Traditional Kintsugi with Urushi and Gold
The real one. Tree sap, powdered precious metal, and a repair that takes most of a season instead of most of an evening. It is slow, it is genuinely food safe at the end, and it will outlive everyone reading this page.
This is a home craft, not a studio one. Nothing on this page happens at The Clay Hole. We are teaching it because members keep asking, because the answers online are a mess of half-truths, and because a broken pot is the most fixable thing in the building.
- 4 to 12 weeks per piece
- Food safe when cured
- Skin hazard, handled properly
Read the list. Then watch somebody do it.
Every section here pairs a short written list with video of a working practitioner. That is on purpose, and it is not just belt and braces.
Video is unbeatable for showing you what a thing looks like. How thick the paste should be when it stops sagging off the spatula. How little gold powder is actually on the brush. You cannot get that from a paragraph, and anybody who tells you otherwise has never tried to describe the color of half-cured lacquer.
What video is terrible at is being checked. You cannot scan a video. You cannot glance at minute 14 with sticky gloves on to remember whether the sanding comes before or after the black coat. It runs at its own speed, it goes past, and the part you needed was thirty seconds you were not looking at. Written steps sit still. You can put them on a phone propped against a jar and look at line four eleven times.
So watch first for the feel of it, then work off the list. The lists on this page are all built to be readable at arm’s length on a phone screen, because that is where they will actually be used.
The videos are other people’s work
Every clip below is from a practitioner who teaches this for a living, most of them in Japan. We did not make any of it and we do not profit from any of it. If a channel helps you, go subscribe to them. The player and its cookies wait until you tap, but the still frame on each card is one request to Google that happens while you read.
Before you scroll
- Three hours of work, spread over eight weeks. That is the honest shape of a traditional repair, and almost all of it is waiting. Start a second pot.
- A crack is a different job from a break. If the pot is still in one piece there is nothing to glue. You feed thin raw lacquer into the line over several sessions and let capillary action pull it down. Then rejoin the main road at the smoothing stage.
- Buy keshifun, the fine matte powder, and buy very little. A tenth of a gram covers a surprising length of seam. A repair finished in plain lacquer with no metal at all is also a real, traditional result, so the gold is never the thing stopping you.
Urushi is poison ivy’s first cousin, and it does not care how careful you feel
Raw urushi is sap from Toxicodendron vernicifluum. It contains urushiol, the same compound that makes poison ivy poison ivy. This is the single most important thing on this page and it is the thing most kits mention in six point type on the back of the box.
- Sensitization builds up. A first reaction does not predict a second. A year without trouble is a year without trouble, not immunity.
- Dry is not cured. It can feel dry and still carry live urushiol. Sanding a part-cured fill throws it into the air.
- It travels. Doorknobs, phone screens, taps, sleeves. Most rashes on a forearm or a face did not come from the brush.
-
The part people get wrong
Sensitization builds up
A first reaction is not a reliable predictor of a second one. People handle urushi for months with nothing, then react badly. Going a year without trouble is not immunity, it is a year without trouble.
-
The part people miss
Curing is not the same as dry
Urushi can feel dry to the touch and still carry uncured urushiol. Sanding a part-cured fill throws it into the air. Do not touch a piece in progress with bare hands, ever, even weeks in.
-
The part that surprises
It travels
Doorknobs, phone screens, the tap you turned on to wash up, the sleeve you pushed your hair back with. Most rashes on a forearm or a face were not put there by the brush.
The setup, every single session
- Nitrile gloves, not latex Urushiol passes through natural rubber. Nitrile is the standard and it is what the kits ship.
- Long sleeves and long trousers Clothes you have written off. Wash them alone, not with anything else.
- Cover the table Newspaper or a plastic sheet you throw away at the end. Not the good tablecloth and not a surface anybody eats at.
- Tie hair back and keep your phone out of reach The two most common ways it gets onto a face.
- Vegetable oil within arm’s reach Oil first, then soap and water. Water alone spreads it around.
- Ventilate A window open, not a sealed room.
- Never in a kitchen Not where food is prepped, not where pets walk, not where small children can reach the cardboard box.
When to not do this at all
If you react to poison ivy, poison oak, mango skin or raw cashew, you are already cross-reactive to urushiol and the odds are against you. If you have had a serious contact dermatitis reaction to anything, or you are immunocompromised, take the synthetic path lower down this page instead. It is a good craft in its own right and nobody is grading you.
