Nineteen at the bottom, four at the top
The end panel is offered up at the end of the run and held plumb. At the top the gap behind it is 4 mm. At the bottom it is 19. The wall leans out fifteen millimetres over 720, which is nothing anyone notices standing in the room and everything once a flat panel is stood against it. The panel is flat because a beam saw cut it flat; the wall is what a plasterer left in 1974.
Two operations decide where that fifteen millimetres ends up. Levelling puts the room's vertical error into the adjustable legs, where nobody ever sees it. Scribing puts the horizontal error into the back edge of a panel, where it becomes a line touching the wall along its length instead of a wedge of shadow. Everything downstream — door gaps, drawer alignment, whether the worktop sits on eleven boxes or on the two highest — inherits what those two decided, and none of it is fixable later without taking something back out.
What follows assumes the room has been set out already: a level line struck round the walls, the runs chalked on the floor, the services located, the floor build-up written on the plaster. That is a separate afternoon with a laser in an empty room and it belongs to its own guide. This one starts where that one stops — boxes in the hall, marks on the wall, first leg about to be wound.
What a levelled box is level in
Put a spirit level across a base carcase, get the bubble central, and you have proved one thing. There are three independent ways a single box can be out, and two further checks that are not about one box at all, and a fitter who tests only the obvious one produces a run that is level and still wrong.
Along the run is the axis everybody checks, and the failure it causes is not the expected one. Two adjacent boxes can each be level to a bubble and still sit at different heights: a step of a millimetre puts the worktop bearing on one and not the other, and puts two drawer fronts meant to read as a pair out of line at eye level. Level the run as a joined ladder instead — clamp and bolt each new carcase to the one already set, front edges flush, wind it, then check both tops with a straightedge long enough to span them.
Across the depth is the axis nobody photographs. A box tipped backwards by a degree hangs every door out of plumb and, more expensively, throws the runners. Concealed and undermount runners work in a carcase level front to back within the tolerance their manufacturer publishes, and the symptom of exceeding it is a drawer that closes itself, or one whose soft-close never engages. Blum, Hettich and Grass all state a figure and not the same one. A tipped box is what you get for free when a run is pushed hard back against a leaning wall and the legs are left where they were.
Racking is the third and the only one that cannot be corrected later. A flat-pack carcase assembled on an uneven floor, or dragged out of true by a fixing pulled tight into a hollow, becomes a parallelogram: still level, its opening no longer square. A door hung in one tapers against its neighbour whatever you do, because a cup hinge translates a door in three straight lines and cannot rotate it. Check both diagonals of the opening before the back is pinned, and again once the box is bolted to its neighbour.
| What is being checked | What it is checked with | What it breaks if it is wrong |
|---|---|---|
| Height along the run, box to box | A straightedge spanning at least two carcases, referred back to the level line rather than to the neighbouring box | A step in the worktop bearing, and a pair of drawer fronts out of line at eye height |
| Level across the depth, front to back | A short level taken front rail to back rail on every box, not on a sample of them | Doors out of plumb, and runners outside the tolerance their maker publishes — drawers that creep, or soft-close that never catches |
| Square in the face plane | Both diagonals of the carcase opening, before the back is pinned and again after the box is bolted to its neighbour | A parallelogram opening. Every door in it tapers, and hinge adjustment cannot rotate a door into square |
| Straightness of the front edge in plan | A string line pulled the full length of the run at front-rail height, or a laser line where there is one on the job | A shadow line that wanders. It reads from the doorway before any height error does |
| Coplanarity of all the carcase tops together | The same string line moved to the back of the tops, or a straightedge walked across every joint in turn | A stone top bearing on the two high boxes and bridging the rest — the failure the worktop section below exists for |
Where the height comes from, and how much of it actually moves
Finished worktop height is a sum of three components and only one of them adjusts. The carcase is a fixed dimension off a beam saw and the worktop a fixed thickness. Between them sits the leg. The coordinating dimensions that make those three add up to a conventional working height — around 900 mm in European practice, 36 in in North American — are the subject of BS EN 1116, worth knowing because it explains why every maker's boxes land within a few millimetres of each other and why none are identical.
The trap is that a leg's nominal height is not its adjustment range. A leg sold as 150 mm gives a nominal 150 and a published travel either side of it, and that travel is a small fraction of the nominal figure. Häfele and Blum both print it for each leg and socket type, and it differs between systems and between the same system's 100 and 150 mm variants. Read it off the sheet for the legs actually in the boxes, because whether this room can be levelled at all is that one figure against the fall you measured.
