEMA 34/906 for the Z, and the Keling digital drivers.

- Is there a consensus on using switched vs. unregulated toroidal DC power?

- Am I right in understanding that voltage primarily limits rapid speed, so if I'm not too worried about that, then 48v DC is OK?

- Does the consensus here support Gecko's assertion that the power supply's amp rating should be 2/3rds the motors' total rated amps? Does that hold for switched power supplies?

- Would it be a problem to use two or more smaller rated switched power supplies - maybe even one per driver/motor?

(On the drivers, etc., it's a few steps off so I'm sure I'll get to the info, but what is the typical micro-stepping setting used with ballscrews?)

[jalessi]

Tomdchi,

Most guys are using the switching power supplies due to the cost differential.
48 volts will be fine if you match the motors correctly.

"Does the consensus here support Gecko's assertion that the power supply's amp rating should be 2/3rds the motors' total rated amps? Does that hold for switched power supplies?" Yes.

"Would it be a problem to use two or more smaller rated switched power supplies - maybe even one per driver/motor?" No issues using multiple power supplies.

On the drivers, etc., it's a few steps off so I'm sure I'll get to the info, but what is the typical micro-stepping setting used with ballscrews?) Its easy to use set the microsteps at 10 , the resolution will be more than sufficient plus the math (steps per equation is simple).

Are you from Chicago?


Welcome to the Zone,

Jeff...

I'm about to dive into this whole process, and I got Hoss' full set last week, which is great, and he's been nice enough to answer some strategy questions as I'm plotting out the conversion.

I'm still working my way through this massive thread (and these answers are probably in there somewhere, but I haven't found them yet) but I thought I'd throw them out to the folks here rather than just bugging Hoss:

DC power for the drivers/steppers:

I'm currently planning on using NEMA 24/570 oz/in for X and Y and NEMA 34/906 for the Z, and the Keling digital drivers.

- Is there a consensus on using switched vs. unregulated toroidal DC power?

- Am I right in understanding that voltage primarily limits rapid speed, so if I'm not too worried about that, then 48v DC is OK?

- Does the consensus here support Gecko's assertion that the power supply's amp rating should be 2/3rds the motors' total rated amps? Does that hold for switched power supplies?

- Would it be a problem to use two or more smaller rated switched power supplies - maybe even one per driver/motor?

(On the drivers, etc., it's a few steps off so I'm sure I'll get to the info, but what is the typical micro-stepping setting used with ballscrews?)

[hoss2006]

I'm about to dive into this whole process, and I got Hoss' full set last week, which is great, and he's been nice enough to answer some strategy questions as I'm plotting out the conversion.

I'm still working my way through this massive thread (and these answers are probably in there somewhere, but I haven't found them yet) but I thought I'd throw them out to the folks here rather than just bugging Hoss:

DC power for the drivers/steppers:

I'm currently planning on using NEMA 24/570 oz/in for X and Y and NEMA 34/906 for the Z, and the Keling digital drivers.

- Is there a consensus on using switched vs. unregulated toroidal DC power?

- Am I right in understanding that voltage primarily limits rapid speed, so if I'm not too worried about that, then 48v DC is OK?

- Does the consensus here support Gecko's assertion that the power supply's amp rating should be 2/3rds the motors' total rated amps? Does that hold for switched power supplies?

- Would it be a problem to use two or more smaller rated switched power supplies - maybe even one per driver/motor?

(On the drivers, etc., it's a few steps off so I'm sure I'll get to the info, but what is the typical micro-stepping setting used with ballscrews?)

What I'll add is 48v is well suited to my recommendations on my electronics page as the motors are chosen with suitable inductance ratings where 48v is near the max voltage needed for the motors. the gecko formula is the square root of the motors inductance multiplied by 32 = the recommended max voltage.
Any higher voltage will just generate more heat in the motor.
I use 1/8 micostepping generally, a 6050 driver for example only gives a choice of 1/2 or 1/8.
Read #6 for tips and settings.
FAQs
Hoss

[tomdchi]

Bah! Drat! Phooey! (Not the exact wording I've been using all day, but I'll be polite here.)

I received and unboxed a G0704 today. Lots of things looked good to acceptable.

But there are a bunch of unground surfaces on the Z and Y ways that look very much not acceptable.

It's very similar to what "wheeliecake" showed on the machine he received back in 2013.



A little closer detail of the machining marks:



And the Y ways:



You'll note that the non-contact surface between them is nice and smooth.... I ran my 0.0001" indicator over the machining marks and the indicator read at least 0.0004 variation for each ridge/groove, plus some because the ball end of the indicator is spherical and can't get all the way down into the grooves. This is not some surface finish issue that can be dealt with via lapping the parts together. On the other hand, the non-contact surfaces between the dovetails were pretty nicely flat when I checked with the tenths indicator.

The horizontal surface on the underside of the X table looked pretty good (some traces of machining marks, but actually smooth/flat):



In an ideal world, they'd throw a replacement column and base on a truck up to me and I'd have it in a few days. From what others have described, if I can get some action, it sounds like they're going to want me to crate it back up and ship it back to them, which I now know is a PITA. I'd be surprised if whatever I get for replacement(s) would have been checked by someone who knows what they're doing before it's sent out.

Wish me luck dealing with Grizzly.

edit: the angled dovetail surfaces look consistently good. I didn't disassemble enough to check the sliding contact surfaces. I pulled out the Z gib strip and it very clearly rocked from corner to corner (I didn't check further.) Also, the unit came with a "Test Record" with blue hand-written numbers showing that the unit is within spec in various ways (some clear and easy to understand, others cause a bit of head scratching.

[Signal9]

Hoss,

If I recall on your video, you are controlling your spindle with a C6 board connected to your KBIC 120 board.

