Studer RHU 450 from the 1960s

[ballen]

I had some time today and since I have not heard back from Richard about this, I went ahead on my own. Originally I had planned to remove material from the lower half of location 4 (just beneath the grinder wheel casting). But after I got the parts off and looked at them, they were too long and too wide for my grinder and for my surface plate. After that I decided to remove material from the upper part of location 2 (rotary table just above machine casting). Here are the bottom and top:

To get a reference surface I first ground the non-working surface above the dovetail, using the rotary plate as a reference. This surface was originally planed and still had the planar marks. By grinding it parallel to the rotary plate, I get a reference surface which I can then use for milling and grinding the rotary plate to a different height.

Then I milled off about 480 microns (about 0.019"). I did this using multiple passes of a small carbide end mill because that made it easier to compensate for some "belly" in my milling machine.

Then back to the grinder. The part is twice as wide as my grinder, so I needed to do half, then rotate and blend. Note that in both the milling photo above and the grinding photo below, I am using the reference surface created earlier to clamp the part down.

The result is not perfect but better than the other alignments in this part of the machine. I had thought about deepening the relieved center portion when I had it on the mill, but I did not want to "erase" the machine number stamped onto the casting, and since the facing part is relieved, I decided not to relieve this part further.

The hold-down dogs no longer had enough range, but they had 14mm holes and a 12mm eccentric drive, so I bushed them in to 13mm. I made the bushings by lightly knurling some 14mm steel rod then boring it out to 13mm. I coated the bushings with permanent loctite and squeezed them in with a vise. Not pretty but works well!

Here it is, 0.52mm = 0.020" lower than this morning.

 

[CharlyDE]

Edit: This messabe was written while Bruce's preceding message has been posted.

Hi Bruce,

I set a dial indicator stand on a pair of 18.00mm high parallels and zeroed the indicator on the wheel hub.

Has the real height of the parallels been checked, too?

After some parts of the machine have been rebuilt (e. g. the table), before removing so much material from the grinding wheel stock, I would measure the geometry of the grinding wheel stock guide surfaces related to the rebuild parts before. I expect, that the relevant surfaces of the grinding wheel stock are within the wanted geometry, but who knows, if there could be found another issue, remaining from further pre-owner's rebuilds.

Comments please. Where is the best place to remove 520 microns = 0.021" of height?

I'm not sure, don't have enough experience; so I'm tending to remove the 520 microns from the top surface that matches to surface 4. In the photograp, that top surface seems to be made for removing some material, because the level is interrupted by a painted lower surface.

Cheers,
Karl

 

[ballen]

Has the real height of the parallels been checked, too?

Yes, it's within a few microns of 18mm.

After some parts of the machine have been rebuilt (e. g. the table), before removing so much material from the grinding wheel stock, I would measure the geometry of the grinding wheel stock guide surfaces related to the rebuild parts before. I expect, that the relevant surfaces of the grinding wheel stock are within the wanted geometry, but who knows, if there could be found another issue, remaining from further pre-owner's rebuilds.

I did study the remaining misalignments before cutting and grinding.

I *only* removed 520 microns, because if I correct the other inaccuracies, one of them will raise the axis 40 or 50 microns, leaving enough material for me to remove to correct other inaccuracies below. It's complicated to explain, but I will put the machine within the Studer and Schlesinger specs now, and can keep it there if I also correct the additional inaccuracies that I found in the grinding wheel slides. Those inaccuracies are:

(a) grinding wheel spindle points down and to the left, because the cross slide on the machine base has worn about 40 microns = 0.0015" more on the left side where it runs on a V-way than on the right side where it runs on a flat way. I could step scrape the lower half of the rotary table to compensate for this. If I do that it will raise the grinding wheel spindle on the left (the main one) and lower it on the right (where it was even higher).

(b)On the upper (manual) cross slide under the grinding wheel the two dovetail faces are out of parallel by about 20 microns (0.0008") over their length. At some point in the future I might correct this, but I don't think it's enough to have any effect on the accuracy of the machine.

