New spindle location to reduce flex

[David Falkner]

I have been thinking about the flex in my Z axis and wondering, designing, and pondering solutions. Today I believe I have figured the easiest and best way to handle this problem. Mind you, the flex is very, very slight but I can see the effects when I plunge full depth into 1/2" Baltic Birch at 175 ipm.

The issue, I believe, lies in the extended distance of the single plate of 5/8" aluminum that came with the tramming kit. Prior to the tramming plate the spindle was mounted about 3" higher than it is now and 5/8" closer to the trucks. To me, and I voiced this to Nate when he sent the tramming plate, the cantilevered effect of a nearly 20 lb. spindle hanging that far below and in front of the trucks is just asking for flex.

So my remedy is to move the spindle mount up as far as I can on the tramming plate. This will actually place it about 3/4" higher than it was originally as shipped. I will have to remove the tramming plate, drill and tap new holes, then mount everything back into place and redo the tramming but in the end I think the effort will pay off.

Here are some supporting photos so y'all can see if you concur with my thinking -

Flex shows in the plunge to begin these slots. I can't ramp because the compound bit with its upcut start causes the face veneer to splinter and I can't have that.
001 - Evidence of flex on plunge cut.jpg

You can see the flex potential here by how far in front of and below the trucks the spindle hangs -
003 - Flex potential.jpg

Proposed location (approx.) -
004 - Proposed location.jpg

Proposed location will still allow access to the tramming bolts -
005 - Proposed location.jpg

Proposed vs. original vs. current locations -
006 - Proposed location.jpg

Thoughts?

David

[Mark Bolton]

Could you just order another spindle clamp and add a second one in the proposed location?

[David Falkner]

That's a good suggestion, Mark, and a few folks have suggested I do that. And that could work but there would really be a hurdle in getting them perfectly lined up with nothing more than a drill press with a less than stellar runout. I am more likely to move the current mount up to the new location than to add a second one for that very reason. Now if I had a milling machine...!

++++++++++++++++++++++++++

Update - since my last post we got an order for a 14" Longworth chuck so I tried something; I put blue tape where the entry points are for the slots so I could go back to a ramp instead of plunging. So I modified the toolpath to include a steep ramp, 45° to get the bit into the wood quickly but not plunge, and made sure in the simulation that all looked correct.

I lightly sanded the area and then used my hard veneer roller to press the tape down firmly and cut the first disc. When I pulled the tape up I pulled up a few fibers of the face veneer but not too much. I also saw that the cuts were perfect! Plunging causes the bit to pull everything to the side by a very tiny amount, probably less than 1/64" and it's only happening at certain slots, the ones where the entry is parallel to the X axis. When the slots begin parallel to the Y axis there is no movement and the cut is very clean.

Also, when the cut starts out across the grain there is a greater likelihood that grain will pull up from the face veneer, as can be expected. Those that begin with the grain never chip or splinter.

I have seen several times where y'all have used some sort of mask, like contact paper, to put over the workpiece before engraving and I guess that's sort of what I did with the blue tape. What is the mask I have seen and where can I get it?

Anyway, I can continue doing this because it works but it doesn't address the flex that I know is there. But maybe I have prolonged the issue and will address it later because I don't really have time right now to pull all of this apart and drill/tap holes to then go through the tramming process again.

Tape at entry point -
001 - Tape at slot entry points.jpg

002 - Tape at slot entry points.jpg

Perfect cuts -
003 - Perfect entry and slots.jpg

Not yet sanded -
004 - Perfect cuts, before sanding.jpg

I appreciate the suggestions and if you have any new ideas on the removing the flex then please post them here. Again, the flex is extremely minimal and unless I plunge parallel to the X axis I never see any flex so it's not like I am dead in the water or producing sloppy work.

David

[Mark Bolton]

When I have those tricky line up issues and dont have the machining equipment I go the counterbore, and oversized through bore route with a thick washer that allows a little adjustment room. At least that way you could have your existing mount for registration and precise alignment and when that is set just lock down the upper one.

Adding more rigidity was my thought but adding more weight I guess also is a downside.

You seem to always land on a more than acceptable solution. Im sure this time will be no different.

[Mark Bolton]

Couple simple gussets? Cheap piece of aluminum from online metals, drill press, set of transfer punches, and drill and tap?Untitled.jpg

[David Buchhauser]

+1 to that!! Much easier and probably just as effective.
David

[David Falkner]

I like the idea behind that, Mark, and the drawing is very good! Since I found that I can use tape and a quick ramp set to 45° that gets into the full depth very quickly but isn't plunging then I think this simple solution will do the trick for now. That doesn't address the flex, though, and I truly don't see it anywhere except when I plunge into BB for these Longworth chuck discs.

One reason I am against doing braces like you've shown is the added weight, even if it is slight. I think the moment of inertia is over the top on this cantilevered mount anyway and it's fixing to get worse. This 3.0kW spindle weighs in at about 18.5 lbs. if I remember correctly and the gantry is plenty stout to handle that. But hanging out there on that single 5/8" plate I don't think the tramming mount is up to the weight task like the gantry is. The reason it's going to get worse is that I have a bearing going out in this spindle and am replacing it with a new spindle, a 3.2kW that weighs just under 22 lbs. and is 1" longer (taller?).

