Help Needed, Troubleshooting My Oil Varnish-Making Process

[ElliotL]

Hello.

 I have been making varnish and trying various adjustments, but recurring issues persist. Specifically, the varnish doesn't dry properly, and a film forms instead. Here is the method I used.

---

Varnish Preparation

1. Rosin Preparation

Heated 50g of rosin at 250°C for 2 hours and 30 minutes.

Dissolved 5% calcium hydroxide in water, mixed it with the rosin, and heated it at 220°C for 1 hour and 10 minutes.

2. Oil Preparation

Prepared 33.3g of Kremer cold-pressed linseed oil, washed it three times with water.

Heated the oil at 250°C for 2 hours and 30 minutes.

3. Mixing

Combined the treated rosin with the treated oil and increased the temperature to 250°C.

4. Aluminum Chloride Addition

When the temperature reached 250°C, gradually added 1.67g of aluminum chloride (2%) in small portions.

5. Final Step

After cooling the mixture to 100°C, added 10g of tinner.

I applied the varnish and cured it using UV lamps.

---

If there are any issues with this varnish-making process or areas that require additions or adjustments, I would greatly appreciate your advice. Thank you!

Edited December 1, 2024 by ElliotL

[DonLeister]

How many coats is that? It looks thick.

I’m unfamiliar with putting calcium hydroxide, aluminum chloride and tinner (sp? Maybe thinner) in so can’t comment on that.

[Dean Markel]

I’m rather new at this too and can’t comment on your recipe but I was surprised how thin one needs to apply each coat of varnish. I make sure I brush (or rub) it out as thin as possible for fear of thick spots that wrinkle or worse, don’t dry.
I think Don is accurate in his guess that thickness may be your issue. 

[uguntde]

2 hours ago, ElliotL said:

Hello.

 I have been making varnish and trying various adjustments, but recurring issues persist. Specifically, the varnish doesn't dry properly, and a film forms instead. Here is the method I used.

---

Varnish Preparation

1. Rosin Preparation

Heated 50g of rosin at 250°C for 2 hours and 30 minutes.

Dissolved 5% calcium hydroxide in water, mixed it with the rosin, and heated it at 220°C for 1 hour and 10 minutes.

2. Oil Preparation

Prepared 33.3g of Kremer cold-pressed linseed oil, washed it three times with water.

Heated the oil at 250°C for 2 hours and 30 minutes.

3. Mixing

Combined the treated rosin with the treated oil and increased the temperature to 250°C.

4. Aluminum Chloride Addition

When the temperature reached 250°C, gradually added 1.67g of aluminum chloride (2%) in small portions.

5. Final Step

After cooling the mixture to 100°C, added 10g of tinner.

I applied the varnish and cured it using UV lamps.

---

If there are any issues with this varnish-making process or areas that require additions or adjustments, I would greatly appreciate your advice. Thank you!

Why Ca(OH)2 and why ALCl3? What do they do?

[LCF]

Adding lime is common enough but I have never heard of adding aluminium chloride to a straight varnish mix. Whose recipe is that?

 

I suspect it ( AlCl3) might be antioxidant or at least oxygen scavenging thus retarding the oxidation and polymerisation of the varnish film. 

 

Also it looks thick, as mentioned.  

[David Burgess]

I'm tending to go along with the other guesses that the application layer is too thick. Oil varnishes absorb oxygen and increase in volume and weight as they polymerize. When the outermost portion increases in surface area during drying, while what's underneath remains liquid enough to allow what's drying on the surface to move, the increase in surface area confined to the same space will tend to produce wrinkling.

[Andreas Preuss]

Though I never went deeply into cooking varnish , what I heard in general from experienced makers is that (somehow) the longer you cook it the better. And this means at rather low temperatures for very long time . (One maker I talked to cooked his varnish for one week) If done correctly the colophany turns red as well. (This seems to be the big secret.) 

