How Do You Specify a Wine Bag That Arrives With the Bottle Intact?

How Do You Size a Wine Bag for 750 ml, 1 L and Magnum Bottles?

A wine bag is specified from the bottle outwards, and it is a packaging component before it is an accessory. Every other decision in this document — the cell width, the divider thickness, the drop height, the handle width and the bill of materials — sits downstream of two numbers on the glass: its height and its widest diameter. Both are available from the glass supplier and both belong on the first two lines of the tech pack, in millimetres, with the shape named.

The trap is that the volume is the same across shapes and the glass is not. A Bordeaux bottle and a Riesling bottle both hold 750 ml; one is short and wide, the other is tall and narrow, and a carrier built to the first will not close over the second. Under the European rules on nominal quantities for prepacked products, 750 ml is a prescribed fill, so the estate can change glass supplier and the volume will not move while the height moves by 40 mm. Ask for the drawing of the bottle that is actually in the case, not for the size printed on the label.

Two things follow from that table and they are the two worth writing down. The internal cell needs 8 mm of space over the widest glass in the range you are shipping rather than over the one in the drawing, and that figure is the working minimum — below it a bottle will not drop in when the label is wet. And the height has to be set by the tallest shape: a bag sized to the tallest carries every shorter one, and a bag sized to the shortest does not carry the tall one at all. The Riesling row is the one that decides, and it is the row buyers read last.

The six-bottle carrier deserves its own paragraph because it is a different object from a two-bottle carrier and it is usually specified as the same one. Six bottles of 750 ml is 7.5 kg of glass plus the bag, held by one or two handles and by a base panel. It is a materials-handling object with a print on it, and the handle construction and the base are the specification. A six-bottle carrier that is a two-bottle carrier with four more cells will come back, and it will come back with the base first.

Cardboard, EVA or Foam: What Should the Divider Be Made Of?

The divider is the structural member. It is usually specified as a separator and it usually fails as a structure: it carries the compression when the bag is stacked, it carries the impact when the bag is dropped, and it is the thing that stops two bottles touching when the bag is tilted. A divider that is a decorative insert will pass an inspection and then fail in the parcel network, which is the worst of the two places to fail.

The choice between them is really a choice about water. Paperboard is rigid, printable and it loses its shape when it is wet; EVA and PE foam are unaffected by water and they are not rigid on their own. A chilled bottle in a warm room produces condensation, and a paperboard divider in a reusable carrier will be soft inside a season. That is the whole decision: a presentation bag carried once, dry, is paperboard; a reusable carrier that will meet condensation, a wet case and a cold warehouse floor is EVA or foam. Everything else is secondary to that one fact.

Thickness is the second half, and it is where most of the sample-to-bulk drift happens. Two millimetres of laminated paperboard is adequate in a two-bottle carrier and it is the normal figure; the same sheet in a six-bottle carrier will bow, because the span is longer and the load is three times higher. The rule is that thickness follows the span, and the span is the cell width multiplied by the number of cells. A six-bottle carrier with a two-bottle divider is a different product with a different requirement, and it is quoted as one because it looks like one on a drawing.

The assembly matters as much as the material. A slotted divider, cut and interlocked and fitted into a base board, holds its geometry under load. A folded divider glued into the shell and held by the lining does not, because the load path then runs through adhesive and through a seam, and a seam is the weakest thing in the bag. And the divider has to be full height. A divider that stops 30 mm below the rim looks identical on the drawing, and it lets the bottles tilt until glass meets glass — which is precisely the damage the divider existed to prevent.

One more, and it is the one that is most often omitted. The divider has to be attached to the base rather than floating inside it. A floating divider converts every handling event into a shearing load on the joint between the divider and the shell, and that joint is a seam. Slot it into the base board, fix the base board to the shell at more than one point, and the load goes into the structure where it belongs.

How Is a Drop Test Actually Set Up for a Bottle Bag?

A wine bag is not qualified by looking at it. The test that matters is a free fall of the filled bag, and it has a published method: ASTM D5276 covers the drop of loaded containers. The reason to cite the method rather than a height is that a height alone does not say which faces, in which order, on how many samples, or what counts as a failure. Four people can each drop the same bag from the same height and reach four different conclusions, and all four will be defensible.