We are a pottery studio, not a medical anything. This is craft safety practice, not medical advice. A severe reaction, anything near the eyes, or any breathing trouble is a doctor, immediately, and tell them it was lacquer from the sumac family.
Two honest paths, and neither one is the fake one
One question decides which path you are on
Almost everything sold as a kintsugi kit is one of these two things. They are not the same craft wearing different hats. They produce different objects, on different timescales, with different rules about what you are allowed to eat off afterward.
Every comparison of traditional and synthetic kintsugi is really one question wearing a lot of clothes: are you going to eat or drink off this pot?
-
Yes, it goes back in the cupboard
Traditional urushi. Two months, a real skin hazard while you work, and a repair that is genuinely food safe and will outlive you. There is no shortcut version of this that is also safe to eat off.
-
No, it lives on a shelf
Synthetic. Two-part epoxy and mica powder, done in an evening, no lacquer rash. Treat it as decorative, keep it off drinking rims, and enjoy it for exactly what it is.
| Traditional urushi | Synthetic | |
|---|---|---|
| What is in it | Traditional urushiRefined tree sap, flour, tonoko stone powder, wood powder, pure metal powder | SyntheticTwo-part epoxy or a cashew-based lacquer, plus mica or brass powder |
| Time to finish | Traditional urushiFour weeks at an absolute minimum, three months is normal | SyntheticAn evening, plus a day or two of curing |
| Skin risk | Traditional urushiReal and cumulative until the piece is fully cured | SyntheticOrdinary epoxy precautions, gloves and airflow |
| Eat off it | Traditional urushiYes, once fully cured. This is the whole reason the method survived | SyntheticTreat as decorative unless the maker certifies food contact, and even then keep it off the lip |
| Dishwasher and microwave | Traditional urushiNeither. Hand wash, no soaking | SyntheticNeither. Metal powders arc, epoxy softens |
| How it ages | Traditional urushiHardens further for years. Museum pieces are centuries old | SyntheticEpoxy yellows in daylight and gets brittle. Brass darkens |
| Cost of materials | Traditional urushiRoughly the price of a nice dinner out for a starter kit, more for real gold | SyntheticRoughly the price of lunch |
| What it is actually good for | Traditional urushiThe bowl that matters. Things you will use for thirty years | SyntheticLearning the shapes, vases, display pieces, and pots you would otherwise bin |
Neither column is cheating. What is cheating is selling a synthetic repair as food safe, or letting somebody eat off an epoxy seam because it looks like gold. The rest of this page teaches the left column first and then the right column properly, because most people should probably start on the right.
One rule that applies to both
A mended pot never goes back in a kiln. Our glaze firings run to cone 5½, and urushi, gold, epoxy and mica all burn away long before anything gets near that. If a piece still needs firing, fix it in the clay stage instead and read how to score and slip a join.
Four kinds of damage, four different jobs
Before you buy a thing, look at the break and name it. The materials are the same across all four. The order of operations is not, and picking the wrong sequence is the most common way a first repair goes sideways.
- Broken, chipped or missing a piece? All three follow the six stages below. You are gluing, filling a volume, and then smoothing it.
- Cracked but still in one piece? Different route. Nothing to glue, so you wick thin lacquer into the line instead, then join the stages at the smoothing step.
-
Type one
Clean break
Two or more pieces, edges that still fit. The easiest and the most satisfying. Glue, fill the hairline, coat, gild.
-
Type two
Crack, still in one piece
A line you can see and maybe feel, but the pot has not come apart. Different route: you are getting lacquer to wick into the crack, not gluing anything.
-
Type three
Chip
A bite out of a rim or a foot, no fragment to put back. You are rebuilding a small volume and then shaping it to match the curve.
-
Type four
Missing piece
A hole where a shard used to be and no shard. Same idea as a chip but bigger, more layers, and much more patience.
Watch each one done properly
POJ Studio in Kyoto made a short film for each damage type. Find yours, watch that one, and you will already know what the finished job should look like before you open a single tube.
Choose an easy first pot on purpose
Stoneware or porcelain, two or three big pieces, edges not chewed. Do not learn this on the mug your grandmother made.