I'm not sure, don't have enough experience; so I'm tending to remove the 520 microns from the top surface that matches to surface 4.

This top surface at location 4 is the grinding spindle casting. It's heavy and hard to clamp and align, so I did not consider this the best option.

 

[Richard King]

I like what I see....sorry about not responding as I am at the cabin. I recall reading your message, but I guess I forgot. sucks getting old. You didn't need my help and I am so proud of what you did. Grinding the top was exactly the way I would have done. Rich

 

[ballen]

Rich, Karl, thanks for your feedback and comments. I spent a bunch of time measuring the geometry of the rotary table and slide, so if I want to make these better, I now have a good idea of where metal needs to be removed. But as far as I can see, it's all within spec and "good enough" as it is currently.

Before reassembly, I "fixed" a slow spindle-oil link in the grinding wheel head. I put this in quotes because the leak was so slow that I am unsure about where it was from. So I won't know if this worked until after the machine is back together for a few days or weeks.

I am going to make one other improvement before reassembly. I noticed that the cross slide is very worn in one corner, which is closest to the grinding wheel and coolant. In the photo this is the bottom left corner.

Another reason it is so worn is because the way oil for this corner comes from an oiling nipple which is completely hidden behind the grinding wheel guard! Even if you know it is there, and want to oil it, that's not possible without removing the wheel guard and the wheel and flange. Definitely a design oversight.

One way to "fix" this is for me to mill a diagonal oil groove that connects this "bottom left" oil ring to the one on the bottom right. But I don't like that much.

So I am going to try drilling a new oil passage as shown with the red lines, to connect this bottom left (in the photo) oil ring to a new (visible and accessible) oil nipple that I will add. I will need to drill three long holes in cast iron, 6mm (1/4") diameter, one with length around 120mm (5") and the other two with length around 105mm. These holes only need to be precise enough to intersect each other and existing passages (say under 1mm wander in 100mm). I will plug two openings with aluminum or brass plugs.

I'd welcome comments on this idea.

[EDIT]

I decided that this was the way to go. Drilled the passage that goes across the cross-slide, went in 110mm = 4 1/4" from each side. The holes met perfectly in the middle . This 1960s Swiss cast iron cuts beautifully. I did these in vertical mode because access and clamping were easier. The third hole I'll do in horizontal mode because the cross-slide is too long (480mm=19") to stand vertically. Or perhaps I'll remove the work table and clamp it directly to the vertical table. Hmmmm...

[FURTHER EDIT]

This part is finished: I did the last oil passage this morning in horizontal mode. Instead of plugging the ends of two of the passages, I enlarged them to 6.8mm and threaded them M8. I will close them off using M8 setscrews secured with thread sealant.

The passage I am drilling here meets up with the oil hole visible in the photo above, which that feeds the oil ring at the top right.

 

[ballen]

Here is a photo of the cross slide with the new (second) oil nipple, where I've indicated the new oil passage with red lines.

Originally I had intended to put this new oil nipple the left end of the cross slide (where the hex-head grub screw is visible, covering the 210mm hole that I drilled to carry the oil to the far side of the cross slide). But I realised that this corner will be covered by a splash shield. So I drilled one more passage about 50mm = 2" long to locate the oil nipple next to the one for the near side of the cross slide.

I had to close off four openings: two on the front of the cross slide and one on each side. The traditional way is with brass or aluminium plugs. I prefer a more easily reversible solution. So I threaded each opening for an M8 grub screw (8 x 8 mm) which is secured with a liquid thread-sealant.

I was wondering about one other point. The angled sides of the dovetail do not have any oil points, and are above the flats, so don't get oil from there. It would not be hard to make a diagonal oil groove in the gib with a hole to carry oil from the other side, and add an additional oil nipple for that. It's harder to do cut an oil groove on the other side of the support, but also possible. On the other hand, most of the weight/load is on the flats, so perhaps it's deliberate that there is no oil groove on the angled sides. Is there any logic to this choice? My idea is to cut a long diagonal oil groove which starts at the top back and angles down to the bottom front, with an oil nipple feeding oil into the top back. Then gravity would carry it down the groove.