For your quick receipe I’d cut down the amount of linseed oil from 1:2 to maybe 1:4 and dilute it in more turpentine. It’s the linseed oil that makes the varnish wrinkle at the surface.

[nrlewis]

I had the same thing happen on violin #2....#1 I used a commercial varnish, #2 I made my own. I'll have to check my recipe when I get home. When I put it on initially, it was fine, but really soft, and over time it started doing the same thing and looked really nasty. I always thought it just took too long to cure, but I also know that I used some of those "interesting" ingredients in that recipe. I also used a hot plate that I don't think got as hot as it said that it did. I recently redid the varnish because I couldn't stand it...I didn't strip it off completely, just sanded it down significantly and applied a new varnish (this one is the recipe from the Michelman book). I guess I'll see how it turns out a year from now, it might take a full strip to get right. Not a perfect solution, but at least I'm not embarrassed by it now. 

I'm starting to go with the philosophy of "less is better" for varnish recipes...keeping it simple. I've recently been reading through this forum to get ideas for a simple but effective recipe....we'll see what I end up doing! Still a good year away from varnishing #4 (#3 turned out well with the Michelman, but it's pretty complicated)

[ElliotL]

On 12/1/2024 at 6:34 PM, DonLeister said:

How many coats is that? It looks thick.

I’m unfamiliar with putting calcium hydroxide, aluminum chloride and tinner (sp? Maybe thinner) in so can’t comment on that.

I tried applying it over a pre-painted surface I had already been using.
That's correct. It's tinner (oldwood).
Also, I applied the coating somewhat thickly.

[ElliotL]

On 12/1/2024 at 7:33 PM, Dean Markel said:

I’m rather new at this too and can’t comment on your recipe but I was surprised how thin one needs to apply each coat of varnish. I make sure I brush (or rub) it out as thin as possible for fear of thick spots that wrinkle or worse, don’t dry.
I think Don is accurate in his guess that thickness may be your issue. 

Thank you for your response.
I’ll try applying it as thinly as possible.

[ElliotL]

On 12/1/2024 at 8:32 PM, uguntde said:

Why Ca(OH)2 and why ALCl3? What do they do?

The recipe I received instructed me to add those ingredients.
It mentioned that rosin is already acidic, so calcium hydroxide is used to neutralize it and improve the durability of the varnish. As for aluminum chloride, I understood that it helps to reduce drying time.
I tried it out to experiment, but it seems I ended up with an unsuitable varnish.

[ElliotL]

20 hours ago, David Burgess said:

I'm tending to go along with the other guesses that the application layer is too thick. Oil varnishes absorb oxygen and increase in volume and weight as they polymerize. When the outermost portion increases in surface area during drying, while what's underneath remains liquid enough to allow what's drying on the surface to move, the increase in surface area confined to the same space will tend to produce wrinkling.

Thank you for your advice.
I’ll give it a try right away.

I'm glad to have learned this now, even if it's a bit late.

[ElliotL]

8 hours ago, Andreas Preuss said:

Though I never went deeply into cooking varnish , what I heard in general from experienced makers is that (somehow) the longer you cook it the better. And this means at rather low temperatures for very long time . (One maker I talked to cooked his varnish for one week) If done correctly the colophany turns red as well. (This seems to be the big secret.) 

For your quick receipe I’d cut down the amount of linseed oil from 1:2 to maybe 1:4 and dilute it in more turpentine. It’s the linseed oil that makes the varnish wrinkle at the surface.

I agree with you, but I also use rosin cooked for 84 hours and 168 hours.
Perhaps the issue lies in the proportion of oil.
Additionally, thank you for your advice.