Choose an easy first pot on purpose
A stoneware or porcelain piece broken into two or three big pieces, with a glaze that is not crazed, and edges that have not been chewed. Do not learn this on the mug your grandmother made. Learn it on something from a thrift shop, break it deliberately in a towel if you have to, and save the one that matters for repair number three.
What you actually need, and what the words mean
The vocabulary is the barrier, not the technique. There are really only three lacquer mixtures in this whole craft, and each one is raw urushi plus one other thing. Once that lands, the rest of it is patience.
All of this is bring-your-own. The studio does not stock urushi, gold powder or any part of a kintsugi kit, and we have no plans to. Buy a kit from a Japanese maker rather than assembling it yourself the first time. The proportions in a kit are already right and getting them wrong is how a repair fails in month two.
Buy a starter kit from a Japanese maker rather than assembling one yourself. The proportions are already right, and getting them wrong is how a repair quietly fails in month two. Add plain wheat flour, ethanol, vegetable oil, wet and dry sandpaper, masking tape and a box of nitrile gloves from your own cupboard.
The three mixtures, decoded
| Mixture | Raw urushi plus | What it is for |
|---|---|---|
| Mugi urushi | Raw urushi plusWheat flour and a little water | What it is forThe glue. This is what holds broken pieces together |
| Kokuso urushi | Raw urushi plusFine wood powder, usually on top of mugi urushi | What it is forThe bulk filler. Deep chips and missing pieces |
| Sabi urushi | Raw urushi plusTonoko, a fine stone powder, slaked with water | What it is forThe smooth filler. Hairlines, pinholes, and the final surface before coating |
On top of those three you use two coating lacquers. Black urushi, the middle coat that gets sanded flat and gives the gold something even to sit on. And bengara urushi, a red-brown lacquer, which is the last wet layer and the one the metal powder actually lands in.
Plus two coating lacquers: black urushi for the middle coats, and bengara, a red-brown, for the last wet layer that the metal lands in.
The kit list
- Raw urushi The base of everything. Refrigerate it, use it within about a year.
- Black urushi and bengara urushi The two coating lacquers.
- Tonoko and wood powder The fillers you mix into sabi and kokuso.
- Metal powder Pure gold, silver, platinum or brass. See the gold section below.
- Wheat flour Plain flour from your own cupboard. It has to contain gluten, so most gluten free flours will not work.
- Spatulas Cypress or plastic, for mixing and applying paste.
- A menso brush A very fine lacquer brush for the final line.
- An ashirai brush or a soft dusting brush For laying down metal powder.
- A silk ball Wadded silk. This is what you burnish the gold with.
- Wet and dry sandpaper Roughly 400 up to 1500 grit.
- A small file or craft knife For carving back hardened squeeze-out.
- Masking tape Not optional. This is what stops gravity undoing your glue-up.
- Ethanol For degreasing the break and cleaning tools.
- Vegetable oil For getting urushi off skin and off brushes.
- An acrylic sheet or glass tile Your mixing surface.
- Nitrile gloves A box of them. You will change them often.
- A cardboard box This becomes the curing box. Next section.
Two words you will meet in catalogs. Keshifun is fine matte metal powder, the beginner default, and it needs no polishing. Marufun is a coarser round powder that gives a deep mirror finish and has to be burnished and consolidated. Start on keshifun.
Urushi does not dry. It cures, and it needs damp air to do it
This is the fact that trips up every beginner in a dry climate, which is to say every beginner in Utah. Urushi does not harden by losing solvent like paint. An enzyme in the sap polymerizes it, and that enzyme needs moisture and warmth. Leave a lacquered pot out on an open shelf in a Salt Lake winter and it can sit there sticky for weeks and then never properly harden at all.
Urushi hardens by absorbing moisture, not by drying out. On an open shelf in a Utah winter it can stay sticky for weeks and then never properly harden at all.
The fix is a muro, also called an urushiburo: a box that holds humid air. It is a cardboard box with a wet cloth in it. That is genuinely the whole invention.
- A cardboard box or a lidded tub, big enough that the pot does not touch the sides.
- A damp cloth in one corner. Damp, not dripping, and never touching the work.
- Somewhere warm and dark, around 70 to 80 percent humidity. Re-damp the cloth every couple of days and otherwise leave it shut.
Build one in ten minutes
-
Find a box that closes
Cardboard, a plastic storage tub with a lid, or a cooler. Big enough that the piece is not touching the sides. Cardboard is traditional because it breathes slightly and will not drip condensation onto the work.