 

[Richard King]

Looks good Bruce. You have it under control :-) I do have some suggestions. Do some measuring and see how far the wheel head slides forward and to the rear. That is probably why they put the circles where they did. Many times I will cut a diagonal oil groove between the 2 circles and in many cases I cut a groove at the end of travel about 1" from the extreme ends of travel at right angle to the travel so the oil spreads evenly as it moves forward and backward. One thing you HAVE to be sure of is you do not cut the grooves out to far long or sideways so the circuit is open to the air as you know all the oil would drain there.

I would also think about running steel or brass or copper oil lines to the rear of the machine with a oil Zerk fitting as I dislike those ball oilers. Your choice though :-) A auto lube pump would even be better :-)

Have you checked to see if the slide ways are flat? They wear high in the middle...why? You should know...lol Think about it before I tell you. lol Where does the dirt wear slides first? On the ends. I would think they (factory) scraped the shorter way lower so as it wore it didn't get high in the middle and rock like grandma's rocking chair. I like the way you are improving your machine. Everyone makes mistakes, even Studer. I have had to re-engineer many machines over the years after I have discovered a flaw in machine manufacturing. You have discovered how one eliminates stick slip by scraping oil pockets in worn ways, but not worn enough to affect the accuracy. Ballen your a good example of a good example :-) Thanks for your thread. Makes me smile and proud. Rich

Ooops I forgot the gib. Many times builders leave the gibs flat, I suppose they figure the .0005" per side will let the oil drain through the gap. I will grind a bevel on the top side of the gib to help the oil to drain down. Also I cut a diagonal oil grove (most of the time I use a die grinder and a Christmas tree shaped carbide burr. Again being very careful not to grind to low so the oil just drains out. If you wanted to get a bit more detailed. Have you seen how a Bridgeport lubes it's gibs?

They feed (oil) the wear side from the back positive side of the gib and it has a groove in it laterally so if you adjust the gib deeper it still gets oil. Many machines also have pressure lube that feeds from a Zerk drilled into the big end in back if there is room. Now you have put your thinking cap on. Rich

PS: check gib to be sure it isn't bent and check in machine with blue when it is in so the center 40% is .001" lower so it wears on the ends. Be sure to check the surface where the positive side rests. They seem to always get high in the middle

 

[ballen]

Hi Richard,

I finally got the slide fixed. I'll post some photos of that here when I get a chance, but the main issue was that the gib was far from flat (only hitting in a couple of places) and also had gotten too thin. In the process of redoing this I added oil grooves on both sides of the dovetail. Some photos of that work are in this thread in the Deckel forum:

Cutting an inside dovetail oil groove on a manual Deckel FP2

Cheers,
Bruce

 

[ballen]

I thought I would update this thread with some pictures of more recent work on the grinder.

One thing I did was improve the lubrication of the fixed (lower) cross slide. This had a reservoir in the back which had to be filled by hand, and which drained over a period of a day or two. Below the slide were channels and tubing to collect this oil and bring it back to the hydraulic sump. So I decided to add automatic lubrication, using oil from the system that feeds oil to the long ways. This created enough oil flow that the drain system could not keep up on the V-side, so I added an additional oil drain.

This is the oil reservoir:

which feeds oil as shown here:

Here are the drains in the back:

and here is where I added an additional drain in the front to the V side. Here is how it looked originally (the round piece is new, because the old one was worn and letting some oil escape over the top):

The aluminium block is a jig that I used to guide the drill bits for the two holes so that they would intersect properly.

Jig in place but displaced (I have already drilled the hole, visible just next to the edge of the jig). Note that I ground a six-sided 1/4" set of flats onto the drill bit so that I could use a 90-degree right angle "hex driver" adaptor to spin it. I didn't have anything else that would fit into that tight spot.

Here is the newly-added oil drain hole. I subsequently rounded this and smoothed over the transitions but can't find a picture.

