Blistered Paint

5 Causes And 1 Myth

And a video demonstration

57 new subscribers this week – it’s great so many people are willing to work hard improving their glass painting techniques.

Make no mistake: working with us is often tough, but you’ll get your own reward.

And I don’t just mean the free video at the end of this post.

I mean: your own achievements.

Today – blistered paint – you know, when glass paint bubbles and cracks in the heat of the kiln: 5 causes and 1 myth.

But first, here’s why you must know how to do it properly …

Thick, dark paint – its many uses

For the stained glass painter, its distinctive use is to “frame” a painted image – to “block in” that image and thus intensify the effect of transmitted light:

Thick, dark paint intensifies the effect of transmitted light

Thick, dark paint intensifies the effect of transmitted light

As you saw last week, the right-hand gargoyle is far more effective than the gargoyle on the left. The thick, dark paint makes everything more dramatic.

But thick, dark paint is often also used within an image (not just around it), as in the gargoyle’s eyes and mouth, and also in this heraldic crest … of which you’ll see a useful video in a while:

The heraldic crest of Hampton Hall

Thick, dark paint is also very useful within an image (not just around it)

Now flooding is a lovely, simple technique, so there are plenty of reasons for you to get it right.

Also, many ways it can go wrong.

Let’s get going.

The myth

Remember the problem we’re examining is why thick dark glass paint sometimes cracks and blisters in the kiln.

And since there are so many new arrivals this week, I’ll spend a moment debunking the myth which your long-standing colleagues know all about.

  • The myth is: you cannot paint on top of unfired glass paint because if you do, the paint will always blister.

You know, I’ve got at least 10 books which say this, and it’s just not true.

Newcomers – check out Part 1 of Glass Painting Techniques & Secrets from an English Stained Glass Studio (free chapter here), and you’ll see why this view is wrong: one of the first things you learn with us is how (when you want) you can do several layers of tracing and shading in a just one firing.

And that’s right: no blistering.

See, the problem isn’t caused by painting several layers. It’s caused by other things.

First the big causes. Second, the video demonstration. Third, the firing schedule.

So let’s now make a start.

Cause #1 – wrong paint

David and I have tested many different brands of glass paint. Our clear preference is for Reusche.

A good all-purpose mix is made from tracing black (DE401) with umber brown (DE402) in rough proportion 4:1 – it’s wonderful for all consistencies of glass paint from undercoats right through to silhouettes.

And yes, even we‘ve had blistering problems with different paints than Reusche’s.

Note for newcomers:

  • We’re independent.
  • We’re not paid sponsership money or advertising fees.

We tell you things because they work, that’s all. And Reusche paint is excellent.

Cause #2 – too much gum Arabic

With too much gum Arabic, your paint is almost bound to crack: it will just bubble and blister in the heat of the kiln.

So what’s the “right amount” of gum?

There isn’t a scientific answer here. Conditions vary. But here’s a useful and all-purpose test:

  1. Prepare a batch of glass paint.
  2. From this batch, prepare a small quantity of undercoating paint.
  3. On the light box, paint a swatch of undercoat, blend, and allow to dry.
  4. Repeat, so you have a double-layer.
  5. Make sure your hands are clean and dry. Gently rub the undercoat with the soft part below your thumb.

If it’s very difficult to bruise this double-layer of undercoat, this tells you there is too much gum Arabic in the paint. If you use this paint for flooding / silhouetting, it is likely it will blister in the kiln.

On the other hand, if the undercoat is easily and immediately rubbed away, this tells you there might be too little gum Arabic in the mix. You may wish to add more.

If the undercoat comes away in a measured and controllable way, that’s when we’d be happy with it.

Cause #3 – badly mixed reservoir of paint

Now what’s the “reservoir”, you ask?


You start with a lump of paint. (Not a teaspoonful – that’d be wasteful. See Part 1 section 1 for why.)

With your palette knife, you cut away some slices, add some water to get the right consistency for flooding, then mix it all together.

This is your “reservoir” – your pool of working paint.

Now the thing about glass paint, gum Arabic and water is … they’re not best friends! The water evaporates. The gum splits off from the water. And the silica and oxides in the glass paint just sink to the bottom.

That’s why, each time you load your brush, you must re-mix the paint.

If you don’t, the gum in particular will not be evenly distributed throughout the paint.

And that means some parts will have a higher density of gum than others.

And this is a very common cause of blistering.

The only way to prevent it is: re-mix your paint each time you load your brush. As you’ll see in the video, it only takes a few twirls.

Cause #4 – bad painting technique

Images are better than words here, so I’ll keep it brief because you’ll see it in the video.