Packing goes under the foot, never under the carcase. A packer slid beneath a gable that is not directly under a leg does nothing once the box is loaded — the load path is four point loads through four feet, and the base panel between them is not a bearing surface. Use rigid full-bearing packers, not folded offcuts. Where the floor is soft, meaning cushioned vinyl or an old covering left down, those four point loads find their way into it over the first year and take the run back out of level. That is an argument for cutting the covering away under the legs, not for winding them harder.
- Wind every leg to the same nominal starting height on the bench, so the run arrives at the wall already parallel to itself.
- Set the box standing on the highest point of the floor first, to the level line, legs near their shortest useful setting — everything else winds up from there rather than down into travel it has not got.
- Work outward from it, bolting each new carcase to the one already set with front edges and tops flush, then winding the new one to the line.
- Check across each new pair of tops with a straightedge before moving on: two boxes each level to a bubble can still be a millimetre apart in height.
- Record the leg setting on the highest and lowest boxes in pencil inside a carcase, so whoever lifts this kitchen next knows which way the floor ran.
When the floor falls further than the legs travel
Two numbers settle this and both take ten minutes. The first is the fall along each run: the vertical distance from the level line down to the floor at both ends and at the worst point between. The second is the published travel of the leg system in the boxes. When the first is larger than the second the room has not got a levelling problem, it has a flooring problem, and saying so on the first morning is what makes it cheap.
It is worth being precise about why a floor that has just passed a specification can still defeat a set of legs. BS 8204 describes surface regularity as deviation under a two-metre straightedge — a local property, a measure of how lumpy the surface is over a short span. Departure from horizontal over five metres is a different quantity, and the two are independent. A screed can be perfectly regular under a straightedge everywhere you put it and still fall twenty-five millimetres end to end, because a smooth ramp is smooth. The flatness class does not answer the question the legs are asking.
There are three honest answers. Level the floor, with compound or a screed, which is a day of somebody else's time plus a drying period the programme may not have. Absorb it in the plinth, which works where the fall is modest and the plinth is deep enough to scribe without dying to nothing at the low end — and cannot work where an integrated appliance aperture is involved, since that is measured from finished floor. Or let the run follow the floor, which is not levelling and should be called by its name before anyone agrees to it.
Not for the whole budget — for one decision. Fill in the run lengths and worktop area you already have, put a floor-preparation figure from a local screeder into the finishes line, and run it twice: once with that figure and once with it at zero. The difference is what levelling the floor costs as a proportion of a kitchen you are about to hang off it for twenty years, which is the only form of the question worth arguing about. A discovery like this is what the contingency line is for, so if it fits inside the contingency already allowed, the argument is over.
Length of the base-unit run.
Carcass, door, drawer and fitting per linear metre.
Length of the wall-unit run.
Typically 55-65% of the base rate.
Worktop area.
Including template, cutouts and fitting.
Everything you are buying, as one figure.
The work behind the walls.
Everything after the units are in.
Kitchens hide more than most rooms.
Total kitchen cost
Needs your rates
This page does not assume a price. Enter yours and the answer appears here.
They open the calculator with your figures already in it
Kitchen Remodel Cost Calculator — shown in imperial, US market. The link sets both, so the result they see is the one on your screen.
What this calculation does not cover
- Excludes structural work — removing a wall, forming an opening, or moving a soil stack are separate and much larger items.
- Excludes the cost of living without a kitchen, which on a long fit is a real expense people do not budget for.
Scribing is subtraction, and there is no second cut
Five things get scribed in a normal kitchen, to three different surfaces. End panels and fillers to a wall that leans or bows. Plinths to a floor that falls or rolls. Cornice, pelmet and light rail to a wall or occasionally a ceiling. The worktop's back edge to the wall as well, though on laminate that operation belongs with the jointing and is covered there. What they share is that everything taken off is gone, and that the cut is made against a line existing only while the panel is held in the position it will finish in.
The geometry differs from coping a moulding into an internal corner, the scribe most carpenters learn first and the one the skirting guide works through. There the shape being followed is the moulding's own profile and the wall is incidental. Here the shape is the wall, the panel is flat, and the operation is the transfer of an irregular line onto a straight edge. Set dividers or a scribing block to the widest gap anywhere along the junction — not the average, and not the gap where you happened to hold it — hold the panel plumb at its final projection from the run, and run the point along the wall with the pencil following on the panel. Every part of the panel then loses something, which is what makes the line continuous.