Here is how I fed in oil up to the reservoir from below:

The fittings have an M5 thread and are oil-tight. I have drilled and plugged passages in the part with the hammertone finish. The V-slide side is the supply and feeds 3mm OD/2mm ID copper tube that comes up to the top of the reservoir. The flat-slide side is the drain, which comes from an "overflow" copper tube that pokes up to the halfway point in the reservoir. So when the hydraulic pump is turned on, it fills the reservoir halfway. Gravity feeds some oil to the cross slide, and the excess goes down the overflow.

I blocked off most of the volume of the reservoir internally with some solid aluminium pieces. So after the pump starts, within a few minutes one can see via the sight-glass that the reservoir is filled to the middle. I experimented with pressurising this feed to "float" the cross slide but the drain system was not good enough to keep up with the oil volume.

This lubrication system works well, and there is absolutely no sign of "stick slip" in the cross slide. I can easily move it in 1/2 micron increments (20 millions of an inch).

In later models of this machine, Studer added automatic lubrication of the cross slide. Probably they had thought about it from the beginning (hence the nice drains in the back) but did not get around to it until the later RHU-500 edition of this machine.

 

[ballen]

While I was working on this, I decided to also add a DRO to the cross and long axis. For the cross axis I used a 1/2 micron (0.00002") resolution SINO scale so that I would have 1 micron (0.00004") resolution on diameter. The scale has a 16 x 16mm cross section, comes with a calibration curve, and has 70mm of travel (I needed 64mm).

Here is the mount for the scale head, which I epoxied into place in the base of the grinder and later secured with two screws from the back.

This shows the scale mounted to the underside of the cross-slide. The two slotted-head screws are there temporarily to keep chips and grit from falling into the passages that I made for automatic oiling.

The aluminium support blocks were attached with screws and epoxy and then I milled slots to support the scale. Here is the scale with the other part of the head mounting bracket attached. Note the paper towel held on with a magnet to prevent chips and grit from falling into the oil passage hole.

For attaching the mounting bracket inside the machine I made a fake head/scale to hold everything in alignment. Here it is with one piece of the 2-part head mounting bracket attached

and attached to the cross slide (now with both parts of the head mounting bracket):

This is what it looks like in place, from behind. If you look carefully you can also see the two lines for hydraulic/bedway oil, one feeding oil up into the cross slide, and the other taking off the excess/overflow.

The DRO scale can be removed and replaced from underneath, by taking off the central screw visible from behind in this photo, and two M4 socket head screws that hold the scale on from beneath.

Retrofitting a DRO to a machine like this which is ~60 years old and was designed before DROs existed is an interesting challenge. In this case, the scale location is not ideal. It would be better to have it nearer the front of the machine, by the lead screw nut, and closer to the V-slide rather than the flat slide. But this location has the advantages that it is entirely shielded from coolant and grit, the head is fixed and the scale moves, and the sealing lips of the scale are above the head.

 

[ballen]

Adding a scale to the long axis was also a challenge. The total travel is 555mm and there is no space for a traditional enclosed glass scale. Here is the base:

and here is the bottom of the table (sitting on my mill)

At first I thought about using a magnetic strip scale, but some experts convinced me that these are not as accurate as one would like and also tend to drift over time. So instead I used a Renishaw RGH24X head with the RGS-20S strip scale (1 micron resolution = 0.000020"). The head has a cross section of 13.5 x 15.8mm and rides 0.8mm (0.032") away from the self-adhesive steel scale, which is 6mm x 0.2mm (about 1/4" x 0.008"). The scale itself has a 20-micron (0.0008") optical grating on the surface. After two-point calibration the scale and head should be accurate to 3 microns (0.0001"). The scale is "mastered" to the machine and will follow its thermal motion as it stretches and shrinks.

To attach the head I fastened a mounting block in place using epoxy

Here is the head mounted in place. The two outer screws are in slots so that the head can be shifted up and down by about 0.5mm. The central mounting hole is a snug fit and was drilled only after I had adjusted the head to the correct offset (0.8mm = 0.032") from the steel scale.

