QCW Parallelogram

After a few years of use I discovered that my QCW (X-50 Woodworker with Z20 upgrade) is not exactly square. I was originally doing a lot of more organic shapes but then got a job that required very precise right angles and discovered the problem. (And it explained a few problems I had in the past.). So, I loosened the bolts, knocked it square and retightened the screws. This worked for a while but it eventually drifted back to its natural ‘not-quite-square in the XY orientation’ position. Tried it again and, naturally, it drifted back again.

Unlike the tramming, there is no fine-tuning adjustment for the XY angle. I finally reached out to 1F support and they told me to send more info and a detailed video. They responded to the detailed video telling me that they didn’t think it was possible for it to move over time and that I shouldn’t have applied any type of force after loosening the bolts to get it back square. After I explained the problem in more detail (like it was probably never square) they suggested that I loosen the bolts, fix it in a square position and retighten the screws. After I reiterated my attempts at fixing it, they told me to completely rebuild the QCW. So, I pulled off the wasteboard strips and rebuilt it. Unfortunately, that didn’t work either. Still just a little out of square.

Anybody have experience with this? I’m thinking of using some very thin washers to shim a few of the QCW cross members to see if that might bring in back in line. 1F support has gone a little quiet on me today and I’m hoping they come back with a potential solution.

1 Like

I am about to build my QCW will check square. Amazon has machinist shims. How far out is your QCW? What size QCW? Is video on youtube?

I uploaded it to YouTube, here’s the LINK

It’s the X50 Woodworker; 32"x32" I think?? As you can see in the video it’s about 1/16" out of square over about a 24" run. Not too much but just enough to throw off projects with tight tolerances. I would definitely duplicate my test before I screw on the MDF wasteboard.

Some aluminum shimming stock (with screw holes cut out) is probably a better way to go; especially because it’s such a small adjustment. Thanks for the idea!
Think I’ll try to put two layers in the far right back of the QCW support (between the support and inside of the tube) and one layer in the middle QCW support. And then just do the exact opposite of that on the front side. That would adjust it in the right way, wouldn’t it? (I was thinking backwards for a while.)

Still no word back from support.

1 Like

Yes you are on correct path. Also look into machinist shim washer or shim washer.

Hey Jim,

I had a look at you video. Really annoying, but the problem is created by the machine on which these parts were machined. So they may be all out of square in one batch of QCW frame, perhaps in all ever produced (*).

I think just inserting a shim would be a too small contact surface. What you would need is a total metalworking artist with a hand file who files everything to an accuracy of .05 mm, or re-milling everything on a CNC milling machine where the milling head cutting edge and profile feed are REALLY at right angles.

The other way would be to relocate the holes for the feet bolts on one side of the machine. You would still not have the t-tracks rectangular to the other axis, but who cares for t-tracks, they’re just there for clamping. The only thing that counts is if you mill a shallow line with the tip of a v-bit around the entire workarea, is that the X and Y are rectangular to each other. The milling path, not the the t-tracks. They don’t matter as long you don’t take them as reference. A reference would always be a fence that you mill on the wasteboard AFTER X and Y axes are rectangular one to the other.

*) Footnote: DISCLAIMER: These are just assumptions. Yet I did not assemble my QCW frame. It’s still on the list for when I will have moved to a new workshop.

I am a machinist by trade. Many production machines have shims in them.

Expecting to have things in place to inspect my QCW around the 7th. I will post in .001" because I live in USA.

1 Like

@Aiph5u Thanks for jumping in. We so appreciate you here on the Forum and I know you haven’t assembled your QCW yet and was afraid you might not chime in. You’re right on with your ‘contact surface’ concern. For example, if I just shimmed one end (or just did them unevenly in general) there would be little gaps where the end poles ‘bridged’ over the squared end of the T-rail cross members and that wouldn’t be as sturdy.

1F support told me “Sure, give it try!” when I told them what I was considering, so I took that to mean they didn’t have any better ideas. Immediately, I realized that I needed custom shims that would fit such that it had two holes, a a lot of contact surface as well as not compress. I found some flashing, grabbed some shears and a drill and off into production mode I went. Here are the results:

I made 12 of them in total. Front far right had no shim; middle right took one, two on the middle left and three on far left. By having them each get thicker by the same amount I was trying to ensure that there would be solid contact. Like @RogueMachinist said, good solid shims have their place in this world. Here are the results:



It was tough to get them aligned and make sure that I wasn’t going to cross-thread the connecting bolts. When I was sure they weren’t turning freely, I took my driver and crunched them pretty tightly. And then, did the exact opposite shim order on the rear frame pole. Then I tacked on some scrap MDF and ran my square tool path fully expecting that I had either done too much or too little adjusting. Here’s the result:

BAM!! Dead on first try!! (Yes, it was mostly luck, I know.). Now I feel both a sense of accomplishment and I little bit pissed at 1F that I had to do this at all. Maybe the degree of accuracy I am looking for is simply beyond the manufacturing tolerance of the QCW. (Support hasn’t answered my inquiry on that yet…). I find that hard to believe because these CNCs are really well built otherwise; very sturdy, no belts, etc.