Cut on the waste side and undercut behind the face. A back-bevel of a few degrees means only the front arris touches, so the joint closes tight along its visible edge and the material behind is not fighting for the last half millimetre. A down-cutting jigsaw blade takes the bulk off a melamine-faced panel without chipping it; a block plane or a belt sander takes the rest. Offer it up dry, find the two or three high spots holding it off, and take those back rather than running the whole edge again.
The last point turns a scribe into a layout decision. A scribe removes only material that is there to remove. An end panel has a limited width to give up before the run behind it shows, and a filler is only as wide as the one you were sent. Measure the widest gap before cutting anything, top and bottom, and compare it with what the strip has got. Where the wall eats the whole allowance the answer is to move the run, order a wider strip or take a unit out — not to cut the panel to nothing at one end and hope. A phone call on the day, a return visit if it is left.
- Set and fix the run's front line first, so the panel is scribed from a position that will not move afterwards.
- Hold the panel plumb in both directions, packed off the wall by the widest gap, before a mark is made.
- Lock the dividers at that widest gap and mark the waste side clearly — a cut on the wrong side of the line adds the wall's shape to the panel instead of taking it off.
- Cut proud, undercut behind the face, bring it back on the dry fit, and seal any raw core edge before the panel goes on for good.
The screw has to reach something
Wall units are the only part of a kitchen capable of injuring somebody, and the load is not the one people picture while hanging them. A run of them fills with crockery, tins and small appliances, and the fixings end up in axial withdrawal from a vertical surface with a lever arm out to the cabinet front. ANSI/KCMA A161.1, the North American standard for kitchen and vanity cabinets, includes a mounting load test for exactly that condition, and BS EN 14749 covers the same ground in Britain and Europe. Neither tests your wall. The cabinet arrives already proved; the fixing is the part nobody certifies.
Which anchor suits which substrate, how to find the timber and what a cavity fixing is genuinely rated for are worked through on the guide to fixing heavy things into plasterboard, and none of it is re-argued here. Two consequences are specific to a kitchen. The first is arithmetic: on dot-and-dab lining the screw has to cross the board, cross the adhesive void and still reach full embedment in the masonry, so its length is the sum of three numbers and the middle one comes from a test hole rather than an assumption, because it varies along one wall. The second is why a proprietary hanging rail earns its price, and it is not that the rail is stronger. It converts fixings pinned to cabinet-side positions into a continuous member you can fix wherever the structure actually is — which, on a wall where the studs, the dabs and the cabinet joints do not agree, is the difference between four good fixings and two.
Before any of it, know what is in the wall. Section 522.6 of BS 7671 sets out what is required of a cable concealed in a wall at less than 50 mm from the surface — the permitted zones it may run in, and the protection required outside them. In the United States, Section 300.4(D) of the National Electrical Code requires cables parallel to framing members to sit an inch and a quarter back from the face or be protected by a steel plate. Both assume somebody will drill, and neither helps where the run was put in by a previous occupant. Photographs taken before the boards went on are the cheap answer; a detector and a shallow first hole is what is left when nobody took any.
| The joint | What carries the load | How it fails |
|---|---|---|
| Carcase to carcase, gable against gable | Connecting bolts or screws through both gables, pulled up with the front edges clamped flush | A run that flexes between boxes and a step at the worktop bearing — the reason a run is levelled as one ladder |
| Base carcase to the wall behind | A bracket or screw into the substrate, holding the box against being pulled forward rather than holding it up | A box that tips when a loaded drawer is fully out, which is the moment nobody tests for |
| Wall unit to a hanging rail, and the rail to the wall | The rail, and whatever the rail's own fixings ended in — the cabinet brackets only ever reach the rail | Creep. It loads gradually over months and lets go without warning, which is why it is the one worth calculating |
| Island block to the floor | Battens or brackets screwed down to the substrate, with the block square and level before either goes in | A block that shuffles under a leaning elbow, taking the sealant line round the top with it |
| Worktop to the carcases below | Screws or brackets up through the rails, sized so nothing breaks the surface — the jointing guide sets that out | A dimple telegraphed through the laminate, or a top free to lift at one end |
| Plinth to the legs | Clips engaging the leg, fitted after the plinth has been scribed to the floor | A plinth that reads as fitted and drops off when somebody vacuums, because a clip was pressed near rather than home |
What the hours actually attach to
Kitchen fitting is quoted by the unit and by the metre, and neither is what the day is made of. Two kitchens with the same schedule of modules can be a day apart: one into a square new-build shell with a power-floated floor and studs the drawings locate for you, the other into a room with a floor falling twenty millimetres, three walls leaning different ways and a corner at eighty-seven degrees. Same boxes, same doors, same delivery note.