Note that the plate on which the head is mounted only makes contact with the aluminium mounting block. By design it does not touch the cast-iron body of the machine: one can slip a strip of paper between the two of them.

The top face of the head is also above the top of the V-groove, which together with the gap described above should ensure that lubrication oil can not flow or creep onto the head or scale.

To attach the scale I mounted a ground steel strip with a cross section of 12 x 3mm (about 1/2" x 1/8") with epoxy and screws to the bottom of the table. At the end this sits on three aluminium standoffs which are held in place with epoxy and M5 SHCS. I picked the cross section small enough that this steel strip would flex and not distort the table when under stress from temperature differences with the table.

This strip had to have ride height variations from the head which are less than 0.05mm = 0.002". This required some measurement and then milling to make it parallel. I did the measurements by attaching a 1-micron graduation dial indicator to the head mounting bracket and measuring the ride height variations every 3cm (= 1 1/4"). Removing this was a challenge because my mill has a travel of 500mm and the strip is 700mm long. But careful measurement and machining got the ride height variation under 0.02mm (0.0008") on the first attempt. I had to remove about 0.4mm (0.016") in thickness from parts of the strip, after which I removed the machining marks with wet sanding using a backing holder to maintain flatness.

Here is the tool that I used to apply the steel scale:

and here is the scale application in progress. You can see the backing paper where it has been removed from the self-adhesive bottom of the scale. The self-adhesive nature of the scale and its fragility means that you have to get it right the first time or the scale is trash.

Here is the scale laid down:

and with the "end clamps" attached:

(At photo limit, to be continued in the next post)

 

[Richard King]

Bruce you should send some of these photo's to Studer and tell them you will rebuild their used machines for $50,000.00 or what ever you think. That machine is better then new. :-) Nice square cutting too and 1/2 moon flaking.... :-)

 

[ballen]

An additional complication of the Renishaw scale on the long axis is the reference point. This is a "third channel" on the scale, in addition to the two quadratures which give incremental position. This reference channel is triggered at a specific point along the scale and can be used to establish an absolute reference. This is useful for example if power is lost, to reestablish the position.

The reference point is established with a magnet located 0.8mm (0.032") from the side of the read head. Unfortunately this magnet sticks up about 7mm from the mounting plane. The scale only has about 2mm of clearance above the hydraulic cylinder, so this meant that the reference point needed to be near the end of the scale, which does not traverse over this cylinder. It also needed to avoid the green sheet metal slide cover seen in one of the earlier photos. Lastly, the magnet has a small screw which shifts the maximum of the magnetic field by a small amount (tens of microns, so under 0.001") and that screw needed to be accessible.

The solution I found was to put this reference-mark magnet into the mounting boss for the hydraulic cylinder rod.

The magnet itself is epoxied onto a small aluminium block, visible sitting on top of the boss here:

This photo shows the pocket which I machined out, into which that block fits.

To give some idea of the clearances, here I am holding the other half of the cylinder-rod attachment point in place next to the head:

I had to mill out some clearance on this part and shift the tenon block slightly to make space for the magnet. I was worried that the push/pull stress from the hydraulic drive would distort this part, so took care that the reference magnet and mounting block only touch the mounting boss and not the mating part shown above.

Here is a photo taken from the end, where you can see the scale, the reference magnet and its adjustment screw as seen from the end of the machine. The head is "buried" inside the machine, and is only accessible if you detach the hydraulic cylinder rods and their mounting brackets, which enables the table to be slid further than the normal operating range. All of this meant that quite a bit of careful measurement and planning was required to get everything to match up the first time. The scale has about 2mm (0.080") of clearance above the sheet-metal end-cover and similar clearance over the hydraulic cylinder. Side clearances are similar.

The "signal quality" is indicated by a colored LED on the read head, which can be green, orange or red. It was very satisfying that when I turned it on for the first time, it was green (indicating proper alignment/offset) and remained that way over the entire travel.

Here is the machine with the DRO mounted and working.