-
Line the bottom
Newspaper or a towel, so the piece is not sitting on bare cardboard.
-
Add water, but not to the piece
A damp cloth in one corner, or a shallow dish of water. Damp, not dripping. The pot must never touch standing water.
-
Aim for warm and humid
Around 68 to 77 degrees Fahrenheit, and 70 to 80 percent relative humidity. A cheap hygrometer from a hardware store takes the guesswork out and costs less than the gold powder will.
-
Close it and leave it alone
Somewhere out of daylight, away from where anybody eats, and out of reach of children and pets. Re-dampen the cloth every couple of days. Opening it to look does not help anything.
Thin layers, always
Because urushi cures from the surface inward with help from moisture, a thick layer skins over and stays soft underneath for a very long time. Every single stage on this page works better thin and repeated than thick and hopeful. If you take one habit away from this section, take that one.
Six stages, and five of them are mostly waiting
Six stages, and one video to work from
Here is the whole method in one place. Each stage has a short list you can work from and video of somebody doing that exact stage. Work down the page in order and do not skip a cure, however tempting it looks.
Do these in order and do not skip a cure, however tempting it looks. Every wait below is in the damp box.
-
Clean, dry-fit, tape
Wash the break faces, degrease with ethanol, dry the pot for a day. Assemble it dry with tape first to work out the order the pieces go in, and cut your tape strips before you mix anything. Grease left in the break is why repairs let go months later.
-
Glue with mugi urushi, then wait two weeks
Smooth flour paste, then an equal volume of raw urushi folded in until the color is uniform. Thin film on both faces, press, tape under tension across every join, wipe the squeeze-out with an oiled cotton bud while it is wet. Box for one to two weeks.
-
Fill any gaps, then smooth them
Only if there is a chip or a missing piece. Kokuso putty packed slightly proud for volume, in layers rather than one thick plug. Then two or three rounds of sabi urushi, wet sanded back at 600 grit between each, until a fingernail does not catch on the seam.
-
Two or three black urushi coats
A lightly loaded fine brush, one smooth pass along the line, cure, then a very light wet sand at 1000 to 1500 grit. Repeat. This stage is what separates a repair that looks professional from one that looks homemade.
-
Bengara, wait for tacky, dust the metal
Paint a thin red-brown line, then wait somewhere between fifteen minutes and two hours until it is tacky rather than wet. Paint the same line on a scrap shard at the same time and use it as your timer. Tap the powder on with a soft brush, do not paint it on.
-
Cure a week, burnish, then wait a month
Brush the loose excess away, box it for a week, then burnish gently with a silk ball. Keshifun needs almost nothing. Wait a further month before you eat off it.
Keep this video open on your phone while you work. It is the whole method, food safe, in one sitting.
Stage 1. Clean and prepare the break
Nothing sticks to grease. This stage feels skippable and is the single most common cause of a repair that lets go three months later.
- Wash everything Warm water, dish soap, a soft brush into the break faces. Old pots carry decades of oil.
- Dry it completely A day in a warm room. Trapped water in a porous body will fight the lacquer.
- Wipe the break faces with ethanol Both sides of every join.
- Dry-fit before you mix anything Assemble the whole pot with masking tape and no glue. Work out the order the pieces go in. A three-piece break has a right order and a wrong one that locks you out.
- Number the pieces Small pencil marks on the inside. You will be glad an hour later.
- Cut your tape now Strips ready on the table edge. You do not want to be tearing tape with lacquered gloves on.
Stage 2. Glue with mugi urushi
Mugi urushi is flour, water and raw urushi. It is the glue, and it is strong enough that a properly done join will break somewhere else next time.
- Make a flour paste first A little plain flour, water added drop by drop, worked with a spatula until it is smooth and about the thickness of toothpaste. No lumps.
- Add raw urushi in roughly equal volume Fold it in on the acrylic sheet, scraping and folding, until the color and texture are completely uniform. Streaks mean it is not mixed.
- Apply a thin film to both faces Thin. A visible bead of glue is too much and will squeeze out into a ridge you have to carve off later.
- Press and hold Line the pieces up by feel, not by eye. Run a fingertip across the join through a glove. Level beats visually straight.
- Tape it under tension Strips across every join, pulling the pieces together. Gravity will open a join overnight if you let it.
