I did some learning today, tramming

I have an idea of what I want in my head, and a napkin sketch. Let me put a little more thought into it.

2 Likes

I usually make use of my table structure, for the LR2 I levelled it, then sort of built the top upside-down. You can do this with any two parallel beams like a pair of sawhorses. If you are looking for a very large table it makes even more sense to sort the frame out first. If you do visit the post below check out the one before and after it for the full picture

1 Like

Uhh…
Before I put my LR3 together I think I want to understand what is happening.
The braces on the x-axis were too tight? That pushed the y-rail held down with the rail blocks further out?
Is it because the table isn’t level so the y-rail undulates along the length?
Or does the y-rail have a slight curve to it?
I’m confused, which is not unusual, but may not be the best state to be in as I close in on upgrading my LR2.

The last 3 of mine were, so that pushed the bottom of the core out about 0.025 degrees on the far end. Normal day to day use never showed this as an issue since the day I built it. This was only noticeable with a 1/2" diameter surfacing bit and even then only slightly.

If you are worried, I think the best course of action is to measure your rails at each brace to see if there are any inconsistencies.

No issues with the Y rail, if that does have a warp in it the machine and the work piece will follow it.

2 Likes

I wonder if it’s not that they were too tight but they were tightened slightly out of line with the others. You could get a whole twist as you travelled x.
Maybe a long straight edge placed vertically on the front edge of the strut on either side would exaggerate the twist so you could see it.

Loosening the screws fixed it.

Yes I think we are saying the same thing. The last three braces were out of tension, compared to the rest. The braces themselves do have a minimum rail position and I thought I designed then in such a way that clamping would not push the tubes out, I thought the clamps would just open up…I was wrong. Actually just found this out here, Something isn't perpendicular

I set me tram near the 0,0 position so that was my reference. I will take some measurements today, and I have a old Beta beam sitting here so I am going to mess with the tensions to see what happens.

Both of my YZ plates are firmly grounded so twist on my build isn’t the typical twist some get when a plate sits crooked and the front of one side can be tapped on the table.

There are a few ways to adjust most aspects of the machine so I try to start with the most direct checks and adjustments first. A good example of this is adjusting the tram with the bottom tool mount. It has the most direct effect of the tool angle. You have an even finer adjustment using the top tool mount, I just never usually tell people to do that.

2 Likes

Maybe. If the beam was twisted the longer the beam is the harder that would be to see, or even measure. This is on the tune of 0.1-0.2mm out over a 200" diameter swing of the tramming tool. So on the 1/2" bit that is something like 0.001mm tilt. If my brain is doing the early morning math right. These weren’t ridges you could catch a finger nail on that was paint soaked would being deli sliced.

1 Like

Maybe a torque spec…? All hardware uniformly tightened. Wow. Almost typed that without laughing.
All kidding aside, following along this thread, I’m impressed it was as close as it was. Nice job with the new design!

Track saws are an option for long rips. I use a ruler with length stops to quickly repeat accurate setting of the track on both ends of the rip, but you need long, straight tracks. Some multiple-piece tracks don’t assemble straight without some effort.

The track saw is easier to handle than trying to manipulate large sheet stock on the table saw.

Question, would putting plugs in the end of the EMT and screwing the ends of the EMT into the perpendicular plates stabilize the gantry ? The side plates would need to be a little larger.

I have not noticed any deficiencies there. With the right type of inserts, that could make for a very solid connection. I do think in most cases that would need very accurate tubing cuts, and that can be hard to do.

Tracks saws and saw guides would work. A table saw really wins for accuracy here though. An overlapping Y piece could work, the longer the overlap the more accurate the joint. Then it just comes down to if you have a table saw build a traditional torsion box, if not slap together your LR on a flat-ish sheet and cut out the design I want to make.

2 Likes

Or mount a long arm to the core that sticks out the front (points in -Y direction).

Then as X sweeps left to right see if the distance from the end of the arm to the table changes. Then you are measuring not just twist in the rails but also if distance between the rails induces a tilt of the core.

If you surface the table and make the arm as long as Y you’ll be able to measure crazy small amounts of differential tilt.

