Test for doneness: Coffeecake

Once again, I am learning from my own mistakes, chalking my baking failures up to “experience.”

I recently picked 15 pounds of organic blueberries, most of which I froze for use in the Winter when fresh produce is expensive and less-than-desirable quality.

But I had some sour cream in the frige that was about to expire, and rather than let it go to waste, I decided to combine the two ingredients and make a blueberry sour cream coffeecake.

I probably have a recipe for blueberry sour cream coffeecake somewhere in my stache of newspaper recipe clippings and multitude of cookbooks, but rather than sift through all of those various resources, I find it all-too-easy to go to allrecipes.com and do an ingredient search.  I love the ingredient search feature when I have ingredients I want to cook or bake with but need some fresh ideas for new ways to use them.  And using the ingredient search, I quickly found the above recipe–it was exactly what I was looking for!

I made the recipe exactly as it was written.  I always tell my baking students: the first time you make a recipe, follow it exactly so you know how it is supposed to turn out.  THEN you can make adjustments for your own preferences.

Unfortunately, though, I made one mistake: I did not bake the coffeecake fully.  This is where it gets tricky and a little confusing. . . the coffeecake passed the usual tests for doneness: 1) a wooden toothpick inserted in the center of the cake came out clean, and 2) the top sprang back when I touched it, indicating that it should have been done all the way through.  Not so.  After I allowed the coffeecake to cool completely in the pan, I inverted it onto a plate and cut a slice to test it.  The bottom of the cake was fully baked and had a nice texture, but the top of the cake was still gooey and dense–it was completely underdone.  YUCK!

I know from my previous experience working in professional bakeries that coffeecakes always take a LONG time to bake–even longer than when they appear to be done.  I should have applied that knowledge and experience to this recipe, too.

Here is where I think my tests for doneness went wrong: I baked the coffeecake in a bundt pan, which is rather deep.  But I tested the cake with a standard wooden toothpick which is not long enough to test all the way down to the bottom of the pan.  The top portion of the cake was fully baked (when inverted, that became the bottom part that was done).  But my toothpick did not reach down to the bottom that was under-done.

Once my cake had cooled completely and I discovered my mistake, I attempted to salvage it by putting it back in the oven to continue baking it.  WRONG.  That doesn’t work, either!  By the time my coffeecake had baked and cooled, the baking powder had been fully activated and would no longer react once I returned it to the oven–regardless of how long I continued to bake the coffeecake.  I think I baked the coffeecake twice as long as it suggested in the recipe, but it never got any more done on the top (the bottom of the pan).  It was a hopeless effort.

Lessons learned: 1) When baking a coffeecake in a bundt pan, use a LONG wooden skewer to test for doneness all the way down to the bottom of the pan.  Be sure it is fully done before removing it from the oven!  2) Once a cake (or other quickbread) has cooled completely, it is not possible to return it to the oven to bake it further if it is not completely done.  The leavening agent (baking powder or baking soda) will not react a second time to make the cake rise and bake fully.

I can’t wait to try this recipe again and bake it properly the next time.  The parts that were done tasted fabulous!