- Wipe squeeze-out now A cotton bud with a little oil, while it is still wet. Ten seconds now saves twenty minutes of carving later.
- Into the box for one to two weeks Yes, weeks. Mugi urushi is slow and rushing it here undoes everything after it.
Stage 3. Fill the volume with kokuso
Only if you have a chip or a missing piece. A clean break with no losses skips straight to stage 4.
- Carve back the hardened squeeze-out first A craft knife held almost flat, working along the join rather than across it. Gloves on: this dust is the classic way people get a rash weeks in.
- Mix kokuso Raw urushi, a little mugi urushi, and fine wood powder until it is a stiff putty that holds a shape.
- Pack it in slightly proud Press it into the gap and leave it a touch above the surrounding surface. It shrinks a little as it cures.
- Shape it while it is soft A wet spatula gets you most of the way to the pot’s real curve. Every minute here is five minutes of sanding saved.
- Build big gaps in layers Never one thick plug. Two or three thinner passes with a cure between each.
- Cure three to seven days per layer Back in the box.
Stage 4. Smooth it with sabi urushi
Sabi is the fine filler. It closes hairlines, pinholes and the texture kokuso leaves behind, and it is what turns a structurally sound repair into a smooth one. Everybody underestimates how many rounds of this it takes.
- Slake the tonoko Tonoko powder with just enough water to make a stiff clay-like lump. Let it sit a few minutes.
- Add raw urushi Work it in until it is uniform and about the consistency of soft butter.
- Skim it along the seam A spatula almost flat, pressing it into the line and scraping the excess off. You want it in the gap, not spread across the glaze.
- Cure two to four days Box again.
- Wet sand back 600 grit or so, with water, working only the fill and staying off the surrounding glaze. Gloves, and rinse the slurry away rather than letting it dry into dust.
- Repeat until it is genuinely flat Two or three rounds is normal. Run a fingernail across: if it catches, it is not flat.
Stage 5. The black coat
Black urushi is the middle coat. It seals the fill and gives the gold a hard, even bed to sit on. Gold laid straight onto sabi looks dull and patchy, because sabi is porous and drinks the lacquer under it.
- Load the menso brush lightly Wipe most of it off. A loaded brush floods the line.
- One smooth pass along the seam Follow the line, do not scrub back and forth. Slightly wider than the fill, so the seam reads as a deliberate line.
- Cure one to three days Box.
- Wet sand at 1000 to 1500 grit Very light. You are flattening, not removing.
- Coat and sand again, two or three times The line should end up glassy. This is the stage that separates a repair that looks professional from one that looks homemade.
- Clean the brush in oil immediately A dried menso brush is a dead menso brush.
Stage 6. Bengara, then the metal
The last wet layer is bengara urushi, a red-brown lacquer. The metal powder does not get glued on afterward. It gets dusted into the bengara while it is still tacky, and the lacquer grabs it. Timing is the whole skill here.
- Paint a thin bengara line Same technique as the black coat, one clean pass.
- Wait for tacky, not wet Put it in the box and check it. Somewhere between about 15 minutes and two hours depending on humidity and the lacquer. Too wet and the powder sinks and goes dull. Too dry and it will not hold at all.
- Test on a scrap first Paint a spare line on a broken shard at the same time and use it as your timer. This one habit saves more first attempts than anything else on this page.
- Dust the powder on A soft brush or silk ball, tapped over the line like tapping ash off a cigarette. Do not paint it on.
- Cover it fully, then stop Once the line is covered, adding more does nothing but waste gold.
- Back in the box for 30 minutes to two hours Then brush the loose excess away gently with the tip of a dry brush.
- Cure a week, then burnish A silk ball, light strokes. Keshifun needs almost nothing. Marufun needs real burnishing over several sessions.
The same six stages, in one sitting, from a second teacher
Two teachers is better than one. Chimahaga is a Tokyo repairer whose full walkthrough covers a broken pot end to end with the same food safe method, and hearing it explained twice by different people is what makes it stick.
A crack is not a break, and gluing it will not work
If the pot is still in one piece but has a line running through it, you cannot get glue between two faces because there are no two faces to get between. What you do instead is persuade thin raw urushi to wick down into the crack by capillary action and harden in there.
- Clean and dry thoroughly A crack holds water for longer than you would believe. Give it several days.
- Thin the raw urushi slightly A drop or two of solvent so it will actually run rather than sit on top.
