Machining 26mtr circle

Friend of mine called the other day
He has a shop that also works on big theatrical constructions and sometimes if he needs someone to brainstorm about a problem he calls This time he was in the far east working on a big turning stage with a 26mtr diameter
The problem was that the steel 300mm wide ways the rollers were on were not perfectly in line The stage was in place already and almost impoosible to remove
They wandered 5mm on hight on that diam Also the inclined angle was not perfect As far as I undertood the ways were made from sheetmetal 35mm thick It was connected to the floor in a irreversable way The filling underneeth was not perfect So holes were drilled in the ways and the gaps under the ways are filled with epoxy He had contacted a compagnie in Holland that could do on site milling But the problem is time and off coarse money
A Japanese compagnie wanted about 5times the amount of the Dutch compagnie
Was there a alternative that he could do himself ?? We discussed several options
A way that seemed possible to me was with a big beltsander 400mm wide A drum as big as possible Driven by big motor 25Kw or so
The whole connected to the stage on a swivelling arm
On the shaft of the grinding drum you connect a tracer that follows a steel ring next to the steel ways say 25mm wide that is adjusted level to the allowed tollerance That will be a challange of its own because of the tight space But they were able to measure it within the needed accurracy so it must be possible to bolt down a adjustable construction out of several segments Probbably needs a ground surface
Then turn the stage with the beltsander on and keep adjusting the tracer till the whole way is ground

What do you experts think of this idea ? Possible or doomed from the get go
Please ask questions if the set up is not clear Englisch is not my best language

Peter

Maybe the fastest/cheapest way to deal with this is to make sheet steel panels about 6mm thick with an ID and OD radius matching the existing circular track, each segment about 2 to 3 meters long. Fit them with set screws or bolts in a pattern that allows the screws to "jack" each panel to be level to the desired specification on top of the track. Once all panels are fitted, use tape or clay to "dam" the sides so that a contained volume is created between the underside of the 6mm panel, and the upper side of the existing track.

Mix a thin, easy flowing epoxy filler that doesn't shrink, pour into ~13mm holes in a grid that pierce the 6mm plates. This will fill the gap between the two surfaces, and when cured the heads of the screws that space the 6mm plates are ground flush by hand. If the epoxy does not totally fill the 13mm holes it's not a problem, the rollers will easy go over them without issue, or additional epoxy can be applied to fill.

Using precision levels and laser transits to fit each panel should allow uniformity to at least 0.5mm, with more patience I'm sure it could be even better. The 6mm panels should be specified as flat from the vendor, if they are bowed or curved in plane it makes fitting more difficult.

An even easier, but likely not as long lasting approach is to do essentially the same thing but just with the epoxy filler - make dams for the inside and outside of the current track to whatever height is needed, then pour a large amount of "self leveling" epoxy into the current track. If done using the right compound it will cure slowly enough that the material will level itself (as water would) before solidifying.

This might be acceptable if the rollers that run on top of it are not steel, but have a softer rubber coating. Steel rollers would likely start breaking down the epoxy through a "wave induced" fatigue process.

A level or two, a straightedge or two, a big handheld anglegrinder or two and lots of hard physical labour should do it.

We discussed both possibilities but were concidered not durable enough
The rollers are Nylon. Every 2 wheels are a tandem Wheelload is about 5 ton each So 6mm will fale IMHO We were talking about 20 mm and turn down the diam of the wheels to compensate

Does anybody has experience sanding down significant volumes

Peter

5mm on a a face 26m wide is not actually that bad (1/500)

Why is such a small variation so important for a theatrical stage ? Surely the stage itself will flex as much ?


Bill
edit . off by a factor of ten! 1/5000

Peter from Holland said:

We discussed both possibilities but were concidered not durable enough
The rollers are Nylon. Every 2 wheels are a tandem Wheelload is about 5 ton each So 6mm will fale IMHO We were talking about 20 mm and turn down the diam of the wheels to compensate

Does anybody has experience sanding down significant volumes

Peter

Click to expand...

Not being able to do the calcs I can't contest your statement that 6mm plate would fail under nylon rollers, but my gut says the nylon would fail first. For abrasive removal of that much material I'd think that heat would be an issue, not to mention large amounts of dust, noise, and time to do the abrading.

Billtodd said:

5mm on a a face 26m wide is not actually that bad (1/500)

Why is such a small variation so important for a theatrical stage ? Surely the stage itself will flex as much ?


Bill

Click to expand...

Told him too it wasn`t that bad But for some reasson they wanted better Also the stage flexes because the ways are out of tolerance It should not flex if the ways were dead on as well as the wheels off coarse
My friend was hired for consulting/trubble shooting
His shop made some if not most of the construction here in Holland
He said that they wanted 0.1mm. I said well good luck then
The ways were galvanised and that gave problems too The zinc flaked off and bonded to the wheels Told him to buy some brooms and connect them to the stage to sweep clean the ways while the stage rotated And a scraper on the wheels if need be

Peter

I wonder if 5 tons/wheel set isn't a little high, perhaps they'd be well off to add additional sets to spread the load. And heck, it's Japan - tell them to put the wheel sets on moving computer controlled rams, then "map" the errors and have the rams move in/out to counter the track error as the stage rotates.