The orientations matter more than the height, because glass breaks at a corner and fabric opens at a seam. The base orientation tells you whether the base panel and its attachment hold. The corner orientation — the most vulnerable bottom corner, which is the diagonal, and usually the one with a seam under it — is the one that breaks the bottle. The flat side tells you whether the divider is holding the bottles apart. The handle orientation tells you whether the attachment holds, and it is skipped nearly every time, because it is the one that fails.

The pass criterion is the bottle, not the bag. Three conditions, and all three in the inspection instruction: no bottle breakage, no opening of a seam or a closure, and no glass exposed through the shell. A bag that lands with its base seam opened 40 mm and the bottle intact has still failed, because the next event will release the bottle. That is the point of writing the criterion down — it is not a cosmetic judgement, it is the question of whether the bag still contains what it was holding.

Two discipline points decide whether the report is worth paying for. Fill the bag with what it will carry: bottles filled to nominal with water, capped, at the weight of the range — because glass failure is a function of contact geometry and not of mass alone, and a substitute mass does not have the shape of a bottle. And use a fresh sample for each orientation, one drop each, because damage accumulates and a bag dropped on its base and then on its corner was tested for the second one with a damaged base.

There is one test that is not a drop and that prevents more complaints than any of them: the pack-out. Six bottles in the bag, tipped to 45 degrees, carried across the room and set down on a table. The cells that are too wide will let the bottles roll and touch; the divider that is too short will let them tilt; the handles that are too long will drag the rim along the floor. It takes a minute and no apparatus, and it will find all three before the drop rig does.

Finally, if the bag is going through a parcel network, the network may hold its suppliers to a transit procedure of its own, and ISTA 3A is the one usually named for individual parcels. Ask the carrier which procedure it enforces and test to that one, because a drop test the carrier does not recognise is a drop test you will end up running twice, at two different heights, with two different answers.

What Holds a Wine Bag Together at the Handle and the Base?

Two parts carry a wine bag and they are the two that come back: the handle and the base. Both are usually specified by description — "a webbing handle", "a reinforced base" — and both have to be specified by construction, because each of those descriptions covers two things with very different strength. A handle is not a piece of webbing; it is a piece of webbing plus a load path. A base is not an insert; it is an insert plus a fixing.

The handle question is a question about the load path, and it has one answer at this weight. Two handles stitched to the side panels put the load into the seam in peel, and peel is the weakest direction a seam has. A continuous loop that runs under the base and up both sides puts the same load into the webbing in tension, and webbing in tension is very strong. The continuous loop costs more, it requires a base strong enough to be part of the load path, and it is the difference between a bag that carries six bottles and a bag that is carried empty.

• The attachment, and it is a bar-tack, a rivet or a stitch-through into the base panel. A handle tacked to the lining alone is not attached to anything that holds a load.
• The hand gap, which is the space between the handle and the rim. Below 90 mm a hand in a glove will not fit, and the bag gets carried by the rim instead, which is a separate failure.
• The width. A 25 mm webbing handle under a 7.5 kg load is a strap that cuts; 38 mm, or a padded sleeve, is the specification at that weight.
• The base insert, 2-3 mm, cut to the same footprint as the cells and fixed to the shell at more than one point.
• Feet on a rigid presentation bag, because a rigid bag is set down rather than dropped, and a bag that is set down wears its base.

The base is the part that gets left out and it is the part that takes the drop. A base board floating inside a fabric floor will shift, and once it has shifted the cells are no longer square and the bottles are no longer separated. It needs to be the same footprint as the cells, it needs to be fixed to the shell at more than one point, and on a rigid presentation bag it needs to be the surface the bag stands on. On a six-bottle carrier the base insert and the handle have to be designed together, because the handle loads the base and the base takes the impact.

There is a version of this worth naming, because it is the most common quiet failure in the category: the padded base that is a piece of foam in a fabric sleeve with no board in it. It feels right in the sample and it does not distribute the load of six bottles, because foam spreads a static load and does not carry a point load. Put a board in it and the foam becomes comfort instead of structure, which is the job it is good at.