Are you running any sort of relay to power on/off, if so where are you interrupting the power?

[hoss2006]

tomdchi, expect grizzly to want the whole mill shipped back, which you should that finish is unacceptable.

signal9, Don't have a relay just a main power switch supplying power to the kb board.
Hoss

[wrljet]

Yours looks even worse than mine did. Grizzly claimed it was intentional, to hold oil like scraping.
Basically told me to pound sand. Chinese machines at this price point are junk with no quality control.

Bill

[KriegKuts]

Can anyone post a picture of what its supposed to look like? I have the same thing going on. Mill is smooth and accurate but the non saddle sides are all rough like that.

Sent from my Nexus 5 using Tapatalk

[hoss2006]

This is what my column ways looked like in 2010, things have gotten worse since.

Attachment 321392

If both mating surfaces are rough it's not good but if one is fairly smooth the "oil retention" of the other could be lived with.
Hoss

[joemosfet]

Mine, hopefully it's helpful:

Photos are before any use, heck even before break in. (Large photos)

[tomdchi]

Hoss - thanks! My initial call was pleasant. If I hadn't learned so much from this thread and other sources, the "tech support" guy's comments might have thrown me off. I don't think he was really trying to make it sound OK, but it was definitely in the direction that someone who didn't have a clue might have gone "Oh, OK, it's not a problem" and accepted the condition of the ways.

Part of that, I should say, is the issue of precise terminology for the parts of the dovetail way. I couldn't find any reference that broke things down to specific terms for the individual surfaces. I think he was trying to make sure that I wasn't just freaked out about a rough non-contact/sliding surface (he couldn't see the photos I had e-mailed in.) I'm describing things as the "male"/non-sliding part and the "female"/sliding portion, each of which have opposed "angled" surfaces, one of which is the gib, and then there are the "flat"(?) surfaces. It seems like everything has names, and I just don't know what the traditional terms are to talk about those individual surfaces so I feel like I'm making it up as I go along.

I e-mailed them the photos and I marked up their parts diagram to clarify where the problem surfaces are. I also realized that I forgot to mention on the phone that there was still some paint in various areas of the machining marks. I infer that they are rough milling the dovetails, then painting, then finishing the dovetail surfaces, which further indicates that these parts were unfinished.

Like I said, this initial call was pleasant enough, but when I asked, he did comment that it can take 24 to 48 hours for someone to look at the e-mails/issue to get back to me. I sent the initial e-mail with the photos while we were on the phone, and he did apologize that his system wasn't hooked up to the e-mail address so he couldn't look at what I had sent. (5 or 10 years ago, that wouldn't be surprising, but today if you've got people trying to troubleshoot problems (and potentially avoid having to return/exchange heavy pieces of machinery) it would be reasonable to give the front-line folks access to that sort of thing...)

Let's see how the subsequent contacts with Grizzly go and how long this takes to be resolved.

For small/light items, shipping them back and getting a replacement isn't a big deal. But I'm not looking forward to re-building their crate (it's mostly together, but partially fell apart getting the mill out) and then lugging the machine back to the garage from the basement, remounting it, getting the chip tray, tool box, etc. back in and bodging the crate back together. Then on top of that, scheduling a work day around the freight pickup and another day around the freight delivery. (oh, and yet more cosmoline to get off...) As long as I end up with a reasonably well functioning mill, that's the key thing.

[Hypnotoad]

I'd be inclined to ship it back to them dissembled in several boxes. There's no reason they need it to come back fully assembled, it's going in a refurb pile anyway (I would hope at least).

[coherent]

Must have been finished on a Friday end of shift when they wanted to meet their build quota and go home... or whatever the equivalent of a Friday end of shift is in China. I've had a few issues with my mill and my lathe and other items I have bought from them and I must admit, Grizzly has been pretty good at working to make things right. Not always as fast as I'd have liked, but with each issue eventually things got sorted out to my satisfaction.

[tomdchi]

Thanks for the photos. Yeah, some trace of the machining (like joemosfet's last photo) I could see as being workable. But on this one the milling marks are deep enough that they wouldn't retain oil, they'd be channels to drain it.

Someone called back this afternoon, but I was meeting with clients so I couldn't take the call, and I didn't get out to call back before 4:30pm eastern time, so hopefully I'll talk with someone tomorrow morning. I really hope I don't get some push that these unfinished ways are OK. For one thing, the surface on the underside of the table is much better so simply being consistent would be a reasonable criterion for what is acceptable, the other is that there's paint in the machining marks, which obviously shouldn't be there on machine ways.

Just like Hoss' photo, the surface between the ways is nicely finished (ground?) and smooth. It makes no sense for the sliding surface of the ways to be a rougher, less smooth surface than a "cosmetic" surface. We'll see how it goes...

edit: followup - well, I got off the phone with Vince from tech support. He was good and very professional, but it's pretty clear that this is what Grizzly is accepting from the manufacturer and sending out to customers. What I got from the conversation is that this is not an anomaly that snuck through quality control, its the norm. He also mentioned several other model numbers, so this is the quality level of this whole class of smaller mills that Grizzly is selling currently. Caveat emptor.

Between the cost of shipping and restocking fee, and the hassle of getting the thing back to the garage and re-crated, I'm chalking this up as a learning experience. (assuming it's not too far out of whack when I measure it's movement, "wiggle" etc.) As Hoss pointed out, as long as there is a reasonably smooth surface on one side and this ridged "surface" opposite it is "flat" it should work... well enough.

I've been doing a lot of reading on dovetail way scraping and watching "MeullerNick"'s videos... I'm not about to dig in and scrape those ways this weekend, but I am on my way to pick up some additional metrology gear so I can quantify how the machine is actually behaving to decide if it's workable or if it's actually too sloppy to tolerate.