[Sean Couch]

I am not an expert by any means with only having finished one violin so far. I made my own varnish with cooked rosin. For my “color varnish” I cooked the rosin for 168 hours at 350F. It reduced by about 24%. Below is the color change progression and what one of the pieces of cooked rosin looked like when backlit with a flash light:

 

I used equal amounts of cooked rosin to washed linseed oil and then added 5% mastic and cooked it for 2 hours at 350F. My varnish isn’t dark enough for my taste. I have since learned that for color, the number of hours that you cook the rosin is not important, it is the reduction that is important. Shoot for a minimum of 50% reduction in weight, 66% would probably be better.

I also made some clear varnish. That was a 2 hour cook for the rosin at 350F. Then again with equal weights of rosin and washed linseed oil and 5% mastic cooked for 2 hours at 350F. I didn’t use lime or any other chemicals during the process.

Both of my varnishes dry to the touch in about 4 hours when in sunlight or my UV cabinet. To pass the “finger print test” it takes about 24 to 30 hours in the UV cabinet so that when you firmly press your finger onto the varnish for 20 seconds, no imprint is left.

The picture of your varnish looks like the pictures I have seen where people are making linoxyn. That would leave me to believe, as others have said, that your varnish was put on too thick. For my violin I did 7 coats of varnish total (clear and color). The weight of the 7 coats of varnish is around 11 grams and is about 0.10mm thick. With the 1:1 ratio of rosin to oil that I used, it is quite thick (thicker than molasses) so to make it brushable, I diluted it 3:1 with high quality turpentine to get it closer to the consistency of honey. 3 parts varnish to 1 part turps.

When I applied it, the turpentine evaporates fairly quickly leaving it difficult to spread by brushing, so I tried to work fast and worked in thirds. Upper bout area, then the c-bout area, and then the lower bout. Get it on and spread out fairly evenly, but thin, before the turp flashes off. I did the ribs first, then the back, followed by the front, and the scroll I did last. 

Once it is all painted, but not totally nice and even, I would go over the entire violin with my finger tip and palm to “pad down” all the varnish. This is how I evened out the varnish. It is going to be messy and sticky, but it works well. You can put the palm of your hand in the varnish where it is thicker and then transfer that excess to where it is thinner.

Keep padding it down and moving varnish round until you are happy with the coverage. There is no rush on this part, you have plenty of time. My goal was to get an even orange peel look to all of the varnish. The varnish will flow back down and the orange peel look with go away.

It took my varnish maybe 5 to 10 minutes to level itself after padding. Pad down the varnish in the same order as it was applied. If you find that there doesn’t seem to be enough “thick areas” left to pad down and transfer to a thin area, paint a bit of varnish on your palm or finger and hit the light spot. Then into the UV box or out in the sun to cure.

[nrlewis]

I looked at my old notebook, it looks like my recipe that I used is the one found here: http://www.makingtheviolin.com/Wood ash oil varnish. Common ingredient in our recipes is the Ca(OH)2, but it looks like that's a common ingredient to help neutralize the acidity of rosin. Looking at my notebook, I cooked this varnish with a new hot plate and just trusted the dial settings--I didn't have a thermocouple to help calibrate. So while the recipe calls for temps in the 250-300C range, I maxed out my dial at "600C" before I even got some fumes. I quickly purchased a thermocouple and learned that my hot plate doesn't nearly reach the advertised temperature, so now I always use a thermocouple to get actual T readings. Mine was likely just undercooked. 

So with that, how much do you trust your T readings? Did you do a string test? I did on mine, but only noted that it was "inconsistent." I'm still fairly new to varnish making, but it seems to me like an inconsistent (failed) string test should have indicated my varnish wouldn't dry properly. Even though I put on thin coats, I put a total of 5 coats on every 2-3 days and let it sit in a UV cabinet in between. Since the previous coat hadn't dried completely before putting the next coat on (I was impatient!) it's basically the same as putting on one super thick coat. Just a few ideas, keep trying and experimenting! It took me a long time to start getting finishes I was remotely happy with, and still a lot of work to go. 

