PROJECT DESIGN REVIEW — MINUTES OF MEETING
Project: heavy gantry machining centre, 6,400 kg total mass
Review subject: center of gravity position, foot print, leveling feet and floor loading
Date: 22 July, 10:00 to 11:50, meeting room 4
Document ref: DMR-2026-0722-COG08
Attendees
| Name | Role |
|---|---|
| W. Bauer | Lead machine designer (chair) |
| D. Petersen | Foundation and installation engineer |
| E. Marchand | Workshop rigger and installer |
| L. Hartley | Structural / FEA engineer |
| A. Kaur | Site services and layout |
1. Purpose of the review
The gantry carries a 900 mm column and a horizontal spindle on a 3.2 m cross rail, and the whole machine rests on six leveling feet on a 220 mm concrete slab. During the design check the model showed the combined center of gravity sitting close to the rear foot line, and the installation engineer wanted the number before the machine was shipped, because a machine whose mass sits near its footprint edge lifts its front feet under acceleration and loses its level. The review was called to fix a leveling and foot-loading strategy before the slab was poured, not after the machine twisted the rail.
2. Review comment CG-01: where the center of gravity actually sits
L. Hartley walked the meeting through the mass distribution, reading from the 3D model table. The column and head, 1,450 kg together, hang at the rear of the machine around 1,100 mm above the floor; the bed and the two platen supports carry 2,180 kg; the cross rail and carriage 1,260 kg; the coolant tank and hydraulics 980 kg; and the enclosure and cladding the remaining 530 kg. The weighted sum put the combined center of gravity 620 mm from the rear edge of the footprint and 310 mm above the floor line, on a footprint 1,600 mm deep by 1,100 mm wide. The number that mattered was the margin, the distance from the center of gravity to the front foot line, which came to 380 mm on a base designed for a 260 mm margin under the previous model. The rear-loading was real, and the front feet carried only 18 percent of the weight while the two rear feet carried more than half.
3. Review comment CG-02: what happens when the mass sits near the rear line
E. Marchand described the failure that had happened on an earlier machine with the same rear-hung column. Under a rapid X-axis move at 14 m/min-squared acceleration, the inertial force at the head height pushed the effective load line back, and the two front feet unloaded to near zero while the rear feet saw 1.6 times their static load. The machine leveled the moment it was trammed and then lost 0.12 mm of level across the Z-axis within the first week as the concrete under the rear feet crept. The review wrote the principle above the numbers: the static center of gravity must sit inside the footprint with a margin of at least fifteen percent of the footprint depth on every side, and under dynamic load no foot may unload below ten percent of its static share.
4. Review comment CG-03: the leveling foot count and the three-point rule
D. Petersen opened with the arithmetic of flat machine bases. A steel base is only ever supported at three points statically, because any four contacting points leave one high and one low in practice. On a six-foot machine, the review must decide which three feet carry the machine and how the other three are set. The proposal before the meeting was to level on the rear pair and the front centre foot, establishing the reference plane on those three, then snug the remaining three feet to a stable 0.01 mm contact without lifting the reference. E. Marchand agreed with a grumble learned from a machine where all six feet were levelled hard, which had bent the bed by 0.18 mm and twisted the rail until the carriage bound at the far end.
5. Review comment CG-04: the concrete slab carries the machine more than the feet do
A. Kaur brought the floor into the review, because a leveling foot standing on a soft patch of slab is a foot that drifts. The site floor is a 220 mm slab of C25 concrete with a measured surface tolerance of 4 mm over the 2.1 m machine footprint, and the specification for this machine calls for the leveling feet to sit on poured pads, not directly on the slab. The pads are to be cast 250 mm square by 80 mm thick in the same pour as the slab, vibrated, and topped with a setting compound at the moment of installation. The review turned the pad design into a number: the machine weight of 6,400 kg divided by six feet gives 1,070 kg per foot statically, rising to 1,700 kg per foot under the dynamic rear condition, and a 250 mm square pad at 0.6 MPa bearing stress is inside the 2.5 MPa the compound allows with the anchor bolts holding the foot stub down.
6. Review comment CG-05: anchor bolts on the rear feet, dowels only in front
D. Petersen read the fastener plan and the room split over the front feet. The proposal anchored all six feet with M16 chemical anchors, because a rear-heavy machine with fast rapids needs the rear foot stubs pulled into the pads or the machine will walk across the floor under reversal loads. E. Marchand objected to anchoring the front feet, because the front pair is the one side that must stay free to take up thermal growth of the base along its length, and welded or anchored feet turn that growth into a bowed bed. The resolution was a split plan: the two rear feet get M16 chemical anchors torqued to 90 N-m into the pads, the two mid feet get dowel pins, and the two front feet sit on the leveling screws without anchors, free to slide a few hundredths of a millimetre as the bed warms.