Cellar-Door Carrier or Presentation Bag: Which Format Does the Order Need?

Four formats cover the category and they are four products. The one thing they share is the bottle; the construction, the finish, the run driver and the unit cost are all different. The expensive mistake in this category is not a bad bag, it is the right bag bought for the wrong job — and it is a mistake that is only visible after the goods have landed.

The cellar-door carrier and the presentation bag are both retail objects and they are usually specified as one. The carrier is bought by the estate and used by the customer for a year, so it needs a divider and a base that survive carrying. The presentation bag is bought for an occasion and used once, and its value is on the outside — a foil crest, an emboss, a ribbon — so it needs a surface that takes a finish and a rigid form that holds it upright on a shelf. Same artwork, two constructions, two prices.

The transit shipper and the reusable carrier get confused in the other direction. A transit shipper is packaging: it exists to survive one network and it is discarded at the door. A reusable carrier is a branded article: it exists to be carried for years and it is judged in the second year. A shipper is a corrugated or foam cell inside a simple shell with a one-colour print; a reusable carrier is an EVA divider in a shell with a reinforced rim. The difference in the bill of materials is the finish and the divider, in that order, and confusing the two is what produces a bag that is too heavy to ship and too soft to keep.

There is a compliance consequence to the format, and it belongs in the same conversation rather than in a later one. A carrier made substantially of paperboard can raise a question at the border between a bag of heading 4202 and a packaging container of heading 4819, and those two lines are not rated alike in every market. That is a broker question rather than a supplier question, and it is a reason to settle the format before the tariff line is settled. The calendar is the other half of the same conversation: wine is bought for a date — a harvest, a holiday, a release — and a presentation bag that arrives after the date has no value at all. That is the working set out in why bulk takes 35-50 days, run backwards from the release.

Which Parameters Drift Between the Approved Sample and the Second Run?

The drift in this category is mostly physical, and it is mostly in the three parts that were specified by description rather than by number: the divider, the base and the handle. Each item below has come back on a second run, and each one can be turned into a single line on the tech pack with a material and a figure.

• The divider material and thickness. A foam substituted for an EVA, or a 2 mm sheet for a 3 mm one, holds the cells the same on day one and differently in month three.
• The divider height, which is the drift nobody measures. Thirty millimetres shorter is invisible in a photograph and it is the difference between separated bottles and touching ones.
• The base insert. A thinner board, or one cut 10 mm inside the footprint, turns a load path into a floating panel.
• The handle webbing width and the number of bar-tacks on the attachment. Neither is written down unless someone writes it down.
• The cell width. Four millimetres is a rounding error on one bottle and it is a rattle across six.
• The finish. A gloss lamination substituted for a matte one changes the colour of the print underneath it, and it is on the outside of the bag where everyone can see it.

The reasoning behind the tolerance sheet is in keeping a repeat order within tolerance; the short version for this category is that a bag cannot be "the same as the sample" unless the sample is in the room and the material is named. A sealed reference sample, a recorded material specification and a measured tolerance on the three load-bearing parts is the whole control, and it fits on one page of the pack.

And the pre-production sample here should be a filled one. Three or four bags from the bulk materials, packed with the actual bottles at the actual weight, then dropped in the orientations that will be tested. An empty sample that looks right has approved nothing, and it is the reason a sample approval and a delivery can be two different bags.

What Are You Really Buying When You Ask for an AQL 2.5 Inspection?

This is the phrase that gets quoted at buyers, and it is worth knowing what those four words commit to. ISO 2859-1 is a lot-acceptance system: it draws a sample from a lot, counts defects against a rule and returns accept or reject. It does not grade the bag, and it does not find the defect sitting in the 94% nobody opened. At General Inspection Level II with AQL 2.5, a 1,201 to 3,200 piece lot draws 125 units and the decision rule is 7 — acceptable at 7 or fewer major defects, rejected at 8.

Four things belong on the purchase order together: the standard, the inspection level, the AQL and the lot band. A quotation that says only "AQL 2.5" has stated one of the four and has therefore not committed to a sample size at all, which is the practical meaning of the phrase and the reason two suppliers can both say it and mean different things.

