The Brazil Ipanema tree dried that was questioned lost 2.0g with two 20.0g samples so about 10%. So it was a bit on the drier end but nothing extreme.

another_jim wrote:It would be nice to have a cheap/quick/reliable way of measuring the moisture content of beans. I would think the drying phase would be proportional to the moisture content.

I have a suggestion that I am trying with some apparent success: when a greens shipment arrives i drop one of those RH/temp panel meters into the beans and seal them up in my 16C constant temp wine cellar. After 1-2 days stabilisation of temps & RH the meter reading for relative humidity gives a rough guide to moisture content - enough to adjust phase 1?

Just an idea given that green beans will take up in humid environments and lose in dry... is it worth developing? No doubt the actual number is bean and temperature specific.

Eric

Hello. 1st post to the forum!

I have also been interested in determining the moisture content of the beans I roast, as well as their density. In my uneducated guessing, I had theorized that the bean density would effect the roast more than the moisture content. Perhaps I am wrong about this? I do know that my density measurements are flawed somewhat by the moisture content, and the idea of drying the beans first seems like a good one to me.

For density, I start by weighing a small amount of green beans on my beam scale, and can get readings to about 5% of a gram. I then measure the bean's volume by fluid displacement in a graduated cylinder, which is fairly accurate to about [EDIT]25% of a milliliter. I then divide the weight by the volume to get density.

I am working off of a new Hot Top - P, and have only a handful of roasts under my belt with a drum roaster (it was iRoar before this). I hope that my density readings will help me fine-tune my roasts for different coffees.

Great first post Eric, welcome!

Thank you, sir. I look forward to learning and sharing with all of you here.

-= Eric

I looked up drying and density at origin in Sivetz's Coffee Technology, which is 1970s paleotech.

For moisture, moisture meters are calbrated against a ground sample weighed before and after a 225F 24 hour bake. Weigh immediately out of the oven to prevent moisture absorption. To get to the nearest 0.1 percent, you need a 0.1 gram scale and a 100 gram sample.

For density, he just took a big can, measured the volume (fill with water, weigh the water), dry the can, fill with beans, and shake until you can get no more in. Not quite a density, but a good relative scale.

Since my earlier post and test in this thread I picked up a new Digiweigh 100g x .01g scale for $25.00 shipped. This allows me to keep my sample size down with accuracy. I still use the espresso baskets and made a forked handle for taking them out of the oven. I've been course dicing them in the whirling blade. 250f 3-4hrs. With summer coming I'm thinking of making a mini oven to hold 4 baskets and heat to 225f. Jim, I'd say Sivetz 225f is about optimum. Much higher and they seem to start cooking a little. I set a cold basket upside down on the scale tray(has stainless tray) and set a sample out of the oven on top of it.

Coffeengineer wrote:I have a suggestion that I am trying with some apparent success: when a greens shipment arrives i drop one of those RH/temp panel meters into the beans and seal them up in my 16C constant temp wine cellar. After 1-2 days stabilisation of temps & RH the meter reading for relative humidity gives a rough guide to moisture content - enough to adjust phase 1?

Just an idea given that green beans will take up in humid environments and lose in dry... is it worth developing? No doubt the actual number is bean and temperature specific.

Quoting myself, sad. No one took this up... definitely the fastest and a quite accurate method of measuring moisture content. What one needs is the sorption isotherm for green coffee beans in question for the measurement temp. Don't know how they change with temp. Measurement only takes 5 mins and requires no heating... I found a sorption isotherm for green coffee at 20C (variety not specified):

The x axis is equilibrium relative humidity (ERH).

Does anyone have access to one of these machines? It takes 24hrs to generate an isotherm that can then be used for rapid moisture content readings.

Now that I have sourced greens from different locations I am measuring ERH in the range 48% to 72%, certainly gives one a hint as to how long the drying phase should be.

Eric
ed: ERH (ratio of water mass in air) and water activity (ratio of water vapour pressures) are technically slightly different.

That's pretty awesome. Do you have a precise version of the graph?

Sorry Jim don't have details - the graph was quoted without a source... hence need access to a machine! It is a tedious exercise to generate the graph manually. A few sample points might be useful. Then again, for our use we don't know exactly how water content affects ramp up so might as well work with ERH directly.

Corollary of course is that green beans are constantly interacting with the environment, picking up and losing water depending upon the local RH, so one should keep greens in sealed containers if you don't want your profile to drift.

Eric

There's a bit of luck here: the moisture content of fresh coffee should be between 9% and 13%, and this is where the graph shows the greatest changes in RH for moisture content.

So I think an RH meter may yield a simple decision rule: Use your standard drying profile between 60 and 75, shorten it if the humidity is lower, lengthen it if it's higher.

Jim I think the number might be a little lower, here is a doc that looks like it was copied from somewhere else that has a sorption isotherm for parchment arabica at 25C... (taken from Stirling's Kenya Coffee, 1974). What was a bit more concerning was the plot of degradation in cupping quality vs moisture content. How can one maintain cupping objectivity over 12 months?