[nrlewis]

FWIW, after this debacle of violin #2, I bought about 3 or 4 super cheap Chinese fiddles in the white, tested a few more varnish recipes, and practiced varnishing on those. That made me much happier when I did violin #3 (Hardanger fiddle). Now I'm working on #4 (another Hardanger) and have more confidence, but still experimenting. But I'd rather mess up on a cheap fiddle that I can then just toss rather than agonizing over redoing one that I had spent all the time on the woodworking side. 

[David Burgess]

19 hours ago, Sean Couch said:

I am not an expert by any means....

Being an "expert" can matter, or not. I think that in your post above, you have made a highly valuable contribution.

[DonLeister]

On 12/2/2024 at 12:03 PM, nrlewis said:

I had the same thing happen on violin #2....#1 I used a commercial varnish, #2 I made my own. I'll have to check my recipe when I get home. When I put it on initially, it was fine, but really soft, and over time it started doing the same thing and looked really nasty. I always thought it just took too long to cure, but I also know that I used some of those "interesting" ingredients in that recipe. I also used a hot plate that I don't think got as hot as it said that it did. I recently redid the varnish because I couldn't stand it...I didn't strip it off completely, just sanded it down significantly and applied a new varnish (this one is the recipe from the Michelman book). I guess I'll see how it turns out a year from now, it might take a full strip to get right. Not a perfect solution, but at least I'm not embarrassed by it now. 

I'm starting to go with the philosophy of "less is better" for varnish recipes...keeping it simple. I've recently been reading through this forum to get ideas for a simple but effective recipe....we'll see what I end up doing! Still a good year away from varnishing #4 (#3 turned out well with the Michelman, but it's pretty complicated)

Nice save on that varnish job!

[ElliotL]

On 12/2/2024 at 10:06 PM, Sean Couch said:

I am not an expert by any means with only having finished one violin so far. I made my own varnish with cooked rosin. For my “color varnish” I cooked the rosin for 168 hours at 350F. It reduced by about 24%. Below is the color change progression and what one of the pieces of cooked rosin looked like when backlit with a flash light:

 

I used equal amounts of cooked rosin to washed linseed oil and then added 5% mastic and cooked it for 2 hours at 350F. My varnish isn’t dark enough for my taste. I have since learned that for color, the number of hours that you cook the rosin is not important, it is the reduction that is important. Shoot for a minimum of 50% reduction in weight, 66% would probably be better.

I also made some clear varnish. That was a 2 hour cook for the rosin at 350F. Then again with equal weights of rosin and washed linseed oil and 5% mastic cooked for 2 hours at 350F. I didn’t use lime or any other chemicals during the process.

Both of my varnishes dry to the touch in about 4 hours when in sunlight or my UV cabinet. To pass the “finger print test” it takes about 24 to 30 hours in the UV cabinet so that when you firmly press your finger onto the varnish for 20 seconds, no imprint is left.

The picture of your varnish looks like the pictures I have seen where people are making linoxyn. That would leave me to believe, as others have said, that your varnish was put on too thick. For my violin I did 7 coats of varnish total (clear and color). The weight of the 7 coats of varnish is around 11 grams and is about 0.10mm thick. With the 1:1 ratio of rosin to oil that I used, it is quite thick (thicker than molasses) so to make it brushable, I diluted it 3:1 with high quality turpentine to get it closer to the consistency of honey. 3 parts varnish to 1 part turps.

When I applied it, the turpentine evaporates fairly quickly leaving it difficult to spread by brushing, so I tried to work fast and worked in thirds. Upper bout area, then the c-bout area, and then the lower bout. Get it on and spread out fairly evenly, but thin, before the turp flashes off. I did the ribs first, then the back, followed by the front, and the scroll I did last. 

Once it is all painted, but not totally nice and even, I would go over the entire violin with my finger tip and palm to “pad down” all the varnish. This is how I evened out the varnish. It is going to be messy and sticky, but it works well. You can put the palm of your hand in the varnish where it is thicker and then transfer that excess to where it is thinner.

