What's new
What's new

assessing wear of Van Norman 22L

Brandenberger

Hot Rolled
Joined
Sep 19, 2017
Hi,

I'm working on this Van Norman 22LU universal mill. It is a ~1944
machine.

I've taken off the table and top saddle so far, leaving the lower
saddle and knee. The upper and lower saddle mating surfaces are
nearly pristine, showing almost no wear and original scraping. Interestingly
there were two copper shims in between these parts however.

The spindle runout in the 50-taper spindle shows .0002" runout on the
taper, and .0005" on the spindle face.

The cutterhead face and mating machined and scraped surfaces to the
ram are hardly worn at all, showing the original scraping. I have removed,
cleaned the bearings of the cutterhead and re-installed.

The ram movement forward and backward shows .001" difference over
12 inches or so of ram travel, indicating the mating surface for the
cutterhead.

VN wear - 1.jpg

The column face and knee show considerable wear and scoring. In fact
the knee surface is covered with sheet metal, that was pinned at the
rear and held on at the front of the knee with a strip screwed into the
front of the dovetail. I am trying to assess the impacts of this wear
and decide whether it makes sense to remove the knee, lower saddle and
address them.

VN wear - 3.jpg

When I indicate the column face from the spindle face (sweeping a .0001"
indicator through an arc) I get .0018" difference from left edge to right edge.
Note the cutterhead angle is against the hard stop for horizontal position, but
not itself indicated to confirm the angle yet, since I'm unsure what surface
to use for this so far.

VN wear - 2.jpg

Indicating the lower saddle surface while moving the saddle forward and backward,
I get .0015" difference over about 8" of travel (can do a more extensive test of
this with a parallel, this was a rough check directly on the scraped surface).

VN wear - 5.jpg
 

Attachments

  • VN wear - 4.jpg
    VN wear - 4.jpg
    88.6 KB · Views: 168
The sheet metal on the knee is badly scored and dented, no doubt from parts
being dropped on it over the years:

VN wear - 6.jpgVN wear - 4.jpg

The sheet metal is .032" thick or so.
 
Nice looking machine, ive not seen one before. What intentions do you have for it? Rebuild or make do and get it up and running?

Starting at the top, in a perfect world all should be square to eachother, your numbers dont sound bad over 10-12" imo. Looks like you have some scraping left at the top of the column, check at the bottom of the column too, its not a given that this is factory but should be a good reference unless someones scraped it bad.
11.jpg

The steel shim on top of the knee sounds like a bodge. I imagine its there to shim the saddle back to height after wear has been machined out of the knee/saddle. Not sure why copper shims where between top and bottom parts of the saddle but is highly likely they shouldnt be there. If it were mine id definitely have that out at least and keep an eye out for other bodges.
The knee should be square to the column both ways, im guessing its dropped and rocking.

How did the bottom of the table/top of saddle look, another likely are for wear.

Cheers
D
 
I agree there's still some scraping remnants on the column face, but it seems odd that it is .0015" off perpendicular to the spindle axis. Maybe first I need to remove the cutterhead again and determine if its scraped face is parallel to the spindle axis?

If the only problem were the column face perpendicularity to the spindle, since this is a universal the table angle setting if indicated to the spindle would compensate I think.

Unfortunately getting the lower saddle off the knee dovetail looks like it requires pulling one or two shafts out of the knee, which means quite a lot of tinkering with the mechanisms in there which look in very good shape.

I'm tempted to remove the saddle gib and try to at least remove the steel sheet, but it seems likely in doing that I will end up with an unavoidable height discrepancy that would cause the power feed gears and shafts to bind.

Has anyone removed the saddle completely and have a procedure for it? Seems like several people on PM have asked about it over the years but I can't find reference to someone who has done it?

Thanks,
Phil
 
I agree there's still some scraping remnants on the column face, but it seems odd that it is .0015" off perpendicular to the spindle axis. Maybe first I need to remove the cutterhead again and determine if its scraped face is parallel to the spindle axis?

If the only problem were the column face perpendicularity to the spindle, since this is a universal the table angle setting if indicated to the spindle would compensate I think.

Unfortunately getting the lower saddle off the knee dovetail looks like it requires pulling one or two shafts out of the knee, which means quite a lot of tinkering with the mechanisms in there which look in very good shape.

I'm tempted to remove the saddle gib and try to at least remove the steel sheet, but it seems likely in doing that I will end up with an unavoidable height discrepancy that would cause the power feed gears and shafts to bind.

Has anyone removed the saddle completely and have a procedure for it? Seems like several people on PM have asked about it over the years but I can't find reference to someone who has done it?

Thanks,
Phil

Not really, indicating the head mounting face as you did doesnt mean much till youve an idea of what the ram if doing, some wear in the ram dovetails could give you that .0015". Worth pointing out again that.0015" in this situation over that distance might only just be outside the tolerance of a new machine imo. Ram/Head mounting face/spindle are 3 separate elements that ideally should be perp to the column.