Here is a reference that states ERH and aw (water activity) are numerically the same. Also shows that isotherms move to the right as temp increases (barley @ 20 & 30C).

I find high quality recent shipments cluster around 61% (59-63) with some outliers possibly representing problems in the supply chain. Water decaf is different, giving a higher reading probably due to the processing. Perhaps you could run through your collection and compare with your roasting notes and summarise.

The RH meter I am using is just a cheapie from fleabay that i picked up on spec to test the theory, not having any absolute numbers to relate to. Looks like it is worth using at least as incoming QA, and also for drying hints. My experience is that over drying is usually more of a problem than under (= flat coffee and extended rest times).

Eric

Fullsack wrote:The other problem created by the high ET charge and fast drying phase ramp was bean color often seemed to go from a greenish yellow to a light brown without ever hitting bright yellow.

I'm just starting roasting with a Toastmaster popper, no thermometer, and I don't generally see bright yellow beans. Looking at the following suggestion for the drying phase from another_jim:

another_jim wrote:Here's how to get some very basic profiling on a completely unmodified popcorn popper:

Drying phase: Run the roaster one minute, then turn it off for two minutes, then restart the roast. Increase or decrease the turnoff time to change your drying time.

I ran my popper for a minute on 100g of beans (Supremo) and turned off the popper, continuing to stir the beans with a wooden spoon for two minutes. I saw some yellowish tint, but no real yellow, so I turned on the popper for 30 seconds then off for a minute and 30 seconds, stirring, and saw some yellow, but not very bright. Another 30 seconds on and then 30 seconds off made the beans even yellower.

Am I on the right track do you think? It took me 6 minutes to get yellow beans -- is that too long for the drying phase?

Buy a thermocouple thermometer, they only costaround $10 to $20.

Mount the TC so it's just above the bean mass at the end of the roast. Run the roaster so it gets to 300F. Turn it off a few minutes, take break, don't stir. Restart the roast. Some of the beans should start going yellowish when you reach 300F the second time, and most should have lost their green color. Don't go for completely yellow.

Forgive my naivete as an armchair roaster, but why would one be content with rough-and-ready bean dryness metrics established before they're roasted? Why wouldn't one be measuring instead the moisture inside the roaster itself with a high-temperature thermo hygrometer?

another_jim wrote:Buy a thermocouple thermometer, they only costaround $10 to $20.

Thanks and I'll try that. The particular device you referred to would have cost $35 with shipping to Hawaii, so I ordered this for $12 (which perhaps displays only centigrade). Hope it works.

another_jim wrote:Mount the TC so it's just above the bean mass at the end of the roast. Run the roaster so it gets to 300F. Turn it off a few minutes, take break, don't stir. Restart the roast. Some of the beans should start going yellowish when you reach 300F the second time, and most should have lost their green color. Don't go for completely yellow.

My first try at this didn't work to produce any yellow beans. It took 2:15 to get up to 300F, by which time the beans were already slightly tan without having gone through any yellow phase. After an off period until elapsed time was 5:00, I continued up to 300F again, seeing only more tan. (First crack came at 388F, second crack at 10:30 elapsed time, and I continued to 12:00 to get a Vienna roast. Batch size 100g green, 83g roasted. Guatemala SHB EP Estate Antigua "Los Volcanes".)

Hi Greg

I'm not sure why you are going doing this for a Vienna roast. Does the roast taste grassy or chlorinated? The trick of turning off the roaster is aimed at removing grassiness from lighter roasts.

On color: The beans start out green, They go yellow to light tan in the low 300s, (I'm not talking saturation, just tint and value), and an even tan color in the mid 300s. Then they get uneven during the first crack, but should come out of it an even brown color. For a roast stopped before the second crack starts, or just at the very first pops, it is best that the drying phase is arranged so the beans go through this color progression at the stated temperatures.

The problem for Vienna roasts is usually ashiness. For these darker roasts, unregulated air roasters do quite well without any special tweaks, since carrying excess moisture into the first crack can reduce ashiness. For such roasts, don't do any stalling; and let the beans go throught their color progression about 30F higher than stated. If the roasts now taste ashy, reduce your load, so that the air temperature blowing in stays a little lower.; if the roasts are too slow finishing, increase your load to raise the supply air temperature (by slowing down the airspeed).

Most of the people posting on profiles are roasting light. If they don't talk about roast depth, they are usually dealing with the tasting issues one encounters in City and occasional Full City roasts.

another_jim wrote:I'm not sure why you are going doing this for a Vienna roast. Does the roast taste grassy or chlorinated?

No, it's just not tasting that great. It's not obnoxious, but it's not very interesting. I'm casting around looking for improvements. Are you saying here that dark roasts don't profit from a controlled drying period? Or that someone who likes dark roasts may as well not bother trying to improve.