I'm almost serious about that last bit, that would be super cool...

metalmachines.com.sg

Try these guys. We use them regularly for 8-10m diameter bearing housing forgings. In-situ portable machines.

Dave

You make me thinking
How about a mecanical way to do so
On each wheel or each tandemset make a excentric shaft that has a lever with a wheel on it that rides on a tracer track
So adjust the hight of the tracer track so the stage rides level
A bit like what Sip does on his jig drilling machines to compensate wear in the acme spindle

Peter

Peter from Holland said:

You make me thinking
How about a mecanical way to do so
On each wheel or each tandemset make a excentric shaft that has a lever with a wheel on it that rides on a tracer track
So adjust the hight of the tracer track so the stage rides level
A bit like what Sip does on his jig drilling machines to compensate wear in the acme spindle

Peter

Click to expand...

Sure, but the loads will be high so the "correcting track" will need to be robust, with bearing followers and actuating mechanism to suit. But with only 5mm or so of correction maybe the leverage is such that it's not too bad. Perhaps a hydraulic setup would work, with each wheel carriage's pump riding on the correcting track.

Hmmmmmmmm- I was thinking spring loaded axles to compensate but it really seems that the stage will have to have enough flex to compensate with no fuss over bearing points.

Is there any indication that the stage is so rigid as to need precise level on ring?
Is the floor so rigid as to have no give?

I am betting if the stage were supported at two points 180 opposed the sag on unsupported sides would be far greater than that 5mm...

I do see post #8 but still...

what about a milling head mounted on a frame with some weight to it to help stabilize it. With rollers that are mounted with a walking beam fashion so that the height errors are cut in half with the mill head in in center of roller sets

map out the highs with a laser level and a nice finely graduated ruler
angle grinder to get them down,
hand belt sander to get them smooth

ETA: or a water level would be very good, too

0.1mm in 26m, thats better than 1 arcsecond across the diameter.

Thats not a rotating stage, it's a precision machine tool!

Peter from Holland said:

Told him too it wasn`t that bad But for some reasson they wanted better Also the stage flexes because the ways are out of tolerance It should not flex if the ways were dead on as well as the wheels off coarse
My friend was hired for consulting/trubble shooting
His shop made some if not most of the construction here in Holland
He said that they wanted 0.1mm. I said well good luck then
The ways were galvanised and that gave problems too The zinc flaked off and bonded to the wheels Told him to buy some brooms and connect them to the stage to sweep clean the ways while the stage rotated And a scraper on the wheels if need be

Peter

Click to expand...

Ooops it's actually more like 1/5000 (5/26000)

Even if it can be levelled ,there will be a question as to how long it will remain flat . The track is presumably only mild steel (else why galvanise) and will be brinelled by the rollers after a while . That will make the plates want to lift off their fixings and buckle.

Grinding the plates will thin the plate and add asymmetry into the mix!

If they want a long term fix maybe a dynamic levelling system would be more cost effective.

I think I'd have designed it to use standard train rails and fixings (all tried and tested technology).

Bill

it is not possible to fix a problem without knowing what the problem is, except by chance.
is it really flexing because of the 5mm hight diff?

how has that measurement been made?

has the ring been "mapped", how many points have been measured and to what accuracy?

what is the foundation the ring is on? is that sagging under load?

is the stage a uniform load, or is it asymmetrically loaded?

what is the bedding of the ring to foundation?

all this needs to be answered before a reasonable strategy for a solution can be arrived at.

having said that, I bet some could level it (to say, 2mm), in a day to a week (depending the amount of metal to be removed)with a 9" angle grinder and a stack of 36 grit cubitrons. keep it cool tho!

If we can assume that the stage is going to be dead level, why not use a slow setting very liquid epoxy levelling compound, pour onto plates using dams to stop it flowing off the edges.
When cured stick down new plates of uniform thickness to the epoxy, using screws to be sure of rigid fixing, make the new plates from pretoughened steel lazer cut to size,in this way they can be thin, compensate for height by reducing roller diameter.

Other easier option, rework the stage and just use pneumatic tyres, its how all fair ground rides generally work, way more tolerant of defects and do a great averaging effect of track errors due to squidgyness! Grease the track with some washing up liquid and you don't even get a tyre squeak.

PS i would then castrate and fire the following, prick that chose galv finish on a roller way, prick that failed to level track, pre concrete pour track inspector, finally not least the moron that thought 0.1mm on a 26 meter tolerance was a good design.