Read the far-right column before signing anything. Two thousand pieces get 6.3% of the units looked at; twenty thousand get 1.6%. That is what the standard is for and it is not a criticism of it — the job is accepting or rejecting lots. Where one failure is expensive on its own, General Level II is the wrong instrument. In this category the instrument you actually want is a different one: a filled drop test on a stated number of bags per lot, at the orientations in the table above. That is not sampling by attributes, it is a test, and it costs what a test costs.

The classification is where the report is won or lost, and it is the part inspection instructions leave blank. Critical is a bag that is unsafe or non-compliant: a base seam that releases a bottle at the drop height, or a material that fails a restricted-substances requirement. Major is a bag that gets returned: a cell 5 mm over width, a divider that is not full height, a handle attachment that fails a static pull. Minor is cosmetic: a print 2 mm out of register, a thread end that is not on a load path. AQL 2.5 governs the majors; criticals are zero-tolerance at every AQL; minors normally carry their own, tighter figure. Put all three in the instruction with an example from your bag, because an inspector left to classify by instinct will classify differently next time.

A competitor will offer AQL 1.5 and it is worth understanding what that buys. The accept number falls, so fewer defects are tolerated; 1.5 is not among the grades normally paired with ISO 2859-1 and ANSI Z1.4 in commercial practice; and the main effect is to reject more lots that were acceptable. It moves the risk between the parties rather than reducing what leaves the building. The arithmetic is set out in AQL 1.5 vs 2.5.

How Long Does a Wine Bag Programme Take From Tech Pack to Port?

MOQ is 500 pcs for one style in one print, and on a repeat it can be discussed down to 300. The constraint is the setup rather than a rule. Each style carries its own cutting pattern, its own divider die, its own print setup and its own assembly jig, and those are paid once and spread across the run. Splitting 500 into two prints of 250 produces two print setups and does not price like the single one, which is the surprise that arrives in the first enquiry more often than any other.

There is a second quantity conversation in this category and it is about the finish. A presentation bag with a foil crest and an emboss carries a block and a die, and those are worth their cost at 500 and not at 100. A transit shipper with a one-colour print carries almost no tooling at all, which is why a shipper programme and a presentation programme are quoted at different quantities. If the two are in the same order, quote them as two lines.

MOQ

500 pcs per style / print (300 on repeat)

Sampling

6-10 working days (12-15 for complex builds)

Bulk production

35-50 days from sample approval

Inspection

AQL 2.5, ISO 2859-1, General Inspection Level II

Payment

T/T 30/70

Incoterm

FOB Xiamen (CIF / DDP available)

Capacity reference

7 production lines, 149 machines, 200,000 pcs per month

Operating since

2014

From sample approval, bulk takes 35-50 days, and the spread is where the work is rather than padding. It holds three things: the wait on the material that was named rather than substituted, because a specified EVA sheet and a specified board are made to order and are not sitting on a shelf; the position in the cutting and assembly queue behind whatever was booked first; and the inspection, pack-out and packing window at the far end. We do not promise the fastest 20 days. We promise a delivery date you can calculate — and in this category the calculation runs backwards from the date the wine is released, which is the method in why bulk takes 35-50 days.

The sampling clock is spent somewhere buyers do not expect, and it is worth saying plainly: a wine bag cannot be sampled from a drawing. The sample is only as good as the bottle it was built around, and the bottle in the drawing is not the bottle in the case. Send two physical bottles of every shape you will ship before the 6-10 working days start, and send the tallest and the widest first, not the house shape. A sampling window spent fitting a bottle that was never in the range is a window spent twice.

And then the date. Subtract customs entry and inland delivery, subtract transit, subtract the 35-50 days, subtract the 6-10 working days of sampling, subtract your own approval time — and subtract the time it takes your glass supplier to confirm the shape, which in this category is usually the last item on the list and occasionally the longest.

Where Does the Money Go in a Custom Wine Bag Quotation?

The divider is the answer in this category, and it surprises buyers who expect the fabric. A six-cell divider is cut, slotted, assembled and fitted, and each of those is paid for in labour; on a six-bottle carrier the divider assembly is usually the largest single block in the bill of materials, ahead of the shell and ahead of the print. A quotation that prices the divider as a material is a quotation that has not read the operation.