Here’s what happens.

The moment the paint leaves your brush, it starts to dry and set.

This means you just have a few seconds to push the paint where you want it. No more than that. Then you must leave it to find its own level.

See, if you move the paint once it starts to dry, you’ll undo your good work in getting the paint well-mixed in the first place. Which again means some parts will have more gum than others. And they’ll crack and blister.

Enough said.

Now watch what I mean.

The Heraldic Arms of Hampton Hall

Yes, it always helps to watch.

And unless you’re lucky enough to have a friendly master glass painter at your call, you won’t ever see close-up scenes like these ones here.

So turn on the volume and watch closely.

Let’s go!

And note the “full-screen” icon on the right …

I hope this helps. As always, questions welcome.

Now for the fifth and final cause, which I can’t show you in the film.

Cause #5 – wrong firing schedule

This is easy: if you fire flooded paint too quickly – meaning, it doesn’t have proper time to dry out – you’ll give your glass every encouragement to blister.

  • So this is an easy thing to put right: just adjust your firing schedule.
  • Make it slower overall. Rise more slowly to your first ramp. Stay there longer. Rise more slowly to your top temperature.

Here’s what works for us. We always start by taking one hour to move to 100 Celsius / 210 Fahrenheit, where we soak the glass for 15 minutes. Then we go to top temperature over three hours.

I hope this helps.

All the best,Stephen Byrne


P.S. For weekly tips and techniques, join here, because if you’re not getting the email tips, you’re missing out on something very special. Don’t miss out.

29 thoughts on “Blistered Paint

  1. As a complete newbie (I’ve yet to set brush to glass, my first class will be in March) it is such a help to me to watch these videos. I’m always shy about taking new classes as well as when I’m attending them so it helps immensely to have some knowledge of technique beforehand.

    As I was watching this video it occured to me that the outline must be dry and as such would serve as a sort of minor barrier as you flood the glass. What I mean to say is if your outline was still wet as you flooded the glass the outline would move as well. Am I correct in assuming that the outline must be dry before you proceed with flooding?

    • Hi Astrid,

      You’re absolutely right in your assumption. And you’ll find the whole process very well described in section 1 of Part 1 of the PDF e-book, Glass Painting Techniques & Secrets from an English Stained Glass Studio.

      Very quickly, it goes like this: start with clean glass, then apply an undercoat which you allow to dry, and then (with the glass on top of the design) you trace the image. Once this line is dry, it indeed serves as a barrier against the thicker paint you use to block in.

      I’m glad you like the videos. It’s great we live at a time where we can explain ideas with words and also images like this.

      All the best,

  2. Hello everyone!

    Stephen, you did it again! This is one more top quality post. You and David keep us always looking for more. Can’t wait to meet both of you!

    Keep it up!

  3. Being, after many months of procrastination one of the 57 newbies I have received and am slowly absorbing the wisdom in the DVD and e-book.

    All I can say is that it is a brilliant and careful instruction in your techniques that I would not do without. Both inspiring and practical, a difficult balance to achieve. I teach leaded glass and as a teacher love the pace of your approach. My 3 year old daughter is also hooked!

    I work in a number of traditional skills including lime plaster and over the last week I have been mulling over the painting information obtained in my evenings reading or viewing whilst lime plastering a 140 sq metre barn ceiling. It was problematic due to the size of the area and the particular form of lime being used. The main issue being how to work forward as the plaster dries behind you, and how much you can revisit already plastered areas before it damages rather than improves. I could not help but draw parallels between this medium of plastering and flooding, which works on the same priciple. You want the material to be laid on without disturbing what is already there, but it must marry continuously. The only answer seems to be careful and precise speed not haste, maintaining constant awareness of the wet edge and working constantly to keep that “wave break” flowing onwards. Possibly needlessly overthought but once I made that link things seemed to work a lot better….. Practice practice practice practice.

    Thank you again for your fantastic altruism and looking forward to the free bonus.


    • Hi Matt,

      Good observation: there are many insights which different crafts share with one another – precisely by virtue of being crafts. And learning how to judge the right pace is definitely a common and important skill.

      All the best,

  4. Hi Stephen,

    Thanks for the post! The Glassworx Kiln Collective will be firing sometime this week for the first time with our wonderful big kiln. With what I’ve learned from you and passed on to the others already, I think blistering is going to be one of the things we DON’T have to worry about!

    And I’ve done up a blog now tracking the beginnings of my painted glass career. Check it out!


  5. Hello Stephen,

    Thanks for the posting and for the prompt DVD delivery, I am enjoying it very much and finding it extremely useful.