- Paint it along the line and wait Watch it sink in. When it stops sinking, add more.
- Repeat over several sessions Three or four rounds until the crack stops drinking.
- Wipe the surface, cure, then join the main road From here it is sabi, black coat, bengara, metal, exactly as above.
A crack that leaks can be stopped
This method genuinely seals a weeping crack, which is the argument for doing it at all rather than just living with the line. It will not, however, save a pot that is crazed all over. Crazing is a glaze fault across the whole surface and no amount of lacquer fixes it.
Gold, and the four decisions nobody warns you about
-
Keshifun or marufun
Keshifun is fine and matte, sits down evenly, needs no polishing, and is right for a first repair. Marufun is coarse and round, has to be burnished and consolidated over several sessions, and gives the deep liquid mirror you see on museum pieces. Marufun is a skill; keshifun is a technique.
-
Which metal
Pure gold never tarnishes and is inert. Silver is cheaper and beautiful and will darken, which some people want. Platinum gives silver’s color without the tarnish. Brass is the budget option, looks close to gold at first, oxidizes over years, and should be kept away from food surfaces.
-
How much you need
Far less than feels right. A tenth of a gram covers a surprising length of seam. Kits ship that quantity for a reason. Most beginners waste half their gold on the first pass by loading the brush and painting instead of tapping and dusting.
-
Whether to use metal at all
A repair finished in plain black or red lacquer with no metal is a real, traditional, complete result. It has its own name and its own following. If the gold is what is stopping you starting, skip the gold and start.
The line you draw is a design decision
Nobody is checking your work against the original crack. Practitioners routinely widen a line where it wants to be seen, taper it where it should fade, and let a seam branch. The Japanese word for the resulting pattern translates roughly as scenery. You are not concealing a fault, you are drawing the pot’s history on the outside of it, and that is the entire point of the craft.
The synthetic path: one evening, real rules
This is the version most people should start with, and it deserves better than being treated as the consolation prize. The shapes you learn here are the same shapes. The judgment about line width, the patience with sanding, the decision about where a seam should be seen: all of it transfers. What changes is the chemistry and the rules about eating.
Two-part epoxy replaces mugi urushi. Epoxy putty replaces kokuso and sabi. Mica or brass powder replaces gold, either mixed into the resin or dusted onto a final coat.
The method
-
Clean, degrease, dry-fit, tape
Exactly as on the traditional path and just as load-bearing. Cut your tape before you mix.
-
Glue clear, not gold
Use a thirty minute epoxy, not a five minute one, so you have room to align. Thin film, press, tape under tension, and leave it for the full cure time on the packet rather than the touch-dry time. Tinting the glue itself is quicker and reads plastic.
-
Fill gaps with epoxy putty
Press it slightly proud, shape it wet with a damp tool, let it harden fully.
-
Sand until you cannot feel the seam
Blade the hard squeeze-out back, then work up through the grits with water. This is where synthetic repairs are actually won. Nearly every disappointing epoxy kintsugi on the internet is a sanding problem, not a gold problem.
-
Paint the gold line, then cure properly
Mica powder in a little clear resin, finest brush you own, one confident thin pass. Then days out of direct sunlight before anything touches it.
-
Clean, dry, dry-fit, tape
Identical to the traditional path and just as load-bearing. Wash, degrease with alcohol, dry properly, work out the assembly order, cut your tape before you mix anything.
-
Pick an epoxy with a working time you can live with
Five minute epoxy is unforgiving on anything more than two pieces. Thirty minute epoxy gives you room to align. Mix small amounts on a scrap of card, exactly to the ratio.
-
Decide: tint the glue, or gild on top
Mixing mica into the adhesive gives a gold line straight out of the join, which is quicker and reads slightly plastic. Gluing clear and then painting a gold line over the seam afterwards is more work and looks considerably better. We would do the second.
-
Thin film, press, tape, wait
Same discipline as lacquer. A visible bead is too much. Tape under tension across every join and leave it for the full cure time on the packet, not the touch-dry time.
-
Fill any gaps with epoxy putty
Press it slightly proud, shape it wet with a damp tool, and let it harden fully.
-
Sand, and then sand again
This is where synthetic repairs are actually won. Scrape the hard squeeze-out back with a blade, then work up through the grits with water until the seam is dead flat and you cannot feel it with a fingernail. Nearly every disappointing epoxy repair on the internet is a sanding problem, not a gold problem.
