|
Post by kiernan on Mar 2, 2021 10:20:16 GMT -5
Hello. Does the type of cracker significantly affect the efficiency of the winnowing? For example, does the consistency and size of the cracked husk change how well you can winnow? I am pretty new to this, so I am using a Champion Juicer to crack. I find I am getting a lot of variety in the size of the husk after cracking (ie. I am getting everything from very large pieces of husk down to too much powdering of the nibs/husks). My homemade winnower seems to work well (though I expect I have to optimize a bit more), but I usually have to do several passes to get a decent winnowing (or I have to increase the pressure and then lose more nib or do a few more passes of the husk to try to extract more nib from the husk).
My assumption is that the variability in the size of the husk makes it difficult to adjust the suction to get it just right to improve the winnowing. Thus I would think that if I had a better / more consistent cracker, I would get less variability and thus could better optimize the suction to get a better result. Am I right on the right track here? If so, is the different sufficiently large that it is really worthwhile investing in a better cracker, even at this early stage for me?
|
|
|
Post by Chip on Mar 2, 2021 10:41:53 GMT -5
Good Morning kiernan, The winnowing portion has more to do with dialing in the winnower than the sizes of the husk fragments. Once you have it tuned in, you will get a good separation and be able to moderately winnow without having to tweak too much more. You also have a lot of variables: size of vacuum you are using; is the vacuum filter plugged; as the filter gets used the effectiveness decreases; rate of pour into the winnower. I have found you will always have a little left over husk that is easiliy removable by the "hair dryer" method. Put it in a large bowl and then use a hair dryer to blow out the left over husk. Hope this helps.
|
|
|
Post by Ben on Mar 2, 2021 14:43:14 GMT -5
I have to disagree with Chip here. The type of cracker can have a significant impact on the efficiency of winnowing. The smaller cracking options out there (Champion, Crankandstein, etc.) all have issues with nib & shell size consistency, too much dust/fines, poor nib/shell separation, etc. Ideally, you want the nib & the husk to be comparably sized with complete separation and minimal dust/fines. Different crackers are more or less successful at these goals. One day, I plan to upgrade to an impact cracker as a few maker friends that I trust speak very highly of them. They apparently get more or less perfect nib/shell separation with good size and minimal dust. Regarding size, many makers sort cracked beans based on size using a series of screens before winnowing to maximize yield. By having consistent size throughout a winnowing pass, you are able to fine-tune the winnower for that size. You can add a 1/4" screen to the output of the champion to get a more consistently sized crack. I've done this and it definitely results in a cleaner winnow in a single pass. It does slow things down a bit and I've found that the yield is somewhat decreased (comparable to the decrease of multiple passes).
|
|
|
Post by Thomas on Mar 2, 2021 14:45:58 GMT -5
The cracker will affect the efficiency of a single flow winnower. By single flow I mean a winnower that does not classify the cracked nibs in any way. Higher end winnower will classify or seperate the cracked beans by size and also have variable air flow or suction for each classified size. No cracker is going to perfectly crack the beans into one size. Some may be better than others. I've used the Champion Juicer, the Crankandstein, and now use one made by CocoaTown. They all work but I ended up with the CocoaTown because I needed a commercial product for my shop. John, the alchemist, has an article about using the champion and it producing powder. I've tried to search for it but haven't found it yet. He compared it to the Crankandstein and found that it didn't really waste nibs. If I remember it correctly. I built my original winnower from here: www.realseeds.co.uk/images/seed%20cleaner2a.jpg. I also use a variable speed vacuum to control the airflow. Multiple passes are easy and you can see the process along with the waste. No worry about the vacuum filter since virtually no waste ever gets there. I'm not sure what type of winnower you made but thought I would mention this since most people seem to make tube type winnowers. I personally never liked these type. Just my thoughts. - Thomas UPDATE: I found the bean cracking comparison between the champion juicer and the crankandstein that I mentioned. chocolatealchemy.com/blog/2006/07/30/crankandstein-vs-champion?rq=champion%20juicer
|
|
|
Post by Chip on Mar 2, 2021 16:41:58 GMT -5
Ben is our "guru" here, so always defer to him. But on a small scale, I use the same equipment (Champion and then John's winnower) and can achieve fairly consistent results most times. Of course, I am a "tinkerer" at heart and never one to need set it and forget it. I am also a very small maker of chocolates compared to Ben.
|
|
|
Post by Sebastian on Mar 2, 2021 20:24:14 GMT -5
I'd suggest that your bean moisture at time of cracking, and bean size uniformity - are two very important variables you will want to be paying attention to. Understand how to control those, and the type of winnower decreases in importance.
|
|
|
Post by kiernan on Mar 3, 2021 10:04:43 GMT -5
Thanks all. This is really helpful. Sebastian: can you expand on bean moisture? Do you mean just ensuring that your roast should shed a certain percentage of water (based on weighing pre- & post-roast, I seem to generally be dropping between 4-6% of water and other liquids) and that the beans should remain at certain dryness by the time they are cracked? Or is there something more specific than that? And then for bean size uniformity, do you mean the cracked bean size uniformity (as Ben and thomas were mentioning)? Otherwise, could you explain how bean size uniformity affects the winnowing? Is it just that uniform bean size improves the consistency of the roast and thus presumably of the cracking, and the cracking itself will be more uniform because of the consistent size, thus giving a consistent cracked size?
|
|
|
Post by Thomas on Mar 3, 2021 10:22:56 GMT -5
Bean size uniformity should have been nib size uniformity. Winnowing is about the weight of the the nibs and also how well the shells have been separated after cracking. The heavy nibs fall through the winnower and the lighter shells and even smaller light weight nibs are blown or sucked into the waste pile. The more uniform the nib size, the more efficient the winnowing process. Hope this clears up any confusion.
