The Preparation You Must Do Before You Do The Work


57 sheets of glass delivered to the studio since January: 33 from Germany, 8 sheets from France, 10 from Poland.

And – in 4th place – 6 (just six sheets: the pity of it!) from our own dear England.

Let’s say: 16 square metres in all.

And now, two months later, only a handful of the 57 sheets remains intact:

Antique stained glass

… while most of it is cut.

From scale designs to full-sized cartoons

Yes, late last year, David took forward these scale designs:

Scale designs for stained glass windows

… and developed them into full-sized cartoons:

full-sized stained glass cartoons

I’ll return to this big theme another day but for now I will say that, on purpose, absolutely on purpose, large areas of the cartoons remain gestural and under-determined:

Stained glass cartoon - close-up

This is paper, after all. And paper is not glass. You might say the paper is a recipe; it is not the meal itself. Therefore we decided it was important to leave large areas under-determined. Our job as glass painters, when the time comes (which it will do very soon) is to “think on our feet” with glass paint and various liquids, and develop the images for this specific medium. Sometimes (not always, sometimes), if too much is resolved on paper, it can hinder what you want to say on glass. And here is one of those times. It would cramp the actual performance.

Working with raw glass – unpainted colours

Now these past two months have required a particular kind of thinking: working with raw (i.e. unpainted) glass, imagining their potential, creating new colours by plating:

It’s not a peaceful experience because we are happiest only when an undercoat goes down and we begin to take control of the ‘canvas’.

And testing is tiring: you’re always probing, asking questions, then waiting for particular replies; replies which, sometimes – that’s the point: to learn – you simply do not get.

Test #1 – does it cut well?

Not all glass is equal. To my eternal embarrassment, the worst offenders are the English. It sometimes seems they imagine it is sufficient that the glass be beautiful. But, horror of horrors, the practical reality of earning money as a glass painter requires that the glass, however seductive, is considered little more than a means to an end.

(Oh yes: dear glass blowers, whatever your undoubted talent, please remember that, when you work for glass painters, it is we who are the artists.)

So if the glass cuts unpredictably, or if it costs us our peace of mind – neither of us, despite the isolation of the studio, enjoys being brought to a state where we use intemperate language – then  it is scarcely any good at all.

But I gladly admit some shapes are more difficult to cut than others. And so, if the requisite depth of colour can only be achieved by thicker glass than usual, I will be unsurprised when some spires or curves prove hard to cut with one glass by comparison with another. We will not accuse a glass on that count alone: we are realistic here.

All along, the main task here is to discover or remind ourselves how each glass likes being cut. (And the less we cuss the better.)

Test #2 – how well does it take silver stain?

A second test has been: establishing how well the glass takes different kinds of silver stain:

Silver stain samples

When a result was disappointing, we might adjust the schedule. Or leave the kiln lid a little over so that more oxygen gets in. (That’s why, above, you see several samples for each kind of glass.) Or try a different kind of stain.

Test #3 – how does it cope with sand-blasting and fire-polishing?

Another kind of test was: how well will the flashed glass ‘survive’ sand-blasting then fire-polishing? (We plan to sand-blast and fire-polish first; then, once the ‘gross’ work is completed, use acid for the subtler variations.)

Some glass (as we knew it might) changed colour with the extra heat which fire-polishing demands. This purple shifted unacceptably towards brown:

Heat can change the colour of the glass

Some glass just couldn’t take the strain:

Stained glass heat-cracks

Do you see the frightful cracks?

Now I assure you we annealed it very well ourselves. So my suspicion is, it wasn’t well-annealed when it was made. It is just as well David discovered this in time. This glass was not robust enough to endure the treatment we proposed for it, so he chose a different orange.

(Note: a tough test this one, because sand-blasting, even with fine sand, is fairly punishing. So we are careful: we simply want to know what the glass is capable of.)

Good fortune

David even found good use for some glass we wouldn’t usually dream of working with:

Unusual glass for us

Look closely at the fillet borders:

Stained glass design: check the fillet borders

Do you see the dark borders on the inside and along the bottom? Well, that glass we wouldn’t usually work with is perfect when slapped down with an uneven wash of paint, badgered roughly, then rubbed by hand when dry:

Unusual glass - perfect now

Test #4 – (finally) painting …

And as soon as the glass passed these first two tests (staining and sand-blasting/fire-polishing), then onwards to the next level where (thank goodness) we can take the raw, crude colour like the sand-blasted, fire-polished off-cut on the left …

Stained glass before and after painting

… and AT LAST start figuring out the techniques to use when painting it.