• The divider assembly, and it is labour rather than material: the number of cells, the number of pieces in the assembly and the number of operations to fit it.
• The base insert and its fixing, which is two materials and three operations when it is built as a load path.
• The handle construction. A continuous loop under the base is one component and two attachments; two handles on the side panels are two components and four.
• The print, by method and by colour count. Each colour in a screen print is a screen and a pass; a foil is a block and a pass; an emboss is a die and a pass.
• The finish. A lamination protects the print and it goes onto the sheet before the bag is made, which is why it has to be specified before the material is ordered.
• The shell, by construction. A laminated shell removes a lining operation and adds a lamination one.
• Carton cube, and it is always available as a lever: a rigid six-bottle shipper ships erected, and erected volume is what the forwarder charges on.

Tooling and testing are the two costs that never appear on a unit-price line and always appear in the accounts. A cutting die for the divider, an embossing die, a foil block and a moulded handle are paid once, on the first invoice, and never again; a drop test and a colourfastness report behave the same way. Insist on both as separate lines. A quotation that has absorbed them into the unit price is a number that will not survive a reorder at a different quantity, and the reorder is exactly where that becomes visible.

Freight is the lever that is nearly always available and nearly always left alone, and on a rigid six-bottle shipper it is a large one. Rigid bags are charged on whichever is greater, the weight or the cube, and the cube wins by a wide margin, because the product is mostly air. A shipper designed to knock down or to nest, in a carton specified to the nested size instead of the erected one, will move the landed figure further than a point off the unit price. That is the purpose of the landed cost calculator: the unit price is one line of eight.

Which Five Design Mistakes Bring Wine Bags Back?

All five are present in the sample that gets approved, and that is what makes them expensive. The inspector works to the approved sample, so an inspection passes all five without any difficulty at all. The drawing review against the tech pack is the only control that catches them, and it has to happen before the 6-10 working days are spent rather than after they are gone.

A sixth that is not a design error but produces the same result: shipping without the drop test. A bag that has never been dropped has not been specified, it has been described, and the first time it meets a parcel network it will tell you which of the five it was.

What Does a Wine Bag Cost to Land, and Which Tariff Line Does It Enter On?

Wine bags are normally entered in heading 4202, and the subheading follows the outer surface. A textile outer surface, or an outer surface of plastic sheeting, usually falls in 4202.92; a bag with an outer surface that is neither of those falls in the residual subheading of the same heading. A carrier made substantially of paperboard raises a separate and genuine question, because it can be read as a packaging container of heading 4819 rather than as a bag of 4202, and the two are not rated alike in every market. Rates are those of the destination tariff and they move, so a rate in an email is an estimate — confirm it against the official tariff or through a licensed broker before the programme is priced.

The reason to get the line right is not only the rate. Origin preferences and duty-relief schemes attach to individual tariff lines, and an error discovered at the border is corrected afterwards as a post-entry amendment rather than settled in a conversation. The Harmonized System is maintained by the World Customs Organization and reissued on a five-year cycle, so the line that cleared last season is not automatically this season one.

Two things to raise with the broker before the goods move. A bag shipped holding wine is a very different entry from the empty bag, and it is the importer entry rather than ours; the bag and its contents may be treated as one article, and that is a broker question. And a shipper intended to be discarded may be read as packaging rather than as a bag, which changes the line and sometimes the duty. The HS code finder will settle the line, and the Incoterm decision tool will say whether FOB Xiamen allocates the risk the way you want on that lane. FOB Xiamen is how it is quoted and how most orders move; it is not right on every lane, and on some of them the buyer is paying a freight margin for the sake of not taking a call.

Eight lines then make the number that matters: unit price, haulage to the port, origin charges, freight on the greater of weight and cube, insurance, destination handling, duty at the confirmed rate, and final delivery. There is a calculator for it, and the two lines it was built for are the two nobody quotes — destination handling and duty. On a rigid six-bottle shipper, it is the freight line that surprises people.

Frequently Asked Questions

People Also Ask