We're getting ready to sign the lease (we hope) in the next week or so for an 1860 square foot space in a tiny little business/industrial park in the San Francisco Bay Area that we hope to transform over the course of the next couple of years into a successful fine chocolate (bean-to-bar) and artisan confection business.
Clay suggested the idea of starting a discussion thread here on Start-up Central as a way of sharing what we learn along the way with others either already involved in their own start-up or still contemplating jumping in. We really liked the idea -- it fits well with our commitment to always 'paying it forward' in life, and with the idea of one hand up, one hand down (one hand reaching up to accept the help/wisdom of others who have something to offer/share and one hand reaching down offering help/wisdom to those with whom we have somethign to offer/share). We'll share what goes right, we'll share what goes wrong -- open source, nothing held back or edited. We welcome your thoughts, questions and suggestions along the way.
Snapshot of where we are today:
- Custom molds are paid for (~$8k), but on hold until our trademark registration is approved (wouldn't be much fun to have molds that we can't use...). Hope to be cleared to move forward by January.
- $1400 trademark search with Thomson Compumark completed. The attorney has reviewed it and is now in the process of filing the registrations (another $1k +/-)
- We're a newly formed S-corp. (Total legal fees to date, including the above ~$4.5k)
- We've designed our packaging, but won't order it until our trademark clears.
- Leslie is going to be full-time, I'll be part-time (mainly on weekends) since I have a day job that (a) I like, and (b) will let us support/sustain the chocolate business through its first couple of years. We'll hire help as soon as the work load dictates.
- We subscribe to the Lean Start-up approach -- an iterative build, measure, learn process; pivot (change direction) when needed, etc. and to the idea (as articulated many times by Brad Churchill in his posts on this site) that making a profit isn't a 'nice-to-have', it is a 'MUST have'.
- Oh yeah, we make a pretty decent bean-to-bar chocolate. At least WE think so. And so do some (possibly biased) friends, family and co-workers. But we all think our own kids are cute, so just to be sure we're not kidding ourselves about where we really are, we're sending bars out this week for formal 3rd party evaluation (~$1k)
- We plan to spend a good part of the first year perfecting our systems. We're not going to scale to any kind of volume until we've worked the kinks out and are capable of delivering a consistently excellent (and profitable) experience to our customers.
- For manufacturing flexibility and resiliancy we plan on using multiple small scale pieces of equipment rather than single larger volume pieces.
- We know that we don't know what we don't know. We plan to contract with Clay for consulting services, both on the plant layout and on the business side.
- We're utilizing SCORE resources to learn more about the accouting, insurance, legal, and sales and marketing sides of the business, and to make contacts.
There's more, but you get the general idea...
Next major milestone for us is seeing if the landlord will accept our proposed lease addendum and mark-ups. That will be our next post...
David and Leslie
The principle is the same as a PVC downdraft winnower, but using suction (think cyclone) rather than blowing air. One thing we're looking into is addressing the laminar airflow pattern around the walls of the pipe that can lead to reduced yield (can't see the effects inside opaque plastic), and we're still working on that. Plus - it's clear, so it's fun to watch and much easier to tune.
A separate part of the project (I am working with someone else on this) is a compact motor-driven auger feeder that can be used to feed a cracker with beans, feed a winnower (any winnower, not just this one) with cracked nib, and feed a grinder with cleaned nibs. Capacity up to about 5 gallons.
Give us time. It won't be long.
Thanks Clay. It all sounds very interesting! I'm especially interested in the feeder.
Yes on the feeder, I know. Feeding is a real hassle and it would be nice to have one feeder that is a multi-tasker.
The laminar flow issue along the tube walls was something that David noticed once he could see the process of separation inside the glass tube. What happens is that there is smooth flow/curtain of air that forms along the walls of the tube - a lack of turbulent airflow. If a piece of shell, or a nib with a piece of shell, gets into this calm space it can fall all the way to the bottom unhindered. By eliminating the calm air space at the edges of the tube you make the entire column of air turbulent - not just the area immediately above and below the "y" where the nib is introduced into the system - and this should improve separation significantly.
Our hearts are beating a little faster now every time we go over to the shop to work on one thing or another -- it's been a long road, but we're getting close enough to almost smell chocolate. This week we ran the power for the oven and turned it on for the first time -- still have to read the manual, but at least we know the lights go on and it makes some noise(!). The contractor is going to give us a quote for running a water line to it now so we can use the steam injection.