Keep padding it down and moving varnish round until you are happy with the coverage. There is no rush on this part, you have plenty of time. My goal was to get an even orange peel look to all of the varnish. The varnish will flow back down and the orange peel look with go away.

It took my varnish maybe 5 to 10 minutes to level itself after padding. Pad down the varnish in the same order as it was applied. If you find that there doesn’t seem to be enough “thick areas” left to pad down and transfer to a thin area, paint a bit of varnish on your palm or finger and hit the light spot. Then into the UV box or out in the sun to cure.

I’m deeply touched by your thoughtful response. One of the steps I intended to follow is included in your advice, and it’s encouraging to hear that your application dried so well. I’ll give it another try.

I also have some mastic resin on hand, so I’m excited to test the effects it might bring. By the way, passing the fingerprint test in just four hours is impressive! I’ll adjust the proportions of oil and resin accordingly. My current ratio is 4 parts resin to 6 parts oil.

In your recommended 3:1 ratio, what solvent did you use to make it easier to spread, and could you share any tips regarding the proportions?

Thank you for your guidance!

[ElliotL]

8 hours ago, nrlewis said:

I looked at my old notebook, it looks like my recipe that I used is the one found here: http://www.makingtheviolin.com/Wood ash oil varnish. Common ingredient in our recipes is the Ca(OH)2, but it looks like that's a common ingredient to help neutralize the acidity of rosin. Looking at my notebook, I cooked this varnish with a new hot plate and just trusted the dial settings--I didn't have a thermocouple to help calibrate. So while the recipe calls for temps in the 250-300C range, I maxed out my dial at "600C" before I even got some fumes. I quickly purchased a thermocouple and learned that my hot plate doesn't nearly reach the advertised temperature, so now I always use a thermocouple to get actual T readings. Mine was likely just undercooked. 

So with that, how much do you trust your T readings? Did you do a string test? I did on mine, but only noted that it was "inconsistent." I'm still fairly new to varnish making, but it seems to me like an inconsistent (failed) string test should have indicated my varnish wouldn't dry properly. Even though I put on thin coats, I put a total of 5 coats on every 2-3 days and let it sit in a UV cabinet in between. Since the previous coat hadn't dried completely before putting the next coat on (I was impatient!) it's basically the same as putting on one super thick coat. Just a few ideas, keep trying and experimenting! It took me a long time to start getting finishes I was remotely happy with, and still a lot of work to go. 

I have always wrestled with two thermometers in hand. Similarly, I have also made the mistake of applying another layer before the previous one dried, but I’m experimenting with many variations on many pieces. I’ve tried applying it thickly, thinly, and adjusting the amount of solvent... In any case, while this work is frustrating, it’s also enjoyable.

Perhaps if the varnish works perfectly, there might even be a sense of longing left behind...

[LCF]

On 12/4/2024 at 2:33 AM, nrlewis said:

I looked at my old notebook, it looks like my recipe that I used is the one found here: http://www.makingtheviolin.com/Wood ash oil varnish. Common ingredient in our recipes is the Ca(OH)2, but it looks like that's a common ingredient to help neutralize the acidity of rosin. Looking at my notebook, I cooked this varnish with a new hot plate and just trusted the dial settings--I didn't have a thermocouple to help calibrate. So while the recipe calls for temps in the 250-300C range, I maxed out my dial at "600C" before I even got some fumes. I quickly purchased a thermocouple and learned that my hot plate doesn't nearly reach the advertised temperature, so now I always use a thermocouple to get actual T readings. Mine was likely just undercooked. 

So with that, how much do you trust your T readings? Did you do a string test? I did on mine, but only noted that it was "inconsistent." I'm still fairly new to varnish making, but it seems to me like an inconsistent (failed) string test should have indicated my varnish wouldn't dry properly. Even though I put on thin coats, I put a total of 5 coats on every 2-3 days and let it sit in a UV cabinet in between. Since the previous coat hadn't dried completely before putting the next coat on (I was impatient!) it's basically the same as putting on one super thick coat. Just a few ideas, keep trying and experimenting! It took me a long time to start getting finishes I was remotely happy with, and still a lot of work to go. 

