Suggestions please Vol2

[Former MemberParticipant @formermember43226]

[JasonBModerator @jasonb]

Well if there was no friction in the bearing you would just add weight to the high leg until it sat horizontally. This would be static balance dynamic may need the weight placing at a different length from the centre with weight increased as it gets closer.

 

Actual measuring could be done with a height gauge.

 

Jason

[Les Jones 1Participant @lesjones1]

Hi Gray,

Could we have some more detail of the nature of the spherical bearing ?

 

Les.

[blowlampParticipant @blowlamp]

If I understand the problem correctly, then provided all the arms are at an accurate 90 deg to one another, you could attach a weight to one of them (à la plumb bob), and let it find it’s own level.

 

Martin.

 

Edited By blowlamp on 05/04/2011 18:22:33

[JasonBModerator @jasonb]

I would have thought a 550mm cross would easily sit on a 600sq plate if set across the diagonals.

 

If the holder is in the way then balance the spider off the holder by suspending it from one of the mulls with the surface plate sat on the table or just move the table about under it and measure upwards.

 

Can you post a picture as we may be missing something

 

Jason

Edited By JasonB on 05/04/2011 18:32:19

[Les Jones 1Participant @lesjones1]

Hi Gray,

Are we allowed to assume that the tables in the milling machines are accurately horizontal in both the X and Y directions ?

Les.

[JasonBModerator @jasonb]

Just use an extended arm on the height gauge to reach beyond the edge of the table or to give enough reach so you can get at the one over the base.

 

In theory the base does not have to be sitting on teh surface plate or even be level as teh bearing will allow the spider to find its own level, the surface plate will need to be perfectly level.

 

Still can’t see why a 550mm cross does not fit on the 848mm diagonals?and if you had it laid out diaginally teh arms would be NE, SE, SW & NW not N ,E, S & W

Edited By JasonB on 05/04/2011 18:59:03

[TerrydParticipant @terryd72465]

hi Graham,

 

I have some ideas in my head but would like to see a picture or drawing of the item we are discussing before firming these ideas up to a workable solution.

 

Best regards

 

Terry

[Les Jones 1Participant @lesjones1]

Hi Gray,

First if there are no existing points at an equal distance from the centre of the spider on each arm and near the outer end of the arms then create such a mark on each arm. Place the spider on its support on the table of the Bridgeport. (I am assuming there is a DRO on the Z axis.) position the table such that the centre of the spherical bearing is directly under the spindle. Without more detail I do not know what the best way to achieve this alignment. Find a way to mount the laser on the spindle pointing down at a shallow angle from the horizontal. I would suggest about 11 degrees as this would cause the spot on the arms of the spider to move about 5 times the vertical distance moved. With the spot projected onto one of the arms adjust the Z axis until one edge of the spot is on the reference mark on that arm. Make a note of the Z axis DRO reading. rotate the spindle to put the spot on the next arm and repeat the procedure. Repeat on the final two arms. A variation on this idea would be to use an arm mounted in the spindle with something like an electronic edge finder on the end and use this to measure the height of each arm.

 

Les.

[chris stephensParticipant @chrisstephens63393]

Hi Gray,

Please bear in mind that none of us have seen the item in question, but if you add four identical weight mirrors on the four legs and shone your laser at them in turn and watched the reflection on a distant wall, it might give you a clue. Think galvanometer. The mirrors are easier to match for weight before fitting by the judicial use of plasticine.

chriStephens

EDIT the symmetrical fitting of the mirrors might be a problem but we can’t do everything for you. 

Edited By chris stephens on 05/04/2011 22:11:33

[Clive HartlandParticipant @clivehartland94829]

The answer is simple, set it up and using a standard Optical level at each end you will be able to sight through your targets on the bars.

You will of course use a precision level that will measure to 100th of a mm. I think a Leica N3(70) will do, they come with a Plan Plate attachment to give added accuracy.

You might need a rise and fall table under the level to give you your hieghting level.

If I were doing it I would use two levels and set them at right angle to each other and collimate each level to the other so they are the same hieght to suit the two targets on the arms. You will be able to focus right through from end to end of your two targets.

This way you can level both arms at the same time.

If you use this method be aware that any heavy moving vehs. close by or even a tidal area will affect the levels when set up.

 

Clive

 

PS I have just seen your cunning plan so a hieghting table under the levels is not needed as the knee of the Bridgeport will do it for you.

No H & S with an optical level nor with a Class three laser.

It is possible to project a laser through an optical telescope and it is aligned with the optical center. These are available for hire, I can give details if required.