Tests, tests, tests and more tests. All the time we are surrounded in the studio by those painfully raw, bright, vulgar colours which (speaking to you as glass painters) we are just longing to cover with vitreous glass paint and silver stain.

Experimenting with techniques – watch the high-speed video

Here’s David, working with propylene glycol on top of unfired water-based (traditional) glass paint, exploring the opportunities for shading; and also finding out how well the sand-blasted, fire-polished ‘canvas’ takes paint:

From January till today: a restless time.

But now we’re nearly at the point where we can start to paint. And not stop till the work is finished.

Rehearsal, preparation, testing, solving problems before they rear their head – and then to work.


I wonder what takes up most of your time before you do the work itself. I’d love to know!


Stephen Byrne of Williams & Byrne the glass painters

P.S. Very soon, we’ll re-open our video course for intermediate glass painters. These demonstrations, discussions and designs are based around the course we ran at Bryn Athyn college in June 2014, which sold out just two hours after booking opened.

Those of you who already joined, you’ll hear from us just after Easter.

There’s also room for a few more: we’ll send out details next month.

25 thoughts on “The Preparation You Must Do Before You Do The Work

  1. So great to hear from you! And so kind of you to share so many details of your trial and errors, so that we can learn from you (as we always do). I appreciate it very much.

    Love, Candida

    • Hey Candida! Well, you know how much we enjoyed your company when you came over and spent those days with us. And so I know you will also understand how, for us, writing and film-making are just different ways of keeping the studio doors open. Of course it’s not the same as face-to-face. But nonetheless it is a happy way of swapping ideas with you: thank you always for your attention!

    • Oh I had a marvellous teacher in Patrick Reyntiens all those years ago when I was a bushy-tailed youth!

      All the same, I do believe I’ll take a more leisurely pace with the window itself.

  2. So interesting. So helpful. Thank you so much for taking the time to document and post your work. Seeing your design and it’s progression was just great. I have a question, if I may, about fire polishing. I’ve been doing some samples to check temperatures but it would be quite expensive to sample a piece of glass the actual size I will be using in my design. Will the size of the piece (a bit bigger than A4) change the polish time/temperature relative to my sample which is about 5cm?

    • Let’s agree you should allow a longer annealing time for the larger piece: it costs you very little to be cautious here.

      Testing the smaller piece, you’re looking for these two things at least.

      Firstly, are you going hot enough to smooth the glass sufficiently so that cement won’t stick?

      Secondly, are you not going so hot that the edges deform in an unacceptable manner?

      I believe you can extrapolate from the smaller piece and be confident about the first point.

      But the second point involves a risk: edges will deform also based on the shape you cut and other factors. I don’t believe all miniature shapes will reliably predict how their expanded counterparts will fire. So, Sophie, there is a risk. But you have to find out somehow. So you might as well run the risk (based on information you previously got from the smaller piece). If it doesn’t work, then your large piece just becomes your test piece; and you will have learned something.

  3. How fortuitous that you posted because I am about to start a glass painting project and have just cut and cleaned the glass.

    I planned to go directly to the scrubbing prep with background (diluted) paint then wiped off which I assume removes any remaining organics from cleaning. This has to produce a compatible surface tension for the next paint applied.

    How important is the fire-polish pre-painting? Several locally in the Pacific Northwest US do recommend that step, others perform well without.

    I am at the crossroads.

    • If we sand-blast, we always fire-polish. Paint is one consideration: the rough surface would usually give us textures which we don’t want. But cement and (post-installation) dirt are the concerns which really matter to us. However, to be absolutely clear, this is us. People must do what works for them and their work. I’m sure there are sealants, for example. I hope this answers your question. On the other hand, because you don’t mention sand-blasting, maybe I’ve misunderstood something. In which case, please tell me!

      • My question was one of process for any glass painting: do you recommend fire polishing stock but cut to size glass (clear, in my case) before any painting is performed? Or is that necessary to just test the heat tolerance of the glass piece first ?

  4. Love these designs and colors! Are these the windows to be installed in August? Hope so!

  5. Hello Stephen, thanks for the above posting. Very interesting. Now I know why my red glass turned yellow after firing! Also why sometimes the paint wouldn’t all fire properly. I suppose I can blame the glass! Or on the other hand me for not testing! Farook.