It seems that there should be some sort of fine tune X/Y adjustment for this like the unit has for the tramming. I asked if they assumed that the QCW would always just simply be square but no answer to that question either. Again, maybe I’m expecting too much precision from these machines and they are supposed to simply be ‘hobby’ units but once you dump the pathetic Makita hand router and put in a gnarly spindle, these things rock.

Also, after a couple of years of ownership, I personally feel like 1F Support has taken on a sort of ‘Jeckyl and Hyde’ vibe. Sometimes I get an immediate, helpful response (even after hours) and other times the reply seems to be the bare minimum that the author could get away with without bothering to read the request.

I always start here in the Forum (couldn’t own this machine without it) because I think we should be as independent as possible so when we do go to support they can help us because they aren’t dealing with questions that have been answered several times over and are really just learning curve issues. And I know I had a lot of those, so thanks to all of you for making this place so helpful!

3 Likes

If it’s not too much of a hassle, would you be able measure the thickness of the flashing you used for your shims? I’m trying to get an idea of roughly how much you actually had to shim to get the frame back into square.

Based on the QCW design, having that rounded machined face mating with the tubes makes keeping the cuts, and as a result the t-slot length, consistent pretty difficult on what I imagine are pretty low volume production orders.

1 Like

Can you be sure that its the QCW frame, and not the machine homing out-of-square?

The shims are .008 thick.

@Machinist Bill, what do you mean “homing out-of-square?” When the CNC cuts a 90 degree angle in simply goes down one rail (with the other not moving) and then the other (with the original not moving). So, what we are effectively measuring is the angle of the X/Y rails, yeah? Lots of posts here talking about the best way to measure/anchor the 1F’s feet in a perfect square without the QWC. But with the QWC the corners have platforms with threads and you put machine screws through the CNC’s feet to attach them. There is a little natural ‘play’ in the CNC otherwise (that folks struggle with to keep square) but with the QCW you have no choice where the feet/corners go as they sit in a very predetermined spot. I was shimming to move those spots.

1 Like

Yes, exactly correct Jim. Even if the QCW may be absolutely square, the gantry can still home unevenly (both sides not fully against the solid stop). This could happen for various reasons, such as wood dust/debris building up in front of the bearing housing (one reason I advocate separate hard-stops).

I don’t believe this phenomenon is very commonplace, but it can happen and is none the less something to be aware of. Additionally, if you’re running the Buildbotics controller with stall homing, there can be errors with insufficient motor torque applied during homing. I don’t run the BB controller, but I have seen conversations in the past related to this.

2 Likes

Hey Jim, hey Bill, hey all,

this is true, it can happen at any time. Usually it depends on how a gantry machine homes. A gantry-style CNC is one with two Y rails. Often the machine only “homes” one of the Y axes. Theroretically it is possible if your machine is a parallelogram, the two Y carriages could alter their home position in a way that the resulting angle is a right angle even if the X and the Y axes form no rectangle, but a parallelogram (software squaring)

Anyway on the buildbotics-derived Onefinity Controller, it can be remedied this way

2 Likes

We need to look at just the QCW no cnc. Aiph5u are you close to getting your QCW up? I am still shooting for the 7th but running behind.

I have a woodworker on a qcw frame and noticed also that it was out of square at about 1/16" in 24" using a pointer in the spindle and a verified 24" square. I found there was enough clearance in the mounting of the Y axis rails to the QCW corner blocks on my machine to adjust out this error.

Gene, when you say “enough clearance” do you mean wiggle room when the screws that connect the feet to the QCW platform are loose?

No, the 8 bolts(4 each side) on the end brackets on the 2 Y axis rails. When I loosen them from the QCW corner pieces that connect the round tube and the Y axis, there was just enough lateral (sideways) movement to achieve square. I moved the X axis to the Y limit and then loosened the 8 bolts and shifted the Y - end to the right which corrected my out of square.
It could perhaps also be achieved by adjusting 1 Y rail either + or - direction but I didn’t try that. I think left or right adjustment is safer since the X axis rail holds the spacing between the 2 Y rails when the bolts are loose.

1 Like