Edited By Clive Hartland on 05/04/2011 22:22:56

 

PS. This method is a ‘No contact method’ so no probes or DTI’s needed

Edited By Clive Hartland on 05/04/2011 22:24:13

[blowlampParticipant @blowlamp]

Another go at this.

1/ Sit the completed unit on the milling machine table.

2/ With a pointer/probe mouted in a drill chuck, position the table and lower the quill, or raise the knee until said probe just touches a suitable reference point on one of the arms and note the reading.

3/ Repeat this operation several times, by spinning each arm into position, to get a good average reading of level for repeat accuracy.

4/ Make suitable adjustments as you go.

 

I think this method is almost immune to outside influences such as an unlevel machine table etc.

 

Martin.

[Les Jones 1Participant @lesjones1]

Hi Gray,

If this device is some kind of levelling device I wonder how free the spherical bearing is. (I have never seen one until I looked on the RS website tonight.) One other non contact way to position “the probe” a defined distance from the arm would be to use electrical capacitance. I remember seeing a design many years ago for device to detect very small changed in the tilt of the earth’s surface. It consisted of two shallow dishes of mercury connected by a thin tube and spaced about a metre apart. There was a metal disc mounted just above the pools of mercury. If the device was tilted the spacing between the disk and the mercury would increase at one end (Decreasing the capacitance.) and decrease at the other. I think a sensor for your purpose could be made with two squares of copper (About 15 mm square.) close together on a piece of printed circuit board. If this was brought close to the arms the capacitance would increase. I will try this sometime over the next few days to see how well it would work.

 

Les.

[Clive HartlandParticipant @clivehartland94829]

Graham, there are many builders Lasers on the market. I have one that has a pencil thin beam that I can see from almost 50 mtrs. It runs on two small AA batts.

and is physically about 9″ long and can have a right angle beam as well. It is set with a bubble and is very accurate.

Cost, about £7.00

Hire cost of a good optical level is about £8.00 a day and it can be delivered and collected.

I have had experience of setting up and adjusting several jigs that are used for the Head up Displays by Marconi, they would come to us annually for setting up. They always came back unadjusted as someone had had a go themselves.

There is always the possibility of Auto Collimation, but that would mean fitting a mirror mount to the job with consequential unbalance problems.

You have not quoted any tolerances, but I would expect that you want a ‘Spot on’ job so mind the draughts as you work!

 

Clive

[blowlampParticipant @blowlamp]

The only potential pitfall I can see now is with the choice of bearing – as pointed out by Les.

All the care in setting up might amount to nothing if it is at all sticky in operation.

It may be that a compass needle type arrangement would be appropriate for minimum stiction?

 

 

Martin.

[Nicholas FarrParticipant @nicholasfarr14254]

Hi martin, although I’ve not actually used these RS bearings, I have used similar types of the rod end variety, and in good condition, although they have enough friction in them to stay put, they are very free to move to any position required with very little force.

 

Regards Nick.

Edited By Nicholas Farr on 06/04/2011 10:51:12

[Les Jones 1Participant @lesjones1]

Hi Gray,

Overnight I thought of a simpler way than the capacitance method. If a tab could be added to the top of each arm near its outer end then a slotted opto device could be mounted on the end of an arm mounted in the spindle. (Or just mounted on the spindle as someone else suggested if the X and Y travel is sufficient to allow the end of each arm to be positioned under the spindle. I you are not familiar with what a slotted opto is it is just an infra red emitter and an infra red detector mounted in a plastic moulding. Anything placed in the slot obscures the passage of infra red between them. If you displayed the current through the detector and moved the device so the tab partly obscured the path to give say 50% of the original reading then this should be quite accurate. To use the capacitance probe method would require careful design of the probe to avoid errors due to stray capacitance as the values being measured would be less than 10 pf If you did decide to use the capacitance method I have seen a simple design design for a capacitance meter with a resolution of 0.1 pf.

Les.

[blowlampParticipant @blowlamp]

Hi Chaps.

 

I haven’t seen the size of the bearing, but I’ve just got this nagging feeling that if it’s got enough friction to self-damp to some degree, then it will more than likely stop in a position that isn’t guaranteed to be consistanly level.

 

Graham.

If you want to guard against your successors’ lack of mechanical sympathy, then rather than a needle point suspension, why not radius the tip ever-so-slightly, so that along with a suitable cup, it would make a tiny ball joint that would be robust enough to stand the odd bit of carelessness?

 

Martin.