7. Review comment CG-06: the center of gravity could also be moved, not only held
L. Hartley asked whether the rear-loading could be engineered out instead of managed. The 1,450 kg column and head could be moved forward on the bed by 150 mm if the bearing spread on the column slide allowed it, which would pull the combined center of gravity forward by 34 mm on the 6.4 tonne machine and lift the front foot share from 18 to 22 percent. The cost was a tighter head travel envelope at the rear and a revised coolant tank position. The review weighed the 34 mm against the effort and decided the anchor and pad plan was enough today, with the column position logged as a possible change for the next model rather than a change to this one.
8. Disagreement: the leveling tolerance, 0.02 mm per metre or 0.05
The review argued fourteen minutes over the level tolerance because the number decides how long the riggers stay on the job. E. Marchand said 0.02 mm per metre across the base is the number that keeps the 3.2 m rail within its twist spec, and the riggers can hold it with a precision level and a two-hour settle. L. Hartley said the machine can run inside its accuracy at 0.05 mm per metre, because the rail twist is dominated by the bed casting, not by the residual level, and the tighter number adds a day to the installation only to protect a tolerance the control system corrects anyway. The vote came to the engineering evidence rather than the vote count: the previous model was trammed at 0.02 mm per metre and still drifted 0.12 mm under the rear feet within a week, which proved the level itself was not the failure. The 0.05 mm per metre was adopted with the note that the level check is repeated after the anchors are torqued and again after the first heated week.
9. Risks and mitigation
Four risks were logged. Risk 1: the slab pads could be cast in the wrong place when the machine bay layouts shift; mitigation is a pad template drawing issued with the foundation package and a pre-pour check against the machine footprint. Risk 2: the rear anchors add a constraint that fights the thermal movement of the base; mitigation is the 90 N-m torque with a collar washer allowing a measured slide, and a re-check of level after the first shift. Risk 3: the front feet free to slide could walk on the pads under reversal; mitigation is a 0.2 mm height lip on the pads and a quarterly foot position check. Risk 4: the centre of gravity number was computed from the model and not measured; mitigation is a corner-weigh at install using four load cells under the feet, with the rear-to-front ratio logged.
10. Actions and owners
| ID | Action | Owner | Deadline |
|---|---|---|---|
| ACT-1 | Issue pad template drawing with foundation package | A. Kaur | 25 July |
| ACT-2 | Cast 250 mm square pads, vibrated, under all six feet | D. Petersen | 8 August |
| ACT-3 | Use three-point reference level at 0.05 mm per metre | E. Marchand | 12 August |
| ACT-4 | Anchor rear feet M16 at 90 N-m, dowel mid feet, free front feet | E. Marchand | 13 August |
| ACT-5 | Corner-weigh with four load cells, log rear-to-front ratio | L. Hartley | 14 August |
11. Decision summary
Level on a three-point reference at 0.05 mm per metre. Cast reinforced pads under all six feet. Anchor the two rear feet with M16 chemical anchors, dowel the mid pair, leave the front pair free. Verify the machine rear-to-front foot load split by corner-weighing after installation. Re-check level after anchor torquing and after the first heated running week.
12. Post-meeting addendum, 5 September
The gantry sat on its pads and the corner-weigh came back at 52 percent rear, 30 percent mid, 18 percent front, close to the model’s 54 / 28 / 18 split. The level held at 0.03 mm per metre after the first week, and the X-axis ran its full 3.2 m with no binding at the far end. The riggers finished a day early, and the only change logged for the next model was the column forward shift that would lift the front share toward 22 percent.
Glossary
Center of gravity: the single point through which the whole machine weight acts. Footprint: the projected area between the outer leveling feet. Three-point reference: the three feet chosen to define the level plane on a machine base. Corner-weighing: measuring the load under each foot with load cells to locate the true center of gravity and check the split.
Appendix. The foot-load worksheet the review signed
Mass break by module, all from the 3D model: column and head 1,450 kg at rear, bed and platen supports 2,180 kg, cross rail and carriage 1,260 kg, coolant and hydraulics 980 kg, enclosure 530 kg, total 6,400 kg. Combined center of gravity 620 mm from the rear edge on a 1,600 mm footprint, 380 mm to the front foot line, 310 mm above the floor. Static foot load at 6,400 kg over six feet: 1,070 kg per foot; dynamic expression under 14 m/min-squared with the rear-hung head: rear feet 1,700 kg, mid 1,000 kg, front 400 kg, which the front pair at 12 percent of static fails the ten percent floor only on paper and passes with the pad lip holding them. Anchor plan: rear pair M16 chemical bolts at 90 N-m, mid pair dowel pins, front pair free on leveling screws with a 0.2 mm pad lip. Level plan: three-point reference at 0.05 mm per metre, re-checked after anchor torquing and after the first heated week.
The chair’s closing remark was written in the minutes as agreed: a machine is leveled once on install and again whenever its mass or its footing changes, and the cheapest insurance on a heavy gantry is a center of gravity drawn on the layout drawing before the slab is poured. The minutes were signed by all five attendees, and the foundation package for the cell now carries the pad template as a controlled drawing.