We ordered version 2 of our glass y-tube winnower last week -- should have it in hand by the beginning of April. In the meantime we're installing version 1 -- hoping to have it up and running this coming weekend. We'll post video shortly thereafter.
The next big project is the bean cleaning table -- we have the table base, but nee to order the top and install the mesh. We're still figuring out exactly how we're going to enclose the area under the mesh and run the exhaust, but those are fun little challenges to have to solve.
After that it's really just moving a couple more tables in, setting up the melanguer and the tempering machine, putting up some shelves and the like. A couple of details like shades for the windows, a good thorough cleaning, and then....turning it all on for the first time.
Hope to have some good news on the winnower next time we post...
David and Leslie
It is exciting to see the end of the tunnel coming and very cool.
We have a few adjustments to make and a couple of improvements to incorporate on version 2, but we're happy to share what we have so far with version 1. In the YouTube video (https://www.youtube.com/watch?v=v3E8i3kySv8) you can see the overall set-up:
- Feed hopper
- Feed auger (KitchenAid mixer with meat grinder attachment -- sans the metal grinding plates)
- Tubing to direct the auger output to the Y-tube feed tube
- Glass Y-tube winnower (version 1)
- Nib collection bucket
- Cone bottom husk collection tank
- Dust Deputy (to capture most of the fine dust before it gets to the vacuum
- Shop Vac
Version 2 of the Y-tube will be longer, dispensing with the plastic tube extension we found we needed on version 1 (in order to get a clean separation). It will also have an integral feed funnel, as well as a couple of ribs at the top to make mounting easier and more secure.
We still have to add a permanent (hinged) flap on the bottom of the cone bottom tank and cut the opening a little larger in order to get a faster/cleaner removal when we're done running).
We started off with a 6.5 peak HP shop vac that we already had, but found out that it is WAY too much for this system. We went down and picked up a 1.75 peak HP vac for $29, hooked it up and it's working great. I wouldn't go any lower than that, though -- we have to run the air adjustment valve almost fully closed to get the separation we're looking for.
We get the cleanest separation at a speed of 1-2 on the KitchenAid. I don't have numbers yet on the throughput, but will measure that sometime soon and will share it. For us it's not so much about the throughput as it is about being able to hit the start switch and go do something else. It can winnow a LOT more than we need right now. Down the road we can look at increasing the capacity by going to a larger diameter winnower with a bigger vacuum.
Other improvements we're looking at:
- Replacing the KitchenAid feeder with a dedicated system (variable speed drive motor and auger) that Clay is working on
- Integration this set-up with the grinder (eliminates one step)
- Possibly adding some ribs near the exit of the feed tube to even out the flow of ground nibs into the "Y"
It's 100% open-source, so any ideas you have for making it better and easy/easier for others to build are more than welcome. We're working with Clay to make the Y-tube available through this forum, but let us prove out version 2 (it should be in our hands by the end of March) before we call it 'production ready'.
David and Leslie
Very nice David. I will definitely be interested.
This looks great, but I'm not sure I see how the Y-tube winnower part is mechanically any different from a pvc winnower that uses a wye at the point where the cracked beans enter the airflow.
It looks like it should work more or less like a winnower design I used for a while:
That one doesn't have the automatic feeder (using the mixer with meat grinder attachment is genius, by the way) and airflow control, but the mechanics of separation appear to be the same.
Am I missing something?
The physics is the same. What's different is visibility. Because you can see the separation process it helps in tuning.
Also - this is V1. There are some improvements in V2 that should make the separation efficiency greater.
Another thing to think of is that the visibility is a big draw for customers.
Thanks Clay. I agree the visibility is very cool. I just wanted to make sure I wasn't missing something with regards to how the actual separation was happening.
How's the feeder coming along?
Slowly. The engineering resources are in Venezuela and things are a little messed up there at the moment.
Well, last week the contractor ran the 240 line and installed the outlet. He installed the pigtail on our shiny new convection oven (it doesn't come with one). This weekend we plugged it in for the first time -- the lights come on, there's a reassuring hum -- it does everything it's supposed to do. Except get hot. The service company is coming out today -- stay tuned...