So I will be going up another 3.5 lbs. and I think that's going to compound the flex even if I don't see any now except in this one instance. For now it will be tape and ramp but when I replace the spindle I will seriously consider moving the mount up where I have indicated.

David

[David Buchhauser]

You're also probably getting a lot of flex (bending) at the edge where your tramming/spindle mount plate meets your larger plate. In other words, your larger plate may be bending there. In the past I have employed dial indicators placed at strategic locations to detect such deflections.
David

[Mark Bolton]

A well designed gusset (not.mine) would only way ounces or a pound maybe. I'd think if your adding spindle weight and power your going to need more than a bracket relocation. That said it looks like you probably have a bending moment just above the plate where the spindle plate mounts to the z plate so whatever option may be best to bridge that joint as well and stiffen the whole thing. You could always hollow out the webs on the spindle side as well and dump even more weight.

[bobby milam]

My clamp has the mounting bolts on the sides of the clamp. You can remove the clamp and spindle as one piece if you wish. It seems that this would be easier to slide onto your spindle while it is mounted and then align the holes to get it more squarely mounted than trying to remount the clamp you have. Here is a picture of one similar to mine, it just doesn't have the holes drilled in the side yet.

https://www.amazon.com/Spindle-Bracket-Aluminium-printing-Engraving/dp/B076CDWXNK/ref=asc_df_B076CDWXNK/?tag=hyprod-20&linkCode=df0&hvadid=309722091285&hvpos=&hvnetw= g&hvrand=11616606373169840137&hvpone=&hvptwo=&hvqm t=&hvdev=c&hvdvcmdl=&hvlocint=&hvlocphy=9027191&hv targid=pla-664088927962&psc=1&tag=&ref=&adgrpid=60439548223&h vpone=&hvptwo=&hvadid=309722091285&hvpos=&hvnetw=g &hvrand=11616606373169840137&hvqmt=&hvdev=c&hvdvcm dl=&hvlocint=&hvlocphy=9027191&hvtargid=pla-664088927962

[richard newman]

David, my thought is to eliminate the tramming plate and bolt a pair of spindle mounts with ears (as above) directly to the z stage plate. The individual mounts could be much thinner to save weight, maybe even half the thickness of the original. Tramming would be trickier and more time consuming, but once done you're set, you're done.

[David Falkner]

David, my thought is to eliminate the tramming plate and bolt a pair of spindle mounts with ears (as above) directly to the z stage plate. The individual mounts could be much thinner to save weight, maybe even half the thickness of the original. Tramming would be trickier and more time consuming, but once done you're set, you're done.

Well, Richard, that is starting to look better all the time. In keeping the tramming plate I get the ease of tramming, but like you said, once it's set I shouldn't have to do it again. However, and I didn't notice this when I changed out the plates, today I realized the original plate is 0.200" thicker than the tramming plates. They are 0.600" where the original is 0.800" and that's a significant difference.

The extra mount should be here in a couple of days and I can see how large the mounting holes are, to see if there's any wiggle room for tramming without the tramming plate. I guess another advantage to using the original plate is that it's already off and will be easy to drill and tap.

Original plate with proposed new mount location (approx.) -
009 - Original mounting plate with proposed second mount location.jpg

Original plate thickness relative to the new plates -
010 - Old vs new comparison.jpg

Close up of the original plate thickness relative to the new plates -
011 - Old vs new comparison.jpg

David

[bobby milam]

I was just reading your original post on this? Are you plunging straight down and what size bit are you using? Have you tried ramping the start? Are you sure that the plate is flexing or could it be your bit? I ramp all my cuts with an endmill. My machine will snap a 1/4" EM like a toothpick but my old machine would have a little flex if I didn't ramp. I have snapped too many bits already from stupid mistakes so I try to take it a little easier on the bits when I can for the same of my pocketbook.

[David Falkner]

In the original post, Bobby, I said, "I can't ramp because the compound bit with its upcut start causes the face veneer to splinter and I can't have that" but said 'compound' instead of 'compression'.

I have seen videos and read tool manufacturers saying you should plunge as fast as you can and I have read that you should never plunge. I don't like to plunge but more than that I don't like splinters and chips in the face veneer.

But in post #3 above you'll see I tried some blue tape where the bit enters and switched back to a ramp and that works great. I have been plunging at 175 ipm into 1/2" BB and cutting full depth at 175 ipm with a 1/4" compression bit for a long time, maybe the last year or so, and haven't broken a bit yet. Watch me break one tomorrow!! LOL!

The blue tape works good enough that I am switching all my toolpaths on these Longworth chucks to 45° ramp in one inch of travel, which is almost like a plunge but it gets past the upcut portion of the compression bit very quickly to lessen the chance of splintering and chipping.

Am I sure the plate is flexing or could it be the bit? The way the spindle is cantilevered so far from the trucks I don't think it can do anything but flex, even if it's just a tiny amount. Is the bit flexing? Likely, I'm asking a lot of it but I don't think it's to blame for what I see when I plunge. I think it's the tramming plate setup.

David

[richard newman]

There's no way a carbide bit could flex that much. They don't really flex at all, they snap. I saw similar looking results back when I used to rout 3/64 grooves (for ebony inlay) in solid wood with a dremel. It was the lack of rigidity in the tool itself, molded plastic bearing seats, etc. Bought a Dotco 1/4" die grinder and made a router base, problem eliminated.