So count what the work is actually made of. Every scribed junction is an offer-up, a mark, a cut, a dry fit and a second cut. Every internal corner is a set-out decision plus a filler. Every integrated appliance is an aperture checked against a datasheet and a door aligned to the cabinet doors either side of it. Every separately levelled run is its own datum transfer and its own two ends. And some operations need two people whatever the price says — a full-height unit onto a rail, a stone top, a tall housing walked past a doorway. Those are the hours; the carcases are the fastest part of the day.
Build the hours from the junction count rather than the unit count: so many hours of levelling and fixing, plus a figure per scribed junction, per appliance aperture and per corner, taken from your own last three jobs rather than from a book. Enter that as hours per person. Then set crew size honestly — once at one for the scribing and alignment, which is a solo trade, and again at two for the half day of lifting that genuinely needs a second pair of hands, and add the two answers. A single run at crew size two prices a second fitter standing about while somebody sets dividers.
The total hours the job is expected to take, per worker.
The rate charged (or paid) per worker, per hour.
The number of workers billed at this hourly rate.
Total crew-hours
40 hours
Figures that depend on a rate wait for yours — this page does not assume one.
They open the calculator with your figures already in it
Labor Cost Calculator: 40 hours — shown in imperial, US market. The link sets both, so the result they see is the one on your screen.
What this calculation does not cover
- One rate is multiplied across every hour and every worker, so there is no tier for overtime or holiday premiums, night and weekend differentials, or a crew that pairs a licensed lead with an apprentice — a mixed-rate job has to be totalled in separate runs and added by hand.
- Crew size acts as a straight multiplier on the hours you entered, which assumes each additional worker stays productive for the full duration: the tasks that will not split across two pairs of hands, the time a crew loses coordinating, and the helper who is only on site for part of the week all leave the total untouched.
- Nothing distinguishes a wage you pay from a rate you are charged, because the same multiplication runs on either. A figure built from raw wages carries no payroll taxes, workers' compensation, insurance or benefits on top of it, while a contractor's quoted rate may already have overhead and profit buried inside — the answer looks identical in both cases.
- Only worked hours are priced. Travel and mobilization, setup and clean-up, tool or equipment hire, disposal, permits and materials all sit outside the figure, and no minimum charge is imposed either — an entry of half an hour returns half an hour of money on a job many trades would bill as a minimum visit.
- The hours you type are taken exactly as they stand, with no contingency for rework, weather, waiting on an inspection or scope that grows once the walls are open, and the rate is held flat for the whole span — a long program approaching the 2,000-hour entry ceiling is still priced at today's number, with no escalation partway through.
The plane the top is about to lie on
Every carcase being individually level is not the same statement as all their tops lying in one plane, and it is the second the worktop cares about. Pull a string line along the back of the tops, or walk a straightedge across every joint in turn, and the box that is a millimetre and a half low shows itself. On laminate that box is a soft spot someone notices leaning on it. On stone it is a bridge: the slab bears on the two high boxes and spans the low one, and the stress arrives at the thinnest section of the slab, the strip between the sink cutout and the front edge.
That is why the last honest chance to wind a leg is before the templater arrives, and why the sequence has the whole run fixed, levelled and proved before anybody measures. Support requirements — continuous bearing, brackets to a proprietary specification, maximum unsupported overhang, the minimum section a fabricator will warrant between two cutouts — come from the fabricator's own installation requirements and the Natural Stone Institute's Dimension Stone Design Manual, and are worth reading before the boxes are fixed.
The levelled run has also just settled the depth of the top, and that is arithmetic rather than specification. A standard 600 mm postformed depth contains four things: the carcase depth, the front standing proud of it, the overhang the top projects past that front, and the gap left at the back for a scribe or an upstand to cover. They add to 600 and no more. Every millimetre you pack a box off the wall to clear a bow comes out of the back gap and lands on the front overhang, and if that varies along the run the top's front edge no longer follows the doors'. Set the front line, hold it, and let the back gap be the variable.
