Chaff collector bypass, Cyclone chaff filter

[billc]

I read this post on the facebook forum a couple of years ago and was going to try to duplicate what Nicholas Smith had done, but I couldn’t find the mini cyclone unit.
I have found a source and ordered one and now, as time permits, I will be doing this mod.

Lately I have been doing 1 kilo back to back roasts of Ethiopian Naturals and the amount of chaff that is produced is huge and I have a feeling that it is affecting my roasts.
I have 2 chaff collectors and screens and can swat them out in less than a couple of minutes. I would prefer not having to stop to do even that.

Here are a few pictures from Nicholas Smiths thread on the FaceBook Aiillio Bullet forum.

He is using the Fan in the Bullet and has had to increase the fan speed by 1 or two in his recipes to accomodate for loss of air flow due to the restriction cuase by the mini cylclone.

I don’t know what the effect on the amount of smoke produced will be.

Looking up from the bottom:

[prehistoricperk]

It’s been on my to-do list for a long time as well. I bought a second chaff collector from ailio s[ecifically for this mod. I even modeled the metal section in cad. But I had the same issue as you with finding the mini cyclone.

[billc]

Here is a link to the mini Cyclone filter.

DIY Aluminum Alloy SN25T5 Dust Deputy Turbocharged Cyclone Dust Collector...

I bought one and have it sitting on the shelf waiting for time to me to fab up the tubing into the extra chaff catcher.

Edit:
I have been contemplating trying to use a Dustopper, available from Home Depot. I will still have to fab up the tubes that will go into the chaff collector, but I should be able to just run hoses to the Dustopper without going through tubing reductions. One of the things that is unknown to me is if there will be enough air flow for the cylclonic action to work and drop the chaff with the Dustopper. I already have a Dustopper and it works great with my woodworking saws but I am using a 5 hp shop vac for the vacuum side of it. With the Bullet, I will only be able ot use the Fan and wouldn’t want to go more than F4 for most of the roast.

One of the other things that I am trying to figure out is how to add the plumbing to the Bullet and still be able to easily access the fan for cleaning.

[billc]

Amazon is carrrying the mini-cyclone dust filter again, and at a better price:

[billc]

After I ran three 1000g roasts of Natural beans, the chaff collector is nearly full of chaff. Here is a photo of the chaff collector after the three roasts:

This is why I want to use the mini-cyclone filter.

I ran some very crude tests with the mini-cyclone filter using duct tape and hose bits that I had on hand.
I ran the test with no heat and just changing the Fan speed.

The results are promising.

This was powered by using the Aillio Fan at F5.

No chaff or dust escaped the mini-cylcone as long as the chaff was below the bottom of the nini-cyclone.

When the chaff got up to the bottom of the mini-cyclone, then it started spewing chaff out of the Bullet exhaust.

I tried other settings and can probably get by with a lower Fan speed. I am not too worried about that. I think that I can adjust my profiles to accomodate, and maybe reduce the batch size slightly.

Now I have to figure out how to adapt the chaff collector.

Note to self: Open Rabbit hole, dive in head first.

[duckduckco]

that’s awesome

[billc]

Sweet Marias does carry this.

I bought one from them in June of 2022.

The complete assembly.

[billc]

I have the adapters modeled up and am getting ready to 3d print the adaptor that attaches to the trapezoid opening.

I will use tubing to go from the two tubes that are shown to the mini cyclone.

The only modifications will be to the extra chaff collector. There will be no modifications to the main part of the Bullet.

I will be able to remove that hackenstein chaff collector and put back the OEM chaff collector in seconds if I need to.

Things I don’t know yet:

If the 3d printed funnel adaptor will stand up to the heat.

If it doesn’t, I will print a 3d mold of the funnel adapter and then use a high temp epoxy to fill the cavity of the mold.

[blacklabs]

@billc

hackenstein chaff collector

[billc]

i just ran my first test with some Natural beans from Columbia. I wanted to use a bean with high chaff output. I have roasted these beans many times before.

Here is a photo of the first real prototype:

I was pretty pleased with how everything fit together and thought that this was going to be a good solution.

It is not a good solution.

I ran a recipe that I had and added higher fan speeds as the roast progressed.
I was not expecting the results that I got.
Preheat was normal.
Charged at 375 IBTS.
I noticed that there was a lot of condensation in the tube and chaff collection jar even before I got to yellowing stage.

When I got to yellowing stage (331F), I was getting a lot of steam coming out of the fan exhaust.

I don’t remember ever seeing steam come out of my Bullet before. Smoke yes, but steam, No.

I kept bumping up the Fan speed and had my microphone in the tryer port so that I could hear fist crack.

First crack started about where I normally notice it: 393F
first crack went into rolling first crack in 30 seconds. Then second crack started and was louder than I have ever heard it before. At first, I thought that I was still in first crack. But when I checked the IBTS, I was already at 430F and a very robust roling second crack.
I dropped dark oily beans at 445F.

I tasted one, and it was not as bad as I was expecting it would be.

The funnel that I 3d printed didn’t hold up to the heat well and was starting to distort and soften.

Here is a photo of condensation in the tubing. Also you can see the jar lid that I printed that just clips in place when you press it down onto the jar.

I am thinking that maybe the steam excelerated the roasting process.
I think that I will patent this as a new process: instead of dry roasting, I will be steam roasting. (I am just kidding)

I will try another roast and start with the fan speed higher, But, I am not very hopeful at this point in time.

Edit: Very little chaff was collected in the chaff collector jar. None of the chaff escaped out of the fan exhaust.

[billc]

Thanks rtuttlejr,

My modification did not reduce the airflow to the back fan.

If anything, there is less restriction without the plastic chaff cover installed.

Regarding the condensation: I am leaning towards a lack of airflow and a venturi effect because of the reduction in the “duct” size forcing the water molecues to squeeze together and bond with each other and condense. The mini cyclone only has a 25mm (1 inch) inlet and outlet. Similar to a contrail made by a jet engine.

I’ve heard of condensation issues with the Bullet when people put an adapter on the exhaust that is sealed to the exhaust housing.

[billc]

Before seeing the condensation in this setup, I had never really paid attention to the amount of water we are removing from the beans.

With weight loss of 12 - 20 %, that is a significant amount of water.
In this case: 132g of weight loss = .over half a cup of water.

[braca19452f9m]

You know, my guess is that the fan is simply not strong enough to cover the distance of the tube and still be efficient. If the fan were placed inside the cyclone collector, I think the story would be different. But that would bring a whole other set of problems. I don’t have much advice, except trying to place it on its side, so the tube going into the fan will be much shorter and not lose so much airflow covering all that intricate itinerary. Nothing else comes to mind. Good luck.

[jacob]

I agree with @braca19452f9m I think the pressure drop from the tubes and the cyclone is too high to be overcome by the fan (which is a forward curved fan, not good for high pressures). There is a large pressure drop in a cyclone, so you might need more than twice the power of what the current setup can supply.

[billc]

It was an interesting experiment.

I have put this on the back shelf in my archives, but I doubt if I will ever come back to it.

I will post the new process when I get further along with it.

I am already working on a new solution and have run some tests. So far, It looks like this new process will work. I just need to run a couple of 1000g roasts consecutively to see if the 3d printed adapters can stand up to the heat.

The new process doesn’t require any permanent modifications to the Bullet with the exception of removing every other blade of the exhaust fan. Which I have already done and re-calibrated the fan. (After buying a back up fan from Sweet Marias)