CNC Build Ep 4: How to Tram a CNC - Cheap and Accurate! About $60 in tools to align the spindle

[@pierre4316]

Thanks a lot for the sharing for all the videos you are doing on the AVID CNC. Very useful !!!

[@CorbinDunn]

Thanks! I've got a bunch more coming..next ones will be about the ATC, and then the spoilboard table build.

[@thebruhhh]

You really answered a lot of questions that I had that other videos just assumed. Excellent video, great work and thank you so much for putting this together. My comfort levels are increasing!

[@CorbinDunn]

Awesome! I'm glad it could help a bit!

[@abrahamlincoln5283]

Some basic concepts: First, we need to make sure the spindle is the same distance from the spoilboard at the extremes of the Y-axis as well as the extremes of the X-axis. This assures the cutting tool, for a given Z-value is co-planer to the X-Y plane. Once this is done you can go about the process of tramming the router head. I have, at last count, 9 CNC milling machines. I use the following method for tramming them: Obtain a length of 1/8" diameter stainless steel tig wire approximately 3 feet long. Bend one end of the wire 2 inches from the end at 90 degrees. Insert the wire in an appropriate sized collet and tighten. The wire extending from the collet effectively is the radius of an arc around the spindle. Next bend down the end away from the spindle down 90 degrees. When the spindle is properly trammed the tips of the wire should be the same distance from the spoilboard throughout the entire arc. Now, instead of trying to reach appropriate tolerance in a 3" radius arc, you are using a 30" radius arc. This is a proven method and you can visually see any errors that are present. In terms of accuracy, let's assume you dial in your spindle using the 3" radius to +/- 0.001". This would have a +/- reading of 0.010" with the 30" radius. It is quite easy to achieve greater accuracy, especially when you consider the "sampled" surface under your granite block or glass plate vs a 60" diameter surface "sample". And, the tig wire is cheap. I use stainless so as to minimize any deflection from magnetic sources.

[@CorbinDunn]

Cool ideas! Doesn't this assume your spoil board is perfectly level to begin with?

[@kurtbilinski1723]

I initially used a granite surface place as is done here but learned a lesson. Unless the surface plate specs state that it is parallel, square, and flat on all six surfaces, errors will be introduced. Most small inexpensive granite surface places have only one flat surface, with all the others only being "meh" in terms of accuracy. Additionally, if the spoil board that the surface plate is sitting on has been surfaced, it's a good bet that the distance between the spindle and spoil board is already constant, so any variation is either due to changes in the spoil board since surfacing - or more likely, that the surface plate's top and bottom surfaces are not parallel.

[@CorbinDunn]

Yeah, you made some good points! I always shim my granite plate to get it to be co-planer with the x/y axis. And yeah, if your spoil board is surfaced, then that distance is constant, but it may have ridges in the x and/or y direction; but I bet it would be minor enough to not affect things, because anything on top would be just sitting on a parallel ridge.

[@kurtbilinski1723]

@@CorbinDunn The issue with surface plates hit me when I placed two on edge, sitting on a third. If the two on edge were really high quality, their faces would be flat against each other. They were not. So, yeah, the top is flat but not guaranteed to be parallel or square to any other side. I finally coughed up the money and bought a smallish granite tri square, because if my setup isn't correct now, everything I make - especially fixture plates - will be wrong, then pass their errors on to everything else. This metrology things really hard to do without spending a lot of money...

[@Festivejelly]

But isnt that irrelevant if you shim it. Since you are making the top precision surface plane parallel to the indicator in each corner. You only need 1 flat surface to do this.

[@lucasthehood121]

Id really like to know where I can find the jog controller you have! Thanks for the videos!

[@CorbinDunn]

Yeah! I got it from VistaCNC: www.vistacnc.com - but lots of people make them, just make sure they work for whatever CNC software you run (mach 4, etc)

[@doyounglee6990]

Thanks for the info!

[@CorbinDunn]

My pleasure!

[@backwardsreaction]

What is the hand held device you use to jog the machine?

[@CorbinDunn]

It's a pendant, and I got this one from vistaCNC.

[@MrDante195]

@@CorbinDunncan you can a video how to installed that on avid cnc

[@CorbinDunn]

@@MrDante195it’s pretty easy; VistaCNC has the directions. They make a Mach 4 plug-in for it.

[@Festivejelly]

The indicator should be held in the collet not clamped to the side.

[@CorbinDunn]

For when you are tramming? Can you explain why you think that is important? When spinning the indicator around, it shouldn't make a difference. When tramming in a vice (etc), it definitely should be clamped to the side and not move.

[@InconnuAnonyme404]

@@CorbinDunn I think you should mesure the runout of your spindle / collet separately. It's necessary for high precision application.