Tuesday, January 21, 2014

Buttoning Things Up? (mystery solved)

In yesterday's post I mentioned that I had at last managed to get the lower spindle height counter on my shaper to reset, to a value of 6.000". While I was glad that the counter could be reset, I was confused as to why the Martin maintenance manual stated that it would reset to 0.00, while the dealer of the machine, in checking a T-26 shaper of similar vintage, found that it reset to a value of 5.000". I was unclear on the concept behind the secondary lower readout, and the differences between my machine, the other machine, and the manual were mystifying.

A reader sent me a note shortly afterwards saying that there were other Martin shaper owners out there who would understand the situation and would possibly be able to help me out. With the mail, he had cc'd a few of those people. This morning I awoke to find a long detailed email from a fellow in California named Scott Slate who runs bespoken woodworks. He explained the situation clearly and now I am no longer in the dark. Awesome!

Here's the low down:

- the reason the T-26 shaper resets to 5.000" while my machine, the T-20, resets to 6.000" is due to the fact that the T-26 is a tilting spindle machine and has less available travel in the height adjustment. It's supposed to reset to 5.000" and mine is supposed to reset to 6.000"

-the manual is simply wrong about an number of things in this regard. The display does not reset to 0.000". I guess I shouldn't be surprised. The manual for my T-54 jointer was so ridden with errors in translation that I rewrote it for Martin.

So, still the question: why the two displays for spindle height? The two counters function a bit differently, and for very good reasons, as it turns out.

The upper display is the 'relative' counter and can be reset at any time to 0.000" by pushing the reset button. It also works in concert with the calibration device to establish the point at which the top edge of the cutter reaches dead level with the table top. If you were using a rebating head, as I have set up in the machine, then once you have the cutter zeroed to the tabletop and registering 0.000 in the counter, it is simple to raise the cutter to the depth of the rebate required, say 0.375" or whatnot, and make your cut. If you then need to make another rebate a bit deeper, then you could either climb directly to that value, or reset the counter to 0.000 and add the extra depth when you raise the cutter. Similarly, if you were using a grooving cutter, you would zero the cutter to the table using the calibration device, then you could raise the cutter to the desired offset. If the grooving cutter were 0.250", and it was offset 0.500" from the edge of the board, you would raise the cutter to a value of 0.7500" on the readout. Same process moving up from there, and you would have the choice to reset to zero for each jump, or perform a bit of addition.

The lower counter, which cannot be set by the calibration device, and only by raising the spindle all the way to the top, is the 'absolute' counter. This is especially helpful when the cutter you are using does not have a flat surface on top, or is of some complicated form upon which you cannot readily obtain a reference. Let's say it's a cutter for forming door or winder rail and stile joints. You position the cutter onto the spindle, atop a specific thickness of spacer, and then lower it to the position you need for the cut, adjusting the fence and table opening as necessary. Then you would make some test cuts until you have found the correct position for the cutter to achieve crisp mating surfaces in the joints between the rails and stiles. At this point, you've done a bunch of work and the machine is finally ready for work. 90% of the time goes into the set up for the cut, and once you're done that you can produce a bunch of work.

A lot of the technological changes in shapers of late have come about in effort to reduce the costly set up times. By having an absolute counter, you have a reference point you can return to next time you have occasion to use the same cutter to make the same doors or windows. You would have a notebook or data sheet, in which you record the specific cutter, its position on the spindle (i.e., the thickness of the spacer below the cutter), and the absolute readout value when you had it dialed in.

Essentially what my machine has is a manual version of the present-day system Martin uses, in which the cutter stays together with the spindle all the time, and the spindle has a chip in it which the machine reads. That enables the machine to automatically reset the fence, the cutter height and possibly tilt, the fence position, and even the table opening size. All of more or less automated in other words. The following video shows depicts this process in the making of a simple window frame - I think it is especially clear just after the 6:00 mark when he refits the fence to the table:


While similar in most respects, that's a newer machine than mine and has a newer type of spindle system which happens to be the same system used on some CNC routing equipment, called HSK. My machine has the 'Dornfix' system. The machine in the video also would sell for at least 3 times the cost of mine, so I'm feeling quite content! The difference in speed in interchanging tooling, comparing the dornfix to HSK systems is there, but it is not a huge difference in time. The chip reading capability and automatic positioning would save a fair amount of set up time, and that makes sense in some businesses, but not so much in mine. I have a manual, low tech version which will take a few extra minutes to set up, and will require taking notes about the sets ups of various tools instead of entering it into the touch screen as another memory setting, and that's fine with me. I actually like the 'low-tech' aspect, although it is all relative. A 2000 Martin T20 is leagues more complex (and safer to operate) than a 1950's Whitney or the like, obviously.

I've noticed that other German shaper manufacturers also make use of a calibration device, like Panhans, which has one called the Zeromaster:


Scott mentioned in a later email that he thought it made sense to have the spindle reset its absolute count at the top of the stroke, and not the bottom, as that would be the only option in some cases if you wanted to reset the counter when a tool was already fitted to the spindle. It might be that the tool is too large to drop down past the table rings. Makes sense to me.

I'm glad to understand this aspect of the shaper function better. I still have a ways to go in terms of learning the ropes with shapers mind you, and I'm looking forward to learning more. Towards that end I've signed up for a 4 day workshop this spring out in Colorado to learn the processes for making energy efficient European tilt and turn windows. After that I should be able to make some informed decisions about tooling.

All for today. Looks like another major snowstorm is rolling in, so battening down the hatches here. Hope it's warmer where you are.

2 comments:

  1. Chris,
    My thoughts about why your machine set absolute positioning at the top of the travel would have to do with loading the spindle travel moving up. As to setting to 6.000 would have to do with using the Cartesian coordinate system where Z up is a positive move. If the vertical travel is 6.000 then when the spindle is dropped to the lower travel, then absolute position would read 0.000. Make sense?
    Jack Ervin

    ReplyDelete
    Replies
    1. Jack, thanks for your comment and observations. It is the case that when the spindle is fully lowered, the absolute measure drops to around 0.000". Another reason for calibrating the spindle at the top of the height, as suggested by Scott Slater is that there would be less chance of debris affecting the positioning at the top. Whatever the case, I'm glad to have the operation of that lower counter figured out.

      ~C

      Delete

All comments are moderated:

Anonymous comments will be deleted

Spam will be deleted

Comments containing links unrelated to blog content will be deleted