Id only suggest removing the steel sheet if youre committed to getting things right, strip and scrape column and knee. The saddle youd get back to height with a way liner, like turcite etc.
 
top of top saddle surface, showing little to no scraping left, and
some minor scoring. Clearly lubrication starved for a while, virtually
all of the bijur lines were clogged.

VN wear - 1 (1).jpg

Underside of top saddle, showing little to no wear. Seems like as you'd
expect the universal feature was seldom if ever used.

VN wear - 2 (1).jpg

Table bearing surfaces, no scraping evident except at the extreme ends,
but no scoring.

VN wear - 4 (1).jpgVN wear - 3 (1).jpg
 
Indicated bottom of ram from the lower saddle. Saddle/knee dip .003" from back
to front over about 12" of travel.

VN wear - 1 (2).jpg


Also seems like indicating the bottom of the ram across the saddle left to right,
there is .0015" or so difference over 8". Perhaps why the copper shim I mentioned earlier
was between the upper and lower saddle (and only on one side).

I am assuming the ram was seldom moved, so the likelihood of wear there is low,
compared to the knee / table.
 
Indicating machined side of ram (where scraping remains, so
a good reference surface) moving saddle from rear forward 12",
looks like the knee twists right .004" or the ram twists left.

ram - 1.jpg

Indicating the side of ram while raising the table, .0000"
movement over 8" of table travel. So the knee elevation
and side of ram are parallel.

ram - 2.jpg
 
Is looking like wear in the usual places. Makes sense the copper shims were in there to try and square the spindle with the table some.
Youve the table off atm but you could strap a length of timber to the knee as a lever and indicate the column for knee rock, as is and with gibs adjusted, might need a thou clock for that one ;).
knee rock.jpg

Could also see where you can get a thin feeler gauge in, dovetails front face of column/top of knee etc.
 
I ended up getting the saddle off (details in another thread on the Van Norman forum (VN 22LU lower saddle removal).

The condition of the underside of the lower saddle and
top of the knee (with the sheet metal shim removed) are thus:

dovetail wear - 1.jpgdovetail wear - 3.jpgdovetail wear - 2.jpg

As it turns out, when I indicate the underside of the ram from the
lower saddle now that it is resting on the actual dovetail way, the
variation is .001" or so, ignoring the scored areas.

dovetail wear - 5.jpg

I checked the column face at the bottom and indeed it shows original
scraping, as it wasn't scored by debris over the years that
got trapped between the column face and the knee at the top.

So now, the big question is how to address the scoring/gouging on the
knee dovetail? The obvious answer is probably to be scrape the top of
the knee and then build up the saddle bottom with one of the various products (moglice, turcite, etc.) and of course install robust wipers.

In terms of establishing the correct height for the saddle--

Given the fairly serious scoring on the knee, it seems like whomever
installed the sheet metal on the knee didn't mill down the knee surface
to begin with, though maybe this machine has been "repaired" more than once.

The only things that seem to matter in terms of the saddle height are the
acme saddle screw and the splined shaft which drives gears which ultimately
drive the table feed.

Thanks,
-Phil
 
Datum Surface for VN22L?

Hi,

I'm trying to determine what the best datum surface is for this VN22LU.

The column surface has some fairly deep scoring, but also shows
what I assume is original scraping from about 12" below the top all
the way to the bottom. I haven't pulled the knee off yet.

casting - 1.jpg

In reading Machine Tool Reconditioning, the usual procedure would be
to use the spindle axis to check the column face.

However since this machine has a cutterhead that can tilt through 90degrees,
in turn mounted to the ram which moves forward and backward, any error in the ram
dovetail could result in a lean in the ram, left to right or front to back.

Anyone have recommendations on what datum surface I should trust, and how I should
assess the first few surfaces?

My overall goal here is to determine if this machine is in pretty good shape
except for the serious knee scoring. If that's all that's wrong with it, I
feel like I can address that somehow (perhaps grinding first or perhaps scraping, or worst-case filling the scoring with an epoxy product, if the knee surface is
otherwise in good alignment?).

Thanks,
Phil
 
I am very slowly going through a similar exercise on a VN #28A. I chose the column face as the initial and key surface. I convinced myself that VN ram motion is a non-precision motion, and that I would probably never attempt to modify a setup into a 2nd setup simply by trusting the ram motion, so Y and Z irregularities didn't bother me, although I did not explicitly consider side-to-side tilt.

On reflection, I should have at least inspected the ram ways and slide more carefully when I had things apart. There was a huge mass of chips thrown up into the ram rack when I received my (very used) machine, and it took considerable pushing fore and aft with long-handling brushes flipping chips out of the teeth, before I got full ram motion back. There was more than ample opportunity for chips to get dragged into the way/slide interface and score things up.
 
I am very slowly going through a similar exercise on a VN #28A. I chose the column face as the initial and key surface. I convinced myself that VN ram motion is a non-precision motion, and that I would probably never attempt to modify a setup into a 2nd setup simply by trusting the ram motion, so Y and Z irregularities didn't bother me, although I did not explicitly consider side-to-side tilt.