[Les Jones 1Participant @lesjones1]

Hi All,

I assume at the moment that the outer of the spherical bearing fits into a hole in the centre of the spider and the support is a 6mm pillar which fits the centre of the bearing. If this was changed so that the outer of the bearing was supported and the spider fixed on top of the bearing on a 6mm rod which extended down through the bearing to act like a pendulum with a weight on the end then I think it would have better self levelling properties.

Les.

[Les Jones 1Participant @lesjones1]

Hi Gray,

I did a very rough test on the capacitance idea. I decided use a single sensor plate rather than the differential one I suggested last night. The sensor plate was made from a 15 mm x 30 mm piece of double sided printed circuit board and connected to the capacitance meter with a short length of coaxial cable. The top face of the sensor plate was connected to the outer of the coax. The outer of the coax and the mill metalwork were connected to the common terminal on the capacitance meter. With the probe about 2 cm from the piece of metal in the machine vice the capacitance meter was set to read zero.

The head on the mill was wound down until the sensor plate looked if it was about to touch the sample piece of metal and the DRO set to read zero. Here are the capacitance readings for various distances from the very rough zero setting.

 

6.0 mm 0.2/0.3 pf

3.0 mm 0.9 pf

1.0 mm 3.1 pf

0.8 mm 3. 9 pf

0.7 mm 4.3 pf

0.6 mm 5.0 pf

0.5 mm 5.7 pf

0.4 mm 6.8 pf

0.35 mm 7.5 pf

0.3 mm 8.6 pf

0.25 mm 9.4 pf

0.2 mm 11.0 pf

0.15 mm 12.8 pf

0.1 mm 15.4 pf

0.05 mm 19.5 pf

 

I think these results indicate that this method could work.

 

Les.

[blowlampParticipant @blowlamp]

Remembering my school days – and just for a lark really, here’s a variation of the idea

http://www.screencast.com/users/Hammering/folders/Jing/media/fdb6c8da-83a0-49ac-9b4a-590170627946

 

Having drawn it, I reaslised it might be useful to make a couple of the Gun Sights (in yellow) adjustable for height, to aid calibration.

 

Martin.

[Michael Ellis 1Participant @michaelellis1]

A 4 tube water level will establish a level plane in space without need to set any reference. Four telescopes (crosshairs?) mounted on vertical stands, 1 at the end of each arm, will optically transfer the level plane to a target on the end of each arm.

 

Bearing hysteresis would be my main concern.

 

You have an interesting problem.

 

Regards, ellis

[MICHAEL WILLIAMSParticipant @michaelwilliams41215]

Don’t understand – if a nominally symmetric object is held by a spherical bearing at the centre then there is no intrinsic reason why it should hang horizontally – it could hang at

at any combination of legs up/down because the object would still be in balance and there would be no restoring force .

 

To consider this problem properly I would be interested to know whether the spider is required to hang :

 

(a) freely in the horizontal plane such that if it is perturbed it moves back to the horizontal position by action of natural or contrived restoring forces .

 

or

 

(b) constrained in the horizontal plane by action of supports such as any combination of conceptual bearings and pins .

 

Notes :

 

(1) When doing any design work where the dynamics/kinetics of the system matter it is always a good idea to establish , if only intuitively , how many freedoms the system has and how many of these freedoms need to be controlled for the system to be stable .

 

(2) The rubber support ball will introduce a force into one of the spiders and an assymmetric force into the spider as a whole even if only making nominal contact . It will be small and it may not matter but it will be there .

[Richard ParsonsParticipant @richardparsons61721]

The answer may be easier than you think. To make certain that the thing is truly level you will need to get its Centre of Gravity below the suspension point. If you hooked 4 identical weights (one on each arm) at the same distance from the centre the thing would then be in balance. The further below the point of suspension the better and the more stable the thing becomes. If you shone a laser (even a cheap one) at a mirror mounted flat on the cross arm, you could see when all movement ceased as the movement of the reflected would be magnified. School-boy stuff, When a mirror is rotated through an angle (say Z° then an incident beam moves through twice that angle (2Z°.

Those confounded smiles are back again.

If you were to mark where reflected laser light ended up on the wall (or where ever) you could remove the weights and use the laser to reset the thing whenever you needed

Dick

Edited By Richard Parsons on 07/04/2011 11:53:56

[Les Jones 1Participant @lesjones1]

Hi Gray,

Going off topic. I have found the article of the “Tilt meter” design (That gave me the capacitance sensor idea.) and scanned it to a PDF file. If you are interested in seeing it PM me with an email address I can send it to. It is claimed to be sensitive enough to detect the tilt of the earth’s surface due to the moon’s gravitational field.

Les.

[Author]

Posts