-
Lay the gold line
Mica powder mixed into a little clear resin or a suitable lacquer, applied with the finest brush you own, in one confident pass. Thin. Build a second pass if you must.
-
Cure fully before anything touches it
Days, not hours, and out of direct sunlight. Then decide honestly what the piece is for.
Say the unflattering part out loud
An epoxy repair is not food safe by default, and no amount of the word artisan on a packet changes that. Unless the manufacturer specifically certifies their product for food contact, treat a synthetic repair as decorative. Keep it off drinking rims, keep it out of the dishwasher and the microwave, and be honest about it if you give the piece away. Vases, planters, display bowls and practice pots are where this method shines.
The studio sells a five minute epoxy at the front counter. It is a general repair adhesive, not a certified food contact product, and the five minute working time makes it better suited to a two-piece break than a complicated one.
What a real traditional repair looks like on a calendar
Total hands-on time for a clean two-piece break is maybe three hours. Spread across roughly eight weeks. Almost all of this craft is putting a box somewhere and going away.
| When | What you do | At the bench |
|---|---|---|
| Day 1 | What you doWash, degrease, dry-fit, tape, mix mugi urushi, glue up | At the benchAbout 60 minutes |
| Days 2 to 14 | What you doIn the box. Re-damp the cloth every couple of days | At the benchZero |
| Day 15 | What you doCarve back squeeze-out, kokuso any gaps | At the benchAbout 30 minutes |
| Days 16 to 21 | What you doCure | At the benchZero |
| Days 22 to 35 | What you doTwo or three rounds of sabi urushi and wet sanding | At the benchAbout 20 minutes each |
| Days 36 to 48 | What you doTwo or three black urushi coats, sanded between | At the benchAbout 15 minutes each |
| Day 49 | What you doBengara, wait for tacky, dust the metal | At the benchAbout 45 minutes plus waiting |
| Days 50 to 56 | What you doCure, then burnish with silk | At the benchAbout 15 minutes |
| Weeks 9 onward | What you doHardening continues. Wait a month before eating off it | At the benchZero |
Start a second pot while the first one is in the box. That is how everybody who does this regularly works, and it is why they never seem impatient.
Nine ways this goes badly, and what each one is telling you
Three ways a first attempt goes badly
-
One
Nothing hardened
Too dry, too cold, or old lacquer. Check the box with a hygrometer before you blame the urushi. Sap more than about a year old may simply be finished.
-
Two
Skinned on top, soft underneath
You went thick. Scrape it back and rebuild in thin coats. There is no shortcut around this one and waiting longer will not save it.
-
Three
The join opened while curing
Not enough tape, or tape not under tension. Gravity has all night and you do not.
-
Four
The seam is a wide ugly ridge
Too much glue and squeeze-out left to harden. Wipe it wet, next time, with an oiled cotton bud.
-
Five
The gold is patchy and dull
The bengara was too wet when you dusted, so the powder sank. Or the black coat was skipped and the porous sabi drank the lacquer. Test on a scrap next time.
-
Six
The gold brushed straight off
The bengara was too dry. There was nothing tacky left to hold it.
-
Seven
A rash, three weeks in
Almost always sanding dust from a part-cured fill, or touching the piece bare-handed because it felt dry. It was not dry.
-
Eight
It let go months later
Grease in the break at the start. Stage 1 is not a warm-up, it is the foundation.
-
Nine
It looks like a repair, not like kintsugi
The line is timid and uneven. Confidence in one pass beats care in six. Practise the line on a shard until your hand stops negotiating.
Most of it is recoverable
Uncured lacquer scrapes off. Sabi and kokuso sand back. A bad gold line can be sanded flat, recoated and re-gilded. The only genuinely unrecoverable mistakes on this page are getting it on your skin and putting a repaired pot in a kiln.
Living with a mended pot
- Wait a month after the last coat before you eat or drink from it. Curing continues long after it stops feeling wet.
- Hand wash only, warm water, mild soap, soft cloth. No abrasive side of the sponge.
- No soaking. A repaired crack in a base should never sit in a sink full of water.
- No dishwasher and no microwave, on either path. Heat, detergent and metal powder are all bad news.
- Keep it out of long direct sunlight. Urushi is genuinely damaged by ultraviolet over time and epoxy yellows.
- Do not use it for hot oil or anything above normal food temperatures.