I was lucky and found a nice mounting bracket on Ebay, so the power cable and two signal cables run though that. The only complication was that the mounting point for the bracket is over the coolant reservoir, so the bracket is mounted on a sub-plate which in turn is attached with bolts that are sealed with thread sealant.

It took a long time to get the scales mounted, but in the end I am satisfied that they will work well and are in protected locations. From the outside, the DRO display is the only part that is visible.

 

[ballen]

Hi Rich,

You should send some of these photo's to Studer and tell them you will rebuild their used machines for $50,000.00 or what ever you think. That machine is better then new. :-)

Thanks! I enjoy doing this as a hobby, but don't think I'm going to stretch beyond that . Also, I understand that there are a handful of small companies in Switzerland that are already doing such upgrades. Probably quite quick and easy for them because after you have found solutions to the problems, it's not hard to do. That's one reason that I have posted the pictures here, to help the next person who wants to do this.

Nice square cutting too and 1/2 moon flaking.... :-)

That means a lot to me, thank you. I'm sure that the 1/2 mooning would look better had you done it, but it works, and that's what matters most.

The original square cutting that you had me do on the long flat and V ways has worked extremely well. Now that I have run the machine for a while I've had a chance to look at how it looks "broken in". It really looks good, I can tell that the oil pockets are doing their job, and can not see any signs of further wear or galling.

An interesting test, now that I have a DRO on the long axis, is to turn down the feed rate and see if there is any stick-slip at low speeds. There is none. I can turn the speed dial down until the table is moving 0.01mm/second, which is 0.6mm/minute (0.024"/minute). You can not see this motion by eye, only watching the numbers on the DRO roll by. And from that you can see that the motion is completely steady at that speed, no grabbing or jerking.

Cheers, Bruce

 

[ballen]

The last work I wanted to show was the repair/improvement of the cross slide, which I had brought along to Richard's scraping class in Austria at the end of last year, but only got started on there. The travel was tight when the slide was pulled all the way forwards and there was some wear in the dovetails near the front. This was certainly from grinding grit getting in that area, but aggravated by poor oil delivery.

After adding more oil passages to the dovetail flats, I scraped the dovetails to improve the geometry, and added oil grooves and oil passages to the dovetails.

I straightened the gib using the Z-axis of my mill (a bit more than 5mm of deflection was required, nearly 1/4")

Here is how the gib was initially hitting

and after some scraping. This rub is very smeary but I didn't take more photos as I was too caught up in making chips and curls. In the end the contact was good and the tight spot was gone.

Here I am adding an oil groove. Oil comes from extending an oil passage that feeds the flat

Note in the previous picture how I have shimmed up one side of the table (the gib is tapered).

Here is the gib "ready to go". I did one thing that Richard does not like. I added a 0.12mm (0.005") shim behind the gib, because it was already rather far in, and after the scraping work on the gib and dovetail it was too much. I'm not proud of this but it works well.

Here is another photo. The shim is visible in the background. It has one hole for the clamping screw and one hole for feeding oil to the gib.

Here's the oil groove that I added on the non-gib side

The work on the cross slide meant that the gear which drives the slide (via a rotary lever handle) was too tight. So I had to move over the gear rack by 0.3mm (0.012"). I did this by first shifting the rack and then reaming the taper pin holes up to the next larger mm size. I also had to widen the pocket by about 0.5mm

I also decided to limit the rearwards motion of the slide by adding an aluminium spacer block visible in this photo. Without this, the slide could be moved far enough back to expose the circular oil groovers on the dovetail flats. The aluminium spacer is just large enough to prevent that in the future.

 

[ballen]

I was asked to provide a photo showing the fixed cross slide from below, as seen from outside the machine. This photo shows the connections for the automatic lubrication as well as the glass scale. The slide is as far back as it will go (hydraulic quick-retract + handwheel). One advantage of the locations I picked is that these elements are partly visible/accessible without disassembling the machine, but are still contained and protected.

 

[drcoelho]

Nice machine, congratulations !

 

[sfriedberg]

Necro-spammer reported.

 

[eKretz]

Necro-spammer reported.

Thank you, sir.