3 Likes

I also have some hand drawn plans for a torsion box table with a CNC, but Schneewittchen was just too small. The idea was to have segments you can puzzle together either with latches or with interlocking pieces and strong magnets (I do love magnets… :D). Never left the drawing stage though, I only cut the holes for the vacuum table with the CNC. :smiley:

This is the table where my idea came from: Modular Torsion Box Table Beds But obviously segments. :smiley:

2 Likes

Ohhhhh, great idea. Slap the longest square I have on the face of the core and sweep back and forth…Nice.

1 Like

Cool! I was thinking something along those lines with a little more LR3 specific flavor!

Fusion is crashing like crazy on me today. I am going to try another computer and if it still keeps crashing I’ll give it some more time.

1 Like

Didn’t you learn in high school or uni that tramming is not an effective learning method? Consistent review and discussion with a peer learning group is much better than tramming before exams…

Wait… wrong forum…

1 Like

Can we have :expressionless: as a reaction for KVCs posts? :smiley:

1 Like

Three years ago you were like “Endstops are not really needed!!”. :smiley: Now this. :stuck_out_tongue:

4 Likes

Yup taking a hard left turn. IF people really want to start chasing zero’s I am not going to shoot them down. If you have noticed, instead of me fighting that lately, I help push them further and the conversations typically end up with them saying “that is enough for my needs”!? If you take out the measurements and bring in real world demo’s of measurements, it seems to really help drive the point home.

Who knew, it’s as if internet people just want to resist no matter what the outcome.

Never surfaced a spoil board in all these years of CNCing. I shaved down a once obvious high spot but it’s a new me. I am adding a chasing zeros, or advanced tuning page to my to-do list.

5 Likes

I think it’s a good idea to encourage that. Even though your machines are entry level machines if you look at the price point, you make sure that they can compete with high-lvl machines if you want to.
I think that a lot of my “problems” are only there because I measured. My first inlays weren’t perfect, I needed to make sure that it was square manually, pushin it against stop blocks, so now I also want endstops, I had to surface the spoil board for my primo because my table was shitty, there were only slight ridges that never bothered me until I did the cutting board. When building the vacuum table it became more than obvious that the table was really crooked when I had to surface the MDF from both sides to let the air flow through, so I told me I needed an absolutely flat table for the LR3.
Surfacing gave me ridges, I trammed it. There is a slight lifting: I need to get that out. There are still ridges on the left side but not in the middle: could the screws still be to tight? It still moves around 5mm back even after twisting the gantry. Maybe it is my strut plates that didn’t cut through completely, maybe I didn’t clean the holes enough and now they are pressing the gantry back (I am going to check that today).
There was no need to do something about most of these problems, after twisting the gantry there is maybe 0.2mm lifting, one of the braces rubs very, very lightly because I had to tighten the core a lot (and I still think it’s the 25mm’s fault, without having evidence).
What I want to say with this: my cuts would still have been good enough to never see any of those issues, my control box, cut with the LR3 is fine, as is the grid of the new vacuum table I cut way too fast and deep just to see whether I could and I am pretty sure nobody would see the problems but me and a very, very trained eye… But somehow your expectancy grows, you want it to be perfect. I have good ears (I used to hear the beeping of those old TVs when it was turned off three rooms over and my family though it was impossible), I hear every tingling sound a bearing makes and it annoys the hell out of me, so at least those have to be spot on. :smiley: I am nearly at a place where I am satisfied with the LR3, and I guess most of the problems are my own doing. :expressionless:

1 Like

I just do not want to push it like it is actually needed. 95% of the time, people will never know if they are way out of tram and pretty far out of square.
Same reason I never want to have a fancy table. I do not want people to see that and think they need to build a fancy table. All the CNC’s work great screwed down to any 1/2" or thicker surface.

Can it do all the fancy stuff, yes, after your first new build, no. Absolutely not, you need to tune any machine you buy to get finer and finer results. I went through most my life never adjusting a table saw fence or my chop saw backstop, heck I made most my stuff with a jigsaw for probably half my life. I do not want someone build the machine and spend weeks on end chasing zeros to maybe possible get them then go to make the first cut and try to cut a square at 100mm/s and tear apart the machine,…just to have to do all the tuning again.

What people NEED, is very hard to determine. If I say it can do 0.01mm it has to be able to do that. So I choose to always ask what people expect, or need for their types of projects. If I lose customers because I do not have misleading specs, I don’t mind.

7 Likes