    I have found it amazing how seeing something on a video has actually given me a much better understanding of flooding. I have had the problems with cracking and tried all of the things with gum Arabic and mixing to resolve it – seeing it on video, I have just been laying it on far too thickly and hadn’t understood that the brush does actually move the wet paint.

    Brilliant, can’t wait to have another go now!

    Do you have anything on video on the site for softening traced lines? (Another skill I’m determined to gain.)

    Many thanks

    • Hello Angela,

      Thanks for your message and that’s great you find the DVD so useful.

      About softened lines: yes, they’re important, and this guide to painting a stained glass face also comes with 96 minutes of online video.

      Which means you see exactly how it’s done.

      All the best,

  6. Hello Stephen,

    Your article on blistering could not come at a better time for me …

    Just this morning I opened the kiln to discover some blistering in my nameplate. The paint I used was a Reusche trace black that I use all the time. This time it was a fresh batch mixed just slightly different from usual.

    Your article on blistering related to a water-based method, of which I have had only some experience. (I’m exploring more). I agree with all that you stated.

    The connection here seems to be both the components, and the need for mixing to distribute them evenly.

    My initial medium was an older Reusche product #8200MB, to which I added just a very small amount of Venice turpentine. My thinner was gum turpentine. Subsequent tests done with different proportions of these ingredients revealed that I might have added just a little too much Venice turpentine.

    The folks at Reusche mentioned that they no longer make 8200MB. (They mentioned the use of clove oil with 8200MB, and that actually works well.) Thay also stated:

    “That when looking for a replacement I would recommend the 1903 oil based medium or the D1368 water-based medium. The 1903 has medium dry time, and the D1368 really doesn’t dry at all, it remains tacky even after a few days.”

    I have not tried either product yet. I would be interested to know if you have tried either of these mediums, especially the water-based one.


    • Hi Ron,

      Thanks for your comment. And first up (as you’ll see why) I’ll say straight away I haven’t tried either the 1903 oil-based medium or the D1368 water-based medium but I will.

      I say this (and mean it) because I don’t want you (or anyone else) to think I am a dinosaur (so to speak).

      But this is what intrigues me: what problems are these media meant to solve?

      And the reason I ask is this (which is how someone might think I am a dinosaur …): they didn’t exist in the Middle Ages (OK OK neither did kilns and kiln controllers, I know, and I wouldn’t want to be without them!).

      And I can understand how the demands of modern style (or even the Industrial Revolution) might require the introduction of new media.

      What I can’t yet understand is the use of modern media for traditional techniques.

      Which is not to say I disapprove of them. Not at all. It’s just I’m really intrigued.

      So can I ask you directly, and in the spirit of pure enquiry, what brought you to using a medium other than water or oil?

      Thanks again for your really interesting intervention here, Ron.

      All the best,

      • We’ve just done a set of tests with the water-based D1368 and it’s wonderful! Thanks for the suggestion, Ron.

        I’ll do a post on it quite soon.


  7. Hi Stephen,

    I would like to download the Heraldic Crest and since I already purchased the Diamond Lights of Hampton Hall, I know that it is free. Not sure how to go about it.


    • Hi Mary,

      You’re right it’s free but … I’m only just editing it right now. Once that’s finished, I’ll e-mail everyone about the free download, then send it off to be turned into a DVD which we’ll then send to you.

      It’ll be worth the wait because there are so many astonishing, lovely and instructive scenes.

      All the best,

  8. Thank you for all the wonderful info!

    You guys make me feel like I’m sailing to Byzantium.

    If you do have a small blistered trace line, would you start over or press on?

    Thanks again!

    • Hi John!

      The answer is, it all depends. Right now, for example, we ourselves our working on the tycoon’s 16 stained glass skylights. And the brief is … the glass must look ancient. So we’re having to find all kinds of ways to rough it up – including intentionally bruising and also blistering the trace lines. See, if we don’t do this, the tycoon won’t be pleased …

      So the general consideration is, Does the blistering damage what you wish to achieve?

      Another example: if the piece is only seen from a distance, and the line itself is beautifully curved but also slightly blistered, does the blistering really matter? It depends of course.

      Always best to know how to achieve what you want to achieve, time and again.

      Also, blistering is only likely to occur with trace lines when you go over a trace line before it’s completely dry.

      In other ones, it’s simple to avoid blistered trace lines: let them dry before you paint over them!

      I hope this helps. Thanks for the question!


  9. Hello there,

    I’m returning to your very helpful page on cracking up again since I have a problem that hasn’t occured before and wondered what your thoughts might be.

    Following your useful paint-mixing instructions has worked extremely well for me up – until the last batch I mixed.