    PS. I was talking to some of our tiny glass guild members here in Brisbane and was delighted to hear three of them had bought your e-books. We have formed a little group.

    • Hello Farook: well, antique red glass (in my experience) is unpredictable. Machine-rolled red is fine; but antique …

      On your P.S.: I think working in a group can be hugely beneficial. David and I have learned enormously from one another: if you like, we have our own group where it is fine to challenge each other because we are both, ultimately, after the same goal – solutions which will work best for our clients.

      So if your group can also work together (with you each giving the one another the occasional guiding hand or even shove), you’re on to a good thing there.

  6. I always test fire the glass with and without paint to see if the glass changes color or if the paint doesn’t work well with that particular glass. Test, test, test.

  7. I am so grateful and have learned so much from your DVDs books and letters but I am plagued by a kiln that runs to hot so all the normal tests you do are multiplied.

    I have just devised a paint firing schedule that works on St Juste flashed glass: the top temperature in my kiln is just 610 c., and I anneal between 515 and 500 c.

    Even so, the last piece of flashed brown on clear shrank and distorted a little at these temperatures, so how much do you reduce the temperature at a time to resolve the problem and is this only at the top or throughout the schedule please? I.e. do I reduce the 515 annealing temp as well, because that is probably also firing too high?

    What do you think I might try as a fire polishing temp for sand blasting given that my kiln is hot?

    Thank you again for all your help.


    • Oh dear, I’m sure you already suspect at least some of what I am going to say: namely, that from where I am in Stanton Lacy there is no knowledge I have which will allow me to guide you reliably here.

      I understand from what you tell me that “610” c. in your kiln equates to roughly “665” c. in one of ours.

      Since glass appears to fire and anneal well in a Speed-burner, I believe your own guess at where you should anneal in your electric kiln is good enough.

      But a paint firing doesn’t subject your glass to as much thermal shock as fire-polishing after sand-blasting.

      So my own inclination for fire-polishing would be to widen the annealing span e.g. from 520 to 470 – just to be on the safe side.

      That doesn’t solve your question of what temperature to use for the fire-polishing itself. And this question, from where I am, just is not something I can help you with.

      If your kiln has a controller, I wonder if you can find a kiln manufacturer who would look at it for you. Or maybe you should get a new thermocouple. But unfortunately these things are far beyond my knowledge.

      Practically, if you can’t straightforwardly do what you want to do, you must either resign yourself to constant anxiety and uncertainty; or you must change something (repair the kiln; use different techniques i.e. gentler sand-blasting, no fire-polishing etc.; use different glass).

      Last word: it is one thing for a kiln consistently to fire hot/cold.

      But it is quite another and wholly unacceptable thing if the kiln becomes unpredictable. So if you have any suspicion it is unpredictable – “it”: the kiln: I don’t mean the glass: but the kiln – then you absolutely must resolve to change something about the kiln.

      When you get through this, you will be stronger.


  8. I noticed you were experimenting with propylene glycol. One of the glass artists in my area and I were talking about the difficulty I was having with the lead free glass paint in getting it to be an even texture without lumps, even after a great deal of mixing. She said that she uses propylene glycol as the base instead of water and it is smoother and in doesn’t dry out in the containers or glass pallet. I wondered if you had ever heard of that being done and if so, what are the problems if any? It would seem that blending might be a problem since it takes longer to dry than with water. I have been experimenting with it and so far am getting better results because I just could not get the texture right (the mixing) with water based.

    • What brand of lead-free paint are you using? The reason I ask is: the lead-free paint I’ve used is too smooth for my personal taste. So just the opposite to the situation you find yourself in! That’s why I’d love to know.


      OK. You wrote to me and I understand you’re using Reusche. The lead-free series which I tested was Series 5 Tracing Black (E401) (results here). And like I say, I found it incredibly smooth.

      In fact, too smooth for me. So, since we protect ourselves from lead poisoning in other ways (e.g. working with a good-sized lump of glass paint, not a teaspoonful that’s forever drying out), we continue with Reusche’s lead-based glass paints, which we absolutely love.

      Now in your email you say you “have a great deal of difficulty getting it smooth without lumps and streaks”.

      I need to ask you to clear things up for me here.

        It’s one thing for someone to find it difficult to get the paint smooth i.e. when grinding and preparing it.

        It’s potentially a very different thing for someone to find it difficult to lay it down without “lumps and streaks”: very different causes might be operating here.

