How do you define the start of first crack? The very first pop? When they first appear with less than x seconds in between?

Likewise for the end of first crack and the beginning of the second... What guidelines do you use to determine the transitions?

By bean temp, but if that's not available, maximum activity.

First pop is "first of ..." or "start of ..."
Ditto last

I don't have a measurement of bean temp.

So if the second pop is 15 seconds after the first you still use the first?

I find this even more difficult when trying to determine the end of first crack, there always seem to be some random pops that go off. So far I have defined the end of first crack as the first period after activity slows down when there is more than 4 seconds between cracks. Not sure if thats a good way of doing it though...

I ask these questions trying to find a better way to go about changing my roast profiles.

15 seconds is an awfully long time between pops. Probably should be discarded as an outlier. If it's discarded, then when the pops occur at short interval should be the start of first.

If you standardize on a 1 second maximum interval between pops, when that interval happens, that's the start and end. You could choose 2 seconds or 0.5 seconds.

If you want to profile accurately, bean mass, and optimally environmental, temperatures are necessary.
Randy G has instructions for adding a bean temperature here

The temperature and bean color at which the cracks start and end changes as roasts get longer.

If you roast fast, the first crack, created by out gassing steam, starts at lower temperatures and lighter colors; while the second crack, caused by the breakdown of the cellulose matrix, starts at higher temperatures and darker colors. So, for fast roasts, there's a distinct gap between the end of the first and the start of the second.

If you roast slow, the out gassing takes longer, the first crack is less vigourous, and takes place at higher temperatures. The cellulose breakdown however, progresses at the regular pace, and the second crack starts at a lower temperature and a lighter roast color. So, on slow roasts, the first cracks runs into the second crack.

If you do not have a thermometer or tryer, it's much easier to get consistent results if you roast fast enough so the cracks are separated by a a couple of minutes quiet time. To do this on a Hottop, roast less weight or drop in the beans at a higher temperature (let the roaster run on empty after it dings for a minute).

Thanks!

Hmm... Perhaps I need to add a temperature probe...

I have a computer a few feet away from the roaster. What do I need to get the roast profile (bean temperature) directly transferred to a file on my computer at regular intervals?

I use an Extech 421509. About USD150.


EDIT : This Excel Workbook takes the work out of plotting and comparing profiles
Roast Data Logger Plotting Excel Workbook
Right click the Link and Save As...

Check out the hottp owners thread on CG, I have pics of my beanmass probe install there.

You could go one further for data logging and real time line graph/roast curve read out by getting a mini thermocouple transmitter system from omega which would send the data to your comptuer wirelessly. The whole system is under $200

That would make for the ultimate probe on your HT and super clean install.

You may also consider changing your P model's board out for the basic which would you give much more control over the roast. I would for sure if it were me.

-j

Dieter01 wrote:I have a computer a few feet away from the roaster. What do I need to get the roast profile (bean temperature) directly transferred to a file on my computer at regular intervals?

As you can see, there are lots of answers to this question.
If you're handy with a soldering iron, and JimG and crew get around to printing another batch of boards, I'd highly recommend using the RoR Meter they designed. My blog entry on it might be the best "short" summary. (I've since installed an ET probe and relocated both probes through the Hottop back wall.)

http://blog.scratchmonkey.com/2010/08/14/coffee-roasting-with-an-arduino-based-rate-of-rise-meter/

The way things are positioned I can do a fairly clean installation even if not going wireless. Extech was expensive when buying from their Norwegian distributor but the HH506RA that Randy G uses (ref link) is available for international shipping.

So basically I if I get http://www.omega.com/ppt/pptsc.asp?ref=HH506A_HH506RA&Nav=teml04 and the KTSS-HH probe supplied by the same company I am all set? Anything else I should consider getting ordering while I am at it???