    I have always used powdered gum arabic before, not having liquid.

    And this time, the gum arabic just seems to separate out no matter how much I mix it in the lump and on the palette. As soon as the water hits it, there is a white layer persistently rising to the top – a particular pain when flooding.

    Am I adding too much water?

    Any philosophical or directive thoughts greatfully received.


    • Hi Angela,

      If it’s separating, there’s surely some kind of imbalance. But more than that, I do not know because I just don’t use gum Arabic in powdered form: sorry!

      I appreciate you already have a batch of paint, so you will not want to waste that, and so it will be important to correct the imbalance if this is possible. Maybe remove a few teaspoonfuls from the “separating batch” and add more paint to it?

      Going forward, it is easy to get liquid gum Arabic, and it really is easier to use than powdered gum Arabic: it’s used by water-colour painters, so any shop or online store which caters for water-colour painters will be able to help you.

      I hope this helps.

      All the best,

  10. Angela,

    The thing with gum arabic is that it’s intensely water-soluble. The best way to get consistent results with the stuff if you can’t buy it in liquid form is to liquefy it yourself – plonk a couple of teaspoons onto your palette or into a mortar and pestle, drop in enough water and grind it with your mixing knife to make a paste. Then (to save it drooling all over your palette), scoop the paste into a jar and stir in a little more water to make a useful runny liquid.

    If you’re having trouble with your paint separating, it’s due to there being too much water in the mix.

    The white layer on top is indeed gum arabic, but it’s there because the few paint particles in your mix have sunk and there there aren’t enough left for the gum to cling around.

    Try lifting the runny paint up with your brush onto your lump of paint and dragging more paint down into the working pool.

    You still have to swish and twirl your brush each time you load it with paint, but it needs to stay mixed in the mixture you’ve put it into until dried. If it’s separating on your palette, it’ll separate on the glass.

    The hallmark of correctly mixed paint is that it sits happily on the palette (and more importantly, on the glass) for the minute or two so without separating until it’s dried.

  11. Hello Stephen and Rtistdug,

    Thanks very much for your comments, I will get on to trying to save the batch today and using the liquid in future. Have to admit, whenever visiting glass supplier stores I had never seen it in liquid form so assumed it was something rare and exotic. Had no idea it was a familiar artists product. So, something else I have learnt.

    Thank you so much.


  12. I’ve been working my way through the Silhouette examples in the first tutorial and I’ve been learning a lot about glass painting. I have a persistent problem. Within the body of the flooded areas I can readily achieve the “millpond” stillness you referred to elsewhere, and indeed when it is fired the body of the flood comes out smooth, uniformly opaque and without any cracks/blisters. My problem is where the wet flood washes up against dry paint. ie- the trace lines, or perhaps another piece of flood that is already dry. The paint is non-uniform in these places and when it fires a crack often opens up along the boundary between the flood and the dry paint.

    I’ve gone though the causes listed and I think my problem comes down to bad technique. I *think* I can fix the problem by having the flood flow on *top* of the trace line. (The theory being that the water in the flood will soften the dry paint and bond to it).

    Is that what you do – or am I missing something?

    Thanks for your help – and for the excellent instructive materials.

    Jason H.

    • I’ve performed some experiments and I think most of my problems stem from too much gum arabic in the paint. I’m using powdered GA. In the part 1 tutorial it suggests 2g of GA per 100g of (dry) paint. If I reduce that to 1g of GA per 100g of paint most of the flood/trace cracking goes away. The adhesion of the paint to the glass is still quite good. BTW – I’m using a milligram accurate chemistry scale to measure the GA and the paint – so the above numbers are suitably accurate and should be reproducible by others.

      • Hi Jason,

        You and I have been in contact via e-mail and I think you’re now well on your way to getting things exactly as you want them to be.

        For now, let me say there are some relevant details in this post right here.

        All the best,

  13. I’ve only recently gotten back into the painting. I have a bit of blistering. Can I re-paint and then re-fire to correct it or is it a ‘what is done is done’ sort of situation?

    • If you’re prepared to use the kiln, it’s usually worth a test. Slightly blistered lines can sometimes be rejoined so they’re now completely black (though not entirely smooth). Blistering in a larger area is difficult to hide / repair.

      Welcome back!


  14. Thank you!! Once again, I ran into some technical issues, and so I went back and searched through both your site and your book – and I found my answers!

    Thank you, thank you, thank you!

  15. Thank you very much for sending the link. It is allways nice to learn something new, and you help me a lot with your lessons.
    Best regards, Adela (Guatemala)

Comments are closed.