My first hotplate for varnish cooking was diy made with the element plus controller from a junked stove and it easily maintained temps near 300C but most new ones bought nowadays have limiters built in to stop them getting too hot. It may be necessary to hack into them and rewire which is probably illegal in some jurisductions, and always dangerous 

I still have the high temp varnish thermometer I bought long ago from a scientific supply co. however digital multimeters often come with a temperature probe or  you can use a cheap pid unit plus thermocouple simply as a thermometer. 

 

[Sean Couch]

On 12/3/2024 at 7:02 PM, ElliotL said:

I’m deeply touched by your thoughtful response. One of the steps I intended to follow is included in your advice, and it’s encouraging to hear that your application dried so well. I’ll give it another try.

I also have some mastic resin on hand, so I’m excited to test the effects it might bring. By the way, passing the fingerprint test in just four hours is impressive! I’ll adjust the proportions of oil and resin accordingly. My current ratio is 4 parts resin to 6 parts oil.

In your recommended 3:1 ratio, what solvent did you use to make it easier to spread, and could you share any tips regarding the proportions?

Thank you for your guidance!

The varnish I made dries to the touch in about 4 hours when exposed to UV light. It takes about 24 to 30 hours of UV to pass the fingerprint test. To be on the safe side, I gave each coat of varnish 48 hours in the UV cabinet. My UV box has four Philips 18W 24in T8 Black Light Fluorescent Tubes and is lined with aluminum flashing.

For diluting the varnish, I use turpentine. Depending on how thick your varnish is, the amount of turpentine will vary. It was recommended to me to dilute it until it is the consistency of honey. For me, that was about 3:1, or 25% turpentine. I don’t think you can go wrong with Diamond G turpentine. I used Chemfulfill branded turpentine from Amazon. I haven’t had any issues with it. It is clear, evaporates quickly, and smells nice. It did come in a plastic bottle, so I put it into smaller glass bottles for storage.

The instructions for the varnish can be found here:  https://www.ruschil.com/blog/making-violin-varnish

In those instructions, the 168 hour cook time for colored varnish is explicitly called out. My 168 hour varnish didn’t have enough color. I contacted the author of the varnish making tutorial and he mentioned that the reduction amount is important and not the time. He recommended to go for at least a 50% reduction in colophony weight and a 67% reduction would be ideal.

I will probably mess around with the varnish cooking when the weather gets warm again. I did notice that as the varnish cooks down, a skin will form on the top of the melted rosin. It will melt again when stirred in, but it seemed to me that I would need to increase the temperature as the rosin cooks down. Can’t hurt to try.

[joerobson]

I think you have received some good feedback. I would add a couple of thoughts.

What is the color of the resin prior to cooking?

Cooking such small amounts of resin makes it difficult to control temperature. What is the cooking vessel?

Long cooking resin vs long cooking varnish produces far different results.

on we go,

Joe

[jefcostello]

Tell me about my experience.
Burning rosin is to harden the varnish, not for color. Color is an added product.

 Many online recipes may only work in their region.
Cook the rosin to 15%-25%, oil:rosin > 0.66:1. This will make the varnish durable

I've seen a lot of high priced violins rot in my area. This may explain why most people now use commercial varnish instead of making their own. Because the maker doesn't know what will happen to its violin in the future

Adding mastic will make the varnish softer, and I recommend not making the recipe too complicated at first.

[JimBonzo]

It's been a long time since I cooked varnish so I can't remember every step but I seem to remember having to cook the oil until it formed a string when you dip a stick into it and allow it to drip. That's when you know it will dry. I could be wrong and there might be other ways to go about it. Cooking varnish for a long time might darken it's colour but it's not neccesary so that the varnish will dry.