An island removes the wall and with it every place to hide. There is nothing to scribe to, all four sides are visible, the carcases have to be square to each other rather than to a wall that is not square anyway, and the level has to be true in two directions because the block is looked at from two of them. It is fixed to the floor rather than to anything vertical, which makes the floor's own flatness a sharper constraint than it is against a wall. Its top is a different depth calculation too — carcase block plus an overhang on each exposed side, plus whatever the stools need — and it should be ordered as its own piece.
Run each top separately, because they do not share a depth. For a wall run, put in the length through the corners and set the depth to what the boxes are actually standing at once packed off the wall, not the nominal figure from the brochure. For an island, run it again on its own: length as the block, depth as the carcase block plus the overhang on each side, and backsplash height at zero, because there is no wall behind it. The area returned is gross with nothing taken out for cutouts, which is what a fabricator quotes against at enquiry — the template that gets cut comes off the levelled boxes on a later day.
The total length of counter along the wall(s), including corners.
Standard kitchen counter depth is about 25-26 in (64-66 cm).
4 in (10 cm) is a standard minimal splash; set higher for a full backsplash to the cabinets, or 0 for none.
Estimated countertop area needed
31.96 sq ft (total)
- Countertop surface
- 27.63 sq ft
- Backsplash
- 4.33 sq ft
They open the calculator with your figures already in it
Countertop Area Calculator: 31.96 sq ft (total) — shown in imperial, US market. The link sets both, so the result they see is the one on your screen.
Estimated cost — your price
This site holds no price list for this material — local prices vary too much to publish honestly. Enter your supplier's price and the result is costed with it.
What this calculation does not cover
- This is GROSS area, with nothing deducted for the sink, the hob, tap holes or drainer grooves — which is how a fabricator quotes at the enquiry stage, because the slab has to be bought whole and the cut-outs are labour rather than savings.
- It is not a template. Stone and engineered stone are cut to a template taken in the finished room once the base units are in and levelled, and that measurement, not this one, is what gets cut.
- Seams, edge profiles and the extra material a book-matched or heavily veined slab needs are priced separately and can move the slab count without moving this area.
Doors last, and never to correct a box
A European cup hinge adjusts in three axes — depth, height and lateral — each a straight-line translation of the door relative to the carcase. The makers publish which screw does what and in what order, and the order matters because two of the three interact. What no hinge does is rotate a door into square. Take that as the operating rule for the whole stage: if a reveal tapers top to bottom, the box is out and the hinge is not the tool. Check the diagonals, check it is plumb across the depth, correct the box, and only then touch a hinge.
Reveals — the gaps between adjacent doors and drawer fronts — are trade convention rather than standard, commonly around three millimetres on contemporary slab fronts and set by the system's own drilling pattern in any case. The figure matters far less than holding one figure across every gap in the room, which is why it belongs on a gauge block rather than in the eye of somebody who has been in this room since seven.
Then do the opposite and use the eye deliberately. Stand in the doorway, at the distance someone entering will stand, and look along the run in whatever light the window gives. A taper of half a millimetre over a 700 mm door is invisible with a tape and obvious in raking light. The same walk finds a drawer front proud of its neighbours, a cornice mitre with a shadow in it, and a plinth scribed at one end and left square at the other.
- Hang doors and fit drawer fronts only once every carcase is fixed, levelled and proved square, and after the worktop is on where the programme allows.
- Set the depth adjustment on every hinge in a run to the same starting position before adjusting any individually.
- Work left to right, setting each door against its neighbour with a gauge block rather than against the carcase edge.
- Check each reveal top and bottom — a parallel gap of the wrong size is a hinge job, a tapered one is a carcase job.
- Run every drawer to its stop and back before the plinths go on and the runners stop being reachable.
One kitchen, three or four separate runs
A fitting price per metre of run is an average, and averaging is the wrong operation on a kitchen because the work is concentrated at the ends. A five-metre straight run against one wall has two ends, one datum transfer and one front line. Break the same five metres into a two-metre return, a galley wall and an island and you have three of everything, six ends instead of two, and an island with no wall to scribe to. The metres are identical; the days are not.