On reflection, I should have at least inspected the ram ways and slide more carefully when I had things apart. There was a huge mass of chips thrown up into the ram rack when I received my (very used) machine, and it took considerable pushing fore and aft with long-handling brushes flipping chips out of the teeth, before I got full ram motion back. There was more than ample opportunity for chips to get dragged into the way/slide interface and score things up.


It seems like the ram (when locked in position) needs to be parallel
to the knee... if not then by extension the spindle axis is thrown off.

My ram was similarly packed with chips, I haven’t pulled it off yet
but this is making me think I should.

My ram doesn’t seem to be tilted, since in the vertical plane
It is aligned to the knee movement. And the bottom and left side
Seem like they are 90degrees to a good tolerance.


Thanks,
Phil
 
VN22LU knee and column update

Hi,

I pulled the knee almost off today, enough to expose the back (mating surface to the column)
to see the wear.

First, the column face shows a lot of original scraping and a few scoring marks. that seems to indicate overall the wear on the column isn't bad (some deeper scoring toward the top.

knee - 5.jpg

The back of the knee shows some more heavy scoring, in specific locations. Probably some chips somehow got behind there, trapped against of of the port hole covers, and scored over time. Also very likely poor lubrication since there is nothing on this machine to feed the column face other than the operator!

knee - 2.jpg

The guiding dovetail side on the column is scored too, no real sign of original scraping.

knee - 3.jpg

The gib side of the column dovetail shows original scraping, not sure why the wear is asymmetric.

knee - 1.jpg

So overall, I'm trying to understand the approximately .004" twist in the knee, and whether
to address that by scraping the back of the knee, or the knee/saddle dovetail. Since the
power feed to the knee is a splined shaft coming up from the bottom of the knee, I wouldn't want to take much if any material off the back of the knee unless I take off alot and turcite it...

To address the heavy scoring on the top of the knee, I ordered some Moglice Score-Ex, and also found someone locally who would surface grind it. Haven't decided which to do yet...

-Phil
 
Finally making some more progress...

I milled the underside of the lower saddle, and that piece is now ready
for eventual turcite. However I also wanted to use it to evaluate the
knee twist.

Here I'm indicating the saddle movement on the guiding dovetail relative to a precision square referenced off the column. It seems like the knee has .001" twist when the knee is low on the column. Knee gibs fully tightened for this.

VN indicating - 1.jpg

Here indicating the same thing toward the top of the column. .0007" twist in same direction.

VN indicating - 2.jpg

Now indicating side of ram relative to the saddle, again sliding on the knee guiding dovetail. .001".

VN indicating - 3.jpg

Now, keeping the ram gib mostly tight, indicating the ram travel relative to the column face. .0018".

VN indicating - 4.jpg

All of these seem consistent... the knee twists .001" toward the ram; the ram twists .0018" toward the knee, and the combination of those nets out to .001" ram relative to the knee.

This is over about 8" of travel.
 
VN knee wear

I mapped out the relative wear on the top of the knee bearing surface. At the
very front there is evidence of original scraping, so I kept that as zero.

VN knee wear - 1.jpg

In these measurements I used a large indicator base, so the deep scoring isn't affecting
the measurements. Overall, about .002" of worst-case wear other than the scoring. The
plan is to totally scrape the surface anyway, but I wanted to map out current condition.
 
finally, I took the knee off, as I may want to scrape the column and/or back of knee before
dealing with the top of the knee.

I generally followed Paolo's procedure, however a few notes for posterity...

Lifting the knee, having removed the SHCS attaching the knee elevation horn to the machine base.
Gib loosened but still in place at this point.

VN knee removal - 1.jpg


The gib lock binder & handle is unscrewed fully and removed. There is a brass plug (part 2L-609) still in the hole, which can be pushed in and out once the knee is high enough for that point to be above the column dovetail.

VN knee removal - 2.jpg


The binder, handle and plug.


VN knee removal - 3.jpg

Once the plug is removed, the gib can be unscrewed fully, avoiding the need to lift the knee all the way off the top of the column dovetail. OF COURSE make sure the knee is high enough to be free of the power feed splined shaft (22L-132) but if you can push the plug in, the knee is already well above that height.

Knee on the dolly.

VN knee removal - 4.jpg
 
how do you flip this thing on its back?

And the column face condition...

VN column - 1.jpg

VN column - 2.jpg

Overall some scored areas, but lots of original scraping too. Will blue up
my 3ft camelback and evaluate wear...

Question-- how do you flip this beast on its back using only a engine hoist?
I would ideally like it horizontal to scrape the column.
 
I wish you would have messaged or emailed me. I would have told you to scrape the column and prepped the knee before match fitting the knee and gib. Its difficult to scrape the knee right if the
Coulum is off or out of parallel and not square. Did you check the back of the knee if it was lower in the middle 40%? Did you at least square cut the smooth ways to give the new oil pockets? Do you have a precision blade square? I will gladly help. Rich
 
Hi,
I haven't scraped anything yet other than machining the underside of the lower saddle... so no damage
done yet! :)

I'm going to blue up the back side of the knee (untouched so far) today and will post some pictures.

Thanks for the offer of help!

-Phil
 








 
Back
Top