Preliminary results are in

Soapmaking Forum

Help Support Soapmaking Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Apr 19, 2019
Reaction score
After 7 months of making soap, I am approaching my 100th batch and have used dozens of different recipes. I’ve made some of the recipes over and over and some I abandoned after the first batch. For each batch I also recorded a qualitative assessment of how the batter behaves with respect to trace. For example, is it slow to trace, like lard, or is it threatening to become soap on a stick (fast) like almost every recipe I’ve made to date with 40% or more palm. It’s all in a spreadsheet and after looking at it a bit, a relationship emerges. This relationship is for my recipes and my methods. I’m definitely not trying to say that it works for anyone else :). I did not include salt bars, 100% OO, or recipes where I added something really strange, like madder in ammonia.

For additional context, during these early months of soap making, I have mostly stayed away from accelerating FOs, but have used all kinds of EOs, food additives, micas, natural colorants and FOs that don’t accelerate trace. The temperature I use depends on the recipe, e.g. cool for lard, 110F or so for palm-rich and even a little warmer for my newest tests with GW 415 soy wax. I have mostly used 33% to 37% lye concentration. I also use a limited range of oils and butters. The oils in my high oleic group are sweet almond, olive (virgin, not pomace), HO sunflower, and HO safflower. I also use avocado oil, but left it out because it is not following the same pattern as the other high oleics. I included lard in with the high oleics because it has a similar effect on trace in my recipes. My usage range for castor oil is typically 0-6% and for CO 20% or less. I highlighted two batches of goat milk soap that have CO at 25% because they moved a little faster than expected. The few batches I’ve made to date with GW 415 are behaving like high oleic recipes.

My main motivation for looking at the data at this point is to try to understand why I can’t seem to come up with a palm-based recipe that I really like. I *think* the answer is that I’m not using enough of one of the high oleic oils listed above. I threw avocado oil out of the high oleic group when a closer look at my data revealed something unexpected. It contains 20% palmitic according to the SMF calculator. The darker (uppermost) green circles on the graph are recipes made with hard fats > 25% and about 45% of an oil from my high oleic group. The lighter green circles and the yellow circles are recipes made with less than about 40% of the oils in my high oleic group, and typically with avocado oil or more recently RBO. For now at least, avocado oil seems to be less effective at slowing trace in my recipes compared with sweet almond, virgin olive, HO sunflower and HO safflower. The black lines in the graph are hypothetical, e.g. I may not be able to achieve a “slow” recipe using 50% palm and the HOs I typically use. If I decide to try, I will use HO sunflower, which is the richest in oleic acid of the oils I use.

ETA: I put the black lines in super late last night. I’m not sure that they’re helpful. At the least, I may need to move them a bit. If you’re reading this, just ignore them for now. I would really love to know if anyone else feels there is a general relationship between speed of trace and the proportion of the high oleic oils I listed above in the recipe.


It would probably be better to look at the relationship based on FA percentages, but it would be a lot more work because I don’t have that info in my spreadsheet. Maybe that’s the next step.

ETA: relationship with FAs now posted below .
Last edited:
I will almost guarantee you will never get a slow trace using 50% palm with any HO oil. I use 45% palm in my vegan recipe with an HO oil including canola, safflower, or sunflower, whichever I have, and slow trace will not happen with or without castor oil.
I will almost guarantee you will never get a slow trace using 50% palm with any HO oil. I use 45% palm in my vegan recipe with an HO oil including canola, safflower, or sunflower, whichever I have, and slow trace will not happen with or without castor oil.
Thanks Carolyn. That’s one less thing to try.
Here are my results based on an analysis of the FA composition of the soaps I’ve made. I get the best relationship for predicting how fast the recipe will move through trace phases for me when I plot total stearic & palmitic FAs (%) versus total oleic FA (%). I did not include the lard soap recipes for this graph because they all fall into my “slow category” regardless of the amount of lard I used, up to 100%. Maybe that’s because my lard-based recipes have not included any butters, palm or tallow. One last thing is that there are over 50 batches of soap represented. You don’t see all of those points individually because multiple batches plot in the same place. I have not had much variability among batches of the same recipe when using the methods I described above.

Last edited:

Latest posts