      Given that, for us, the lead-free paint is actually too smooth, I’m just wondering if something else is going wrong.

      So please say more: thanks.

  9. Thanks for your reply. Sorry I didn’t get a notification that you had replied and kept checking but for some reason it didn’t show up until today, on here, still no email notification.

    I looked at your review of the lead free paint and your experience was completely different from mine. I live in a dry climate (California), so maybe that has something to do with it. But when I tried mixing it, I found it nearly impossible to get it lump free and when applying, streak and lump free. I am not sure what I was doing differently. Maybe I am mixing differently? I will search your site, and check on the DVD I have (the “Lights” one) and see if I can discern any difference, though I have done that before, but not in a while. I don’t recall the “grinding” part, so maybe that is the difference? In my experience it dried quickly also, but like I said, I live in a very dry climate compared to yours.

    When I used the propylene glycol, it mixed better, especially if I left it overnight or longer. It keeps wet for a long time, but because of that, blending is pretty much impossible. It just smears if still damp, which is why I am asking for help because I really don’t want to keep using it, but need a solution to getting a good consistency.


    • Thanks for your helpful answer, Caryn. And, yes, England is damper than where you are!

      Now, two important features about how we work.

      1. We mix a lump of glass paint, not a teaspoonful: this means that, carefully attended for sure, the paint within the lump has time and opportunity to swell up and soften.

      2. We apply the initial undercoat with a large Ron Ranson hake. This type of brush is a joy. I’ve used other brushes and done well enough. But this brush is amazing.

      Why do I mention these two points? Because they make a real difference to getting the paint nice and smooth (point 1) and also to laying it down as evenly as is humanly possible (point 2).

      But all the same I don’t mean this dogmatically. What I mean is: this works for us. And so I wonder how you mix your paint and what you use to lay down your undercoat …


      P.S. This day one year ago I was rather jet-lagged but deliriously excited to wake up on the first day of my first trip to your magnificent country. I can’t wait to return. Not to escape the dampness but to be amongst your countrymen and -women again.

      • Thanks, Stephen. When I ordered the glass paint it came in a little jar, so little it wasn’t enough to make much of a lump, so I just made it up as a very thick paste, then worked it on glass like you do with the lump, but it was too small of a lump to be very effective. Perhaps that is my problem. Maybe I need to order a large amount so I can make a large lump?

        The brush I use is similar to yours. It is natural bristle but is in a fan shape. It was one of my mother’s she used for her paintings. Very soft and flexible. Maybe I will order the one you use and try that.

        Mostly what I have been doing so far is using the paints to put in details on my fused glass pieces, since I was having such bad luck with doing anything more complicated. I would like to do some of the kind of work you do, if I can get it to work better. I loved your Martha face video.

        Glad you liked our country. I’d love to visit yours someday too. My son, an animator, directs a studio in Paris, so someday I might head that direction to visit. He doesn’t seem to be in any hurry to head home, and they keep signing him up for new films, so I might just have to come over the pond to visit.


        • Hi Caryn,

          Thanks for this information about the quantity of paint you have and also the brush you use.

          OK so the first thing I’d definitely suggest is: invest in the same brush as us. Namely, the large Ron Ranson hake.

          My reason is: we know it works.

          So using it will just kick away one of the factors which might be giving you trouble. (Maybe the brush you have is fine. My point is, we know for sure the Ron Ranson hake is great for loads of things. If you use the same tools as us, we can work together to help solve difficulties which let’s face it are going to arise because we all work in different climates, with different characters and anatomy, on different projects.)

          So I really suggest you start there. With the brush.

          Now, if you also wish, get more paint. Good idea, but not absolutely necessary.

          Thing is, we know Reusche paint is excellent. So you just have a small amount. But it’s always possible to work with smaller quantities of this excellent paint – whereas it’s often a needless nightmare working with a brush or tool that isn’t the best it can be for the job in hand.

          That OK and clear?


          P.S. Are you happy with your badger blender?

          • Ok, thanks. I checked online for a Ron Ranson hake and there are a bunch of different ones. I’ll check on your videos and see if I can figure out which one you use.

            Not sure if the badger blender is the issue. Will find out as you suggested.

            Also you mentioned grinding in the preparation? Do you show this on your site somewhere? This may be the problem, or one of, if there is more preparation before mixing than I knew of. I’ll see if I can find something on it.


Comments are closed.