- If a gold line wears thin in ten years, it can be recoated. That is normal and it is part of the deal.
Traditional urushi really is food safe
Fully cured urushi has been used on Japanese tableware for thousands of years and is inert once polymerized. That is not marketing, it is the entire reason this method beat every alternative for centuries. The caveats above are about longevity and appearance, not about toxicity. The same sentence cannot be written about an uncertified epoxy.
Two things that stop people before they start. This is a home project, not a studio one, on both paths. And if the pot has already been super glued, the traditional path is closed on that piece, because urushi will not bond reliably to cured cyanoacrylate. Pick a different pot.
The honest answers
Can I do this at the studio?
No. This is a home project on both paths. We are not set up for lacquer work, we do not stock any of the materials, and a shared studio is exactly the wrong place for a substance that sensitizes people over time. Read it here, do it at your own bench.
Will you repair a piece for me?
We do not offer a repair service. If you want a professional traditional repair on something precious, look for a practitioner who works in urushi and expect it to take months and to cost more than the pot did. That is not a markup, it is what eight weeks of somebody’s attention costs.
Is the gold real?
In a traditional kit, usually yes, and usually about a tenth of a gram. That is a tiny amount of metal doing a very visible job. Brass and mica versions exist and are honest choices as long as nobody claims they are gold.
Can I speed the curing up with a hair dryer?
No, and it will actively hurt. Heat without humidity is the opposite of what urushi needs. If you want it faster, the answer is thinner coats, not warmer air.
What if I already glued it with super glue?
Then the traditional path is closed on that piece, because urushi will not bond reliably to cured cyanoacrylate. You can either accept it and do a synthetic gold line over the existing seam, or try to release the join with acetone and start over, which often damages the edges. Learn the lesson and pick a different pot.
Does it work on glass?
Not with ordinary urushi, which does not grip smooth glass or a very glassy glaze well. There are specific glass lacquers sold for this and they take even longer to cure. Stoneware and porcelain are where beginners should stay.
How much does it cost to start?
A traditional starter kit from a Japanese maker is roughly the price of a nice dinner out, and it will do several small repairs. The synthetic route is roughly the price of lunch. Neither one is the expensive part. The expensive part is the two months.
Can I put a kintsugi piece back in the kiln?
Never. Our glaze firings run to cone 5½ and every material on this page, gold included, burns off or vaporizes long before then. Once a pot is mended it is finished being fired, permanently.
Is it true that kintsugi is about embracing imperfection?
The philosophy attached to it is real and it is also considerably more recent and more Western than the internet suggests. What is unarguable is the practical bit: it was a way of keeping expensive tea ware in service, and it chose to make the repair visible instead of hiding it. You are allowed to enjoy the meaning. You are also allowed to just want your bowl back.
My pot is bisque, not glazed. Does that change anything?
Porous bisque drinks lacquer, so expect more coats and slower progress, and expect the surrounding surface to stain where lacquer wicks out of the seam. It is doable. It is not a good first project.
Where to go from here
-
Clay Hole University
The whole curriculum this lesson belongs to, from wedging a lump of clay to firing it.
Browse the lessons -
Fix pottery after a firing
For damage that happened in the kiln rather than on the floor. Fused feet, drips and sharp edges.
Read the guide -
Score and slip joins
The repair that happens while the clay is still soft, which is the one that never needs gold.
Read the guide -
What a kiln firing actually does
Why a mended pot can never go back in, explained from the chemistry up.
Read the guide -
The many myths of pottery
Including a few about kintsugi that will not die.
Read the guide -
Drying pottery properly
Most cracks are made weeks before the kiln. This is how to stop making them.
Read the guide
The pot was already broken. Everything after that is a choice.
There is no version of this where you make the break not have happened. What you get to decide is whether the thing goes in the bin, or goes back on the shelf with a line of gold running through it that says somebody thought it was worth two months.
Start on something you do not love. Break a thrift shop bowl on purpose if you have to. And if you want the long version of any of this, with every stage filmed and the whole failure list, flip the switch at the top of the page.
Start on something you do not love. Break a thrift shop bowl on purpose if you have to. Do the synthetic version first and learn where your hand shakes. Then, when you are ready, go slowly at the one that matters.
Which version would you like to print?
Videos come out as a labelled box with the title and channel to search for, since paper does not play them.