-Thomas
|
|
|
Post by kiernan on Mar 5, 2021 10:42:40 GMT -5
Yes. That's what I thought. Thanks thomas. Any idea about what Sebastian meant with respect to moisture at time of cracking? Is it just that the beans have to remain sufficiently dry and thus brittle to crack and winnow properly? or is there an actual goldilocks level of moisture for this?
|
|
|
Post by Sebastian on Mar 5, 2021 16:22:42 GMT -5
So - ultimately what you're trying to do is remove your shell from your nib - but nib moisture will cause the starches in the shell to act very much like a glue, making the shell not want to release. Alternatively, if you dry your bean out too much - you're going to simply turn it into dust when you crack it - so your losses will be inordinately high. Is there a goldilocks range? Sure - but the kicker is it's different for each winnower - as they all use different types of technology - so any numbers i give you may or may not work optimally for your specific piece of equipment. Keep good records, test your bean moisture at time of winnowing, record your settings and correlate them to your results - taht's the real answer. Generally speaking, i'd encourage you to have a bean moisture in the 1-2% range, no higher than 2%.
As far as bean size goes - what your winnower wants to 'eat' are uniform beans. The closer they are in bean size and shape, the more uniformly your winnower will perform. If you have a wide range of thin beans, fat plump beans, and itty bitty tiny beans - your winnower's crusher settings can't really be optimized to maximize efficiency for each of those - so you'll get a mediocre performance out of it at best. If you can't control your bean raw material stream from your vendor - consider doing a post - roast sizing to create large, medium, small 'streams' of product that each get batched through your winnower, with process settings that are optimal for them.
If you've got flat beans mixed in with your good beans - they may or may not get cracked - depending on their orientation when going through the winnower. If the pass through with the thin side facing the crusher, they'll likely jsut fall through uncrushed. If they happen to present themselves to the crusher in the other orientation, with the flat side perpendicular to the crusher - then they'll get crushed. Remove that 'chance' if you can by controlling your raw material stream, would be my advice.
|
|
|
Post by Thomas on Mar 5, 2021 18:32:27 GMT -5
|
|
|
Post by Sebastian on Mar 5, 2021 19:55:07 GMT -5
A good one is going to run you a few thousand dollars at minimum - the ones you linked to are more for bulk grain moisture analysis, not precision ingredient analysis Tell you what, lets work up a test where you send some samples to me vacuum sealed (whole beans, roasted, vac sealed right before you'd be putting them into the winnower) - and i'll test them in exchange for whatever delicious chocolately goodness you choose to send along with the samples?
|
|
|
Post by lilypa on Mar 9, 2021 13:28:11 GMT -5
Hey All, It's been a while since I've posted here, but I've been making choco at home the entire time regardless of my lack of presence here. I have some ideas that may help kiernan and Thomas. Kiernan: I've used both the Champion (with and without 1/4" screen) and the Crankandstein. For my setup, I've found the Champion without the screen and a few other things, which I'll talk about later, to give me the best recovery rate. I did a test years ago and found the Cranksandstein to be slower and more inconsistent with separating husk from the nibs. I found that using the 1/4" screen with the Champion to lead to more powder thus increasing my nib losses. In general, my nib recovery has ranged between 75 - 80%, with 78-79% being the average. So, what may help? I use three sieves produced by Gilson (used for classifying rock materials) #4, #6, and #10 to "classify" (organize by size) my cracked nibs/husk prior to feeding into my winnower. www.globalgilson.com/test-sieves I specifically use full-height 12" ASTM Test sieves as they fit perfectly over a $1.99 Home Depot or Lowe's bucket. This (classifying) has really helped get my recovery up and more consistent. With the three sieves, I end up feeding four different sizes of cracked beans into my winnower. With my winnower I use a piece of wood as a slide gate to vary the amount of suction depending upon the size of nib/husk being fed into the winnower. With the #10 sieve (finest size) I've found you can either just throw those out if you want to save time or winnow them. I always winnow them as I usually obtain about 4-6% of my nibs in this size. However, you will get most of the cocoa bean radicles in this size which doesn't contribute to flavor and will make your chocolate less smooth. Thomas: A friend of mine mentioned to me that you can do a quick and dirty moisture test of your unroasted beans by cracking a few by hand (say 25 - 50 g; basically make sure it's a KNOWN quantity) and then bake them at I think it was 300 F for 3 hours. (It may have been 270 F for 3 hours, but somewhere in that range.) You're basically going to burn them without creating a billowing cloud of smoke in your oven. Then weigh/mass them immediately afterward to calculate the difference in mass. That'll give you a general idea of pre-roast moisture content. Assuming you store them in an airtight container between this test and roasting you can then calculate your post-roast moisture content with the difference in bean mass pre- and post-roast. If you store your post-roast beans in an unsealed manner that can result in them absorbing room moisture, you can always do the burn-test again to get your post-roast/pre-winnowing moisture content. Hopefully that could help. Remember, it's quick and dirty, but can be effective. Lastly, to store my raw beans, my roasted beans, and my cracked beans I use a few different sized zip-seal buckets from Uline. These may be useful to you all as well: www.uline.com/BL_8173/Screw-Top-Pails?keywords=Buckets+With+Lids I guess they call them Screw-Top pails. Cheers, Dave
|
|
|
Post by kiernan on Mar 9, 2021 16:31:34 GMT -5
thanks all. This has been really helpful.
|
|