So split the job before pricing or reviewing it. Take each run separately — its length, its scribed junctions, its corners, its appliances — and work out what a metre of that run carries. Three numbers differing by a factor of two tell you something you can plan with. One averaged number tells you the total, which you already knew, and hides which part of the room will overrun.
The linear items follow different lengths again, which is the second reason a single per-metre figure misleads. Plinth follows the toe recess rather than the carcase face and returns at every exposed end, so an island's goes all the way round while a run between two walls has none at its ends. Cornice and pelmet follow the wall-unit run only, usually shorter and broken in different places. Scribe strips exist only where a run meets a wall. Each is bought in stock lengths and cut with waste at every mitre, and none of their lengths is the number on the cabinet quote.
Use it three or four times rather than once. Enter the whole cost of one run — the fitting hours for that run plus its own trim, fillers and end panels — against that run's own length, then do the same for the galley wall and the island. What you want is not the average but the spread between them, because the highest is where a day gets lost and the lowest is the one people quote from. Keep the segment figures on the sheet after you have a single price, since they are the version that still means something on the next kitchen.
The total quoted or actual cost for the whole run.
The total length the cost covers.
Cost per linear foot
20 $/linear ft
- Total length
- 150 linear ft
They open the calculator with your figures already in it
Cost Per Linear Foot Calculator: 20 currency/linear ft — shown in imperial, US market. The link sets both, so the result they see is the one on your screen.
What this calculation does not cover
- A per-foot rate carries no specification. Fence height, post spacing and embedment depth, picket gauge, gutter profile and trim size all change what a foot costs while leaving the run length identical, so two quotes only compare on this number when they describe the same product. A 6 ft board fence and a 4 ft one around the same perimeter are different rates, not a better and a worse price.
- The ground the run crosses is not in it. Rock, tree roots, slope, a hand-dig where no machine can reach and spoil that has to be carted off all sit in the labor half of a per-foot price. A rate lifted from a job across flat lawn does not transfer to a run along a bank or to a boundary a mini excavator cannot reach.
Before the van goes
The last half hour is worth more than any before it, because everything found in it is still cheap. Open every door and drawer to its stop. Put a flat hand across each carcase joint in both directions, since a fingertip resolves a step an eye standing up does not. Look along the front edge from the doorway. Push the run at one end and see whether anything moves. Check every plinth clip has engaged rather than been pressed near.
Then leave the room able to explain itself. Write the finished worktop height and the plinth height, both referred to the level line rather than the floor, in pencil inside a cupboard where the next trade will find them, and note which end the floor fell towards. Somebody will lay a new covering here in eight years and have to decide whether the dishwasher still comes out; that note and the appliance instructions left in a drawer are the difference between a five-minute answer and a cut floor.
Six measurements taken while the boxes are still loose
All of them before a single fixing goes into a wall, and every one capable of changing what happens next rather than merely describing what already went wrong.
- Fall along each run, from the level line to the floor at both ends and at the worst point between — Against the published adjustment travel of the leg system in the boxes, which is a small fraction of the leg's nominal height and is printed in the maker's technical data. Larger than the travel and this is a flooring job before it is a fitting job.
- Widest gap between the levelled front line and each wall, measured at the top and at the bottom — A leaning wall gives two different numbers and the scribe has to be set to the larger. Compare it with the width of the scribe strip or end panel you were actually sent, before anything is cut.
- Both diagonals of every carcase opening, and of the room where an island is going — A racked box is a parallelogram and no hinge will square a door in one. The room's diagonals decide whether an island can be parallel to two walls at once, which it usually cannot.
- Dab void depth behind the plasterboard at each wall-unit fixing position — From a test hole, not from a drawing. The fixing has to cross the void and reach full embedment in the masonry behind, so screw length is board plus void plus embedment — and the middle term is not constant along one wall.
- Deviation across all the carcase tops together, taken on a string line rather than box by box — Individually level is not coplanar. A stone top bears on the high boxes and bridges the low one, and the stress from that span arrives at the thinnest section of the slab, beside the sink cutout.
- Front-to-back level on every base carcase, against the runner maker's stated tolerance — Blum, Hettich and Grass each publish a figure and they are not the same figure. A box tipped past it gives drawers that creep or soft-close that never engages, and the cause is invisible once the fronts are on.
Opens the calculators above on one screen with the dimensions from this article already filled in. Quantities only — this site publishes no price list